The State of AI Search 2026: LinkedIn, Zapier & ElevenLabs on How to Get Recommended by AI Search

Hello everyone. Welcome to today's state of AI search. Um, I'm just going to let lots of people pour into the the meeting room. Now, my team have said that I have to keep all the speakers on on track today, right? I've got to make sure that everyone stays within their allotted time slot. And the one that I'm most worried about, honestly, is myself. Um, I'm so excited About AEO and and SEO and marketing right now that it's hard to keep myself to half an hour, but I'm going to try and do my absolute best on

that. Um, so for those of you that don't know me, my name is Chris Donley. I'm uh the CEO and co-founder of Searchable. Searchable was founded on two premises. One was the nature of search and the nature of discovery of products and services has fundamentally changed. we are now having a conversational um interaction with AI Uh to surface and find us what we're looking for. And the second part is that AI has if managed really well, if given a really good harness and really good context, it has allowed us to do things faster and better

than ever before. I don't know about all of you, but um we're doing a a webinar next week, I think, on how searchable uses searchable. And so um maybe I can move myself to spotlight for everyone. There we go. Um sorry about that. Here I am. Um uh while everyone pours in, um I'd love to know where you're signing in from. So um let me know in the chat where you're signing in from. Uh but I don't know about all of you. I now work almost 100% in tandem with AI and next week we'll be

talking about how searchable uses searchable to drive this AI search which is going to be a really cool session because you get to hear from JP who is quite literally a genius in AEO um and SEO. It's been in the game As as long as I have basically um which makes me feel old. Um so today we're going to talk about how you can drive your business and how you can drive um your success from AI search. We have quite literally the most insane speakers today. Okay. So we have um Dan Slagen who's a CMO

and CIT at Zapia where he leads marketing and AI automation. We have um Jonathan uh Craftfort who's the VP of marketing um at One Mind which is incredible. We have Um Luke Harris who's the head of growth at 11 Labs, one of the fastest growing AI companies in the world. And then we have um Andrew Mono who's a VP of marketing at LinkedIn joining us today for the AI uh state of AI search which is incredible. And then we have um a good friend of mine uh Joe Hail who's the founder of Verde which is

an incredible AI search agency. Um and Joe and I used to work together back in the day at my uh SEO agency. And so again The people I want on our talks are people who understand the world of SEO because I feel like unless you understand um unless you understand SEO, the way you approach AEO is going to be so biased. And so that's how we like to come at things. We like to be incredibly analytical. Um, and so let's get into my presentation because I'm already doing it. I'm already talking too much. If you

have questions, put them in the Q&A. If you spam the chat um with uh links or Anything that's um, you know, uh, untoward or annoying, we will just remove you immediately. Um, and let's get into it, ladies and gentlemen. So uh uh AI search is a new discovery system. So I've been working in SEO and digital marketing for 15 years. I've sat on the board of um so many global companies in the CMO position and I'm totally obsessed with it. And I think that if you remember nothing else from today, um I want you to

think that AI search is Not just SEO, right? AI search is a completely different discovery system. It has um very different uh sources as to where answers are created from and the way that we improve our um SEO and AEO is fundamentally different and if you can learn how to do it, you will drive success to your business. And so what are we covering today? So first we're talking about why these engines don't behave like one channel. I see a lot of people talking about like AI Search strategy and they're completely um sort of oblivious

to the fact that optimizing for claude for a B2B SAS company and optimizing for you know chat GBT for a consumer e-commerce company is a very different thing and so AI search is not one channel um Google rankings do not guarantee uh uh performance in AI search they are a component part of it and they matter to an extent but they do not guarantee AI visibility and we're going to look into that together. Um, And uh, we're going to look at what to do differently in AI search with your business to drive success. And then

we're going to look at uh, where to start and what you can actually do next to drive outcomes for your business. So before this call, what we did is we looked at um, 250 brands, 6.1 million answers, five engines tested um, up until August recently. And we watched um how people actually typed questions. We watched how um people interacted with Their different chat agents and we looked at the different questions that the um uh the AI platforms were searching themselves. And so, you know, there's a great part in searchable where you can look at the

query fan out based on certain prompts. It's really interesting to look at that because ultimately I think the way that people think about prompts and AI search is fundamentally wrong and it's very instinctled which is not necessarily how This game works. So um ask the same questions across four or five different models and you will get very different answers. You can run this yourself, you can run questions about your own brand um and you can get a very very different answer. And so if I was to search the question best CRM implementation services for B2B

in these four different models, I'm getting very different answers and only two of the brands appear in all four engines. And On this prompt specifically, we've spoken about this before in our webinars like HubSpot incredibly strong brand and Salesforce incredibly strong brand. And so you could do this for your own sector and see realistically how um you are coming up across the different engines. Now, I think it's really important to to um to focus in on the fact that these AI agent AI platforms, they rarely agree. Um so like 14% of questions returned no overlapping

brand at all across the four Engines. Um so one visibility score looked at on aggregate across multiple models is actually a very misleading score. often um it's indicative, it's useful. But if you want to understand how to perform in Claude, you are going to have to look at Claude individually as a tool. Look at how it's um creating its sources, look at how it's um choosing its sources uh and then understand how to optimize um that uh specific channel. So each engine is Rewarding something quite differently. And I thought I would give a an overview

a topline view of this here which is that you know chat GBT is um is rewarding uh you know community and review content. Complexity uh is looking at the the most authoritative sites um on the internet. Uh Google AI overviews is already is looking at already successful SER features. And for those of you that don't know what SER means, search engine results pages. So the um You know looking at the previous Google index uh Gemini again is using because it's a Google product is again using a huge amount of the Google stack and then Claude

is using its own training memory mirrored against Brave the search engine. So you can see that things um are looking uh very differently. You can see that the the uh you know what they're prioritizing is um is very different. So ranking well um no longer means being recommended. So 67% of pages That rank top three on Google never make it into the answer. So just think about that for a second. You know, for for 12 years of my life, pretty much all we did was help companies get into the top three results. And so 67%

of pages that are ranking in the top three never make it into the the AI answer. Um so in 95% of those situations, AI read the site. they just chose a different page or they chose a different source. Um, and so SEO and AI are related, but they are Certainly not the same thing. If you understand one, you are most likely to be better at the other. And so those who come from the SEO world who are open to change, as in there's a lot of people in the world um, who are very negative about

AEO, which I've never really understood. Um, I think it's like uh uh people don't like change. Um but if you understand SEO and then you accept the fact that in the modern world of search the biggest change um is the fact that The models are using multiple different sources much beyond your own website. And if you can get your head around that all that really means is that brand matters more, social matters more, UGC matters more, editorial matters more. It just means that the websites and the sources that the AI are using to create their

answers aren't necessarily exclusively your website anymore. Um, not I mean not exclusively that not never the case, but >> [snorts] >> um so I want to look at um a um uh an example here. So David and Goliath story. So Nike versus uh the niche. Um, so let's take Nike, one of the largest foot and uh sportsware brands ever, doing about uh you know 45 billion a year and they make uh good trail running shoes for example. And one of the most common questions runners ask is um the best trail shoes for rocky and muddy

terrain. So, for these prompts, as in That question I just referenced there for uh the best trail shoes for rocky and muddy terrain, Nike uh only shows up about 4% of the time for a business that is doing 50 billion a year in sales. And in fact, their website hasn't been cited once all summer. And so the sources that AI actually leans on are run repeat um which is a lab that buys shoes and cuts them in half and it's cited 70% of the time. Then um outdoor gear lab which uh is um Uh another

uh like foot doctor office in Michigan. So Nike wasn't um necessarily outranked here. They just were outsourced. you know, they were outsourced by more relevant source material that answers the question. Um, and it's it's worth saying that this is a category by category situation. And so, you need to do your own research. Like, I strongly advise you if you're not tracking this stuff, sign up for a searchable account, check out how you Perform within AI search. Understand the sources that matter. Understand which sources are being used to answer which questions about your products and services.

um because this is so important and you can see here that you would just expect Nike to have serious dominance within um AI answers because it has dominance within Google but it doesn't for core terms. Um so this is a this is a a wider uh this is a wider problem in general Which is like I said before the idea that it's looking at different sources and putting different weights on different sources um it is um forming a balanced opinion and you can see in in searchable you can see the exact sources that every single

answer is looking at. So, it's it's it's very useful to contextualize that it's it's a it's a a massive area for growth for your business. And just because you're performing well in SEO, um doesn't mean You're going to perform incredibly well for AI. Um so, every industry loses top ranking pages to AI. So, when I say top ranking pages, like top, you know, Google results, uh top top pages that have great Google results. Um, so the share of top three Google pages that never appear in the AI answer. Um, and so you know there's uh

I mean there's literally thousands of people on this call but you can see across these different sectors Um you know health and beauty, art and entertainment etc. um like how much um of uh the sort of core success of Google um is not carrying through to uh the success of being in the AI answer which is really important for us to consider. Um and searchable now works with I'm actually in um Salt Lake City in Utah because I'm opening the American office for searchable right now. Um so we've hired an incredible team out here. Um

and I think you know what I want to um Uh sort of focus on is that like we we work with I don't know one and a half thousand customers now. Um and it's the it's a dedicated focus to understanding that optimizing for AI search is um about onethird SEO, one/ird adapting and improving your pages and about 2/3 um websites and content that you don't control. So websites and content like um you know blog posts or um editorial sites or um user generated content or Reddit or Trustpilot like you don't control those and that's about

twothirds of the battleground for AEO and that's why a lot of the time we're helping customers to um refocus their teams, refocus their priorities and and and and really um take that on. So AI um isn't necessarily like picking a ranking. It reads sources and then it builds an answer. And so for uh a single answer, an engine reads about 10 sources and credits about one of them. And the Other nine um are shaping the the answer, but they're completely invisible. Um and so again, it's like it's very important that you know you as a

brand start to decode what these sources are. um and start to think about creating or influencing content that is being used in the the source material. How am I doing on time? Yesesh, I'm running behind. Um so AI is looking beyond your website. And so you need to look beyond your website because this is Going to be an earned visibility problem. Like you're you're going to need to look at the um uh the the websites that make the biggest difference. the and I'm going to show you in a second how it's so easy to do

this stuff um now that now that we have the tools for it. Um, I think the the other thing I want to say at this stage is like for for those of you that are taking on a EO yourselves or doing it for your Customers when searchable came into the market, a lot of people worked with uh companies like Profound etc. and most of our customers have come from there. Um, and what they were doing was they were tracking 300 to 500 prompts total and it was costing like I don't know like 35 to 50k

a year. And that price was set really arbitrarily high um at the beginning of their journey as a company. But in reality tracking 300 to 500 Prompts for a meaningfully sized business is completely inaccurate um and quite like dangerous way of understanding the success of a business. And so all the customers that we on board from profound, we normally um synthesize for them a much wider prompt universe. And so why do we do that? We do that because we want to look at their Google search console data. We want to look at clickstream data and

we want to then model their entire prompt universe. So the entirety of demand that sits there in the um uh AI engines for their business. And so when customers come to us, they normally see a like three to five times increase in the amount of prompts they're tracking for the same or lower price because their pricing is like crazy. Um, and so once you're tracking three to five times more prompts, what are you getting from that? You're getting much greater sense of query fan out. So you see the machines And what they're searching on the

back end, which tells you what the machines are looking at. it gives you a much richer uh selection of sources. And so why is that important? Well, if I'm tracking 300 prompts for a big e-commerce business or a big B2B business or whatever, 300 prompts can't even cover my products. It can't cover my services. It's sort of ridiculous really to think about it. Um whereas if you're tracking thousands, you can get a Breakdown of top of funnel, middle of funnel, bottom of funnel for all of your products and services. And at the same time, you

get all of the sources for all of those prompts, which means you have so much more context to make good decisions for the business and what content to create. Um, and so just to um uh move on because I'm being told off by my team. Um, so sometimes the short list for uh an answer forms before the search starts. This is really really Interesting. Again, not many people are talking about this at the moment. Um, but take Chat GBT for example. Uh, one in six unbranded buying questions have chat GBT searching for a specific brand

that has never been mentioned. So, think about that. It is injecting a brand that you never mentioned into an unbranded query. So, best running shoes for comfort, it searches um, AS6. Uh, best payroll software for small business, it searches Gusto, QuickBooks, ADP. Most Reliable travel provider, it searches Expedia and Booking. Um and so why is that? Like what is what is driving that? And ultimately the the answer is brand authority um is very significant in the memory of these machines um and very significant in in how they're making their decisions. Um Claude plays by entirely

different rules. Um and so um half the time Claude's not even searching the web. It's just using its own memory. Um And over 60% of what Claude fetches sits in Brave as in Brave the um the search engine. Um and so if you're going to optimize towards Claude, you are playing a completely different game. Now just watch out. A lot of these tools out there are tracking what we would consider like Claude Light as in nonweb search Claude. And so again, really not the way to do that. If you're going to track Claude, do it

properly. track where it searches the web itself so you Can understand the sources. Um so what to do differently in um AI search moving forwards like how how can you improve? Um I think you firstly need to find where you're not showing up. Um, and then you want to find out uh who AI is trusting uh and what sources it's looking at, which competitors are being mentioned um where you're absent and then fix what you can influence. So your own content and your presence that you can influence that's very much within Your um remitt. And

we work with a lot of brands who have seen almost 100% increase in citations over 60 to 90 days just updating their websites. Um and so you know we were talking about earlier that third that you can influence via you know more traditional SEO methods still very meaningful still incredibly meaningful. Um but the twothirds of you know sources and how it creates its answers is much more to do with um uh your um your Offsite your social UGC editorial etc. So, one thing you can change today, one thing you can change today is make your

content easier to site. So, um, structure a page so that it's genuinely easy to quote. So, lead with a question phrase the way that people ask it. Like, if you look at your core prompts, the prompts you really care about, and if they're humanized prompts, um, and they have high volume prompts, you might as well think to structure a lot of your Pages around answering key prompts and key questions that you're already tracking. Um, and we tested this with our own site recently. Obviously, we get most of our customers through um um like a variety

of different means, but we're obviously interested in driving our own performance. Um, and the same uh the same information on these pages just restructured around questions. We saw 2.6 times more citations in four days. And So, if you think you can't do this stuff, you very much can. Um, and the most used part of our tool, which now like loads of people are trying to rip off, is the agent, which has perfect context on your brand, how you write, how you think. Um, and it will help you recreate and amend this content. So, I want

you to start by asking four questions. So, the questions are, do you know where your brand appears in AI? Do you know which engines you appear in and Which you don't? Do you know which sources AI sites instead of you? And do you know where your visibility is improving or decreasing? Um, one of the questions, one of the things I want to put out to people is, um, a lot of the customers we take over from, uh, profound, they come and they track five models. They track like AI overviews, claude, co-pilot, chat, GBT, Gemini, and

they only track like, you know, a limited amount of prompts. I would say That honestly, you should think long and hard about which models you care about most, which are most relevant to you. If you're a B2B SAS company that is looking for customers to find you through um Claude, then you you have to um work on Claude. That is going to be your absolute focus because 78% of B2B buying vendor decisions are happening there. However, if you're like a BTOC consumer business doing ecom or whatever, it's going to be like trackbt Gemini AI Overviews.

So don't just track all these models for the sake of um you know these tracking platforms like you want to pick the ones that are most relevant to your business and then go deep on answering those. Um so most brands truly have the same fixable problems. So um when we audit sites we see that um nearly all sites that we audit have like fail to give a clear reason why a customer should choose them. So most are missing like basic FAQs, you know, basic Objection handling content. Um, and these days, um, a lot of these,

uh, fixes could be yourself working with the searchable agent for a week. You know, it's going to create schema for you. It's going to create meta descriptions for you. It's going to create FAQs for you. And I just want to give you a frame to think about using AI in general, which is 108010% incredibly hard work, focus, dedicated work up front yourself. It's where People go wrong. They trust AI too much. 10% upfront yourself doing the work, giving the context, thinking about the strategy. 80% you can give to AI. So whether that's like claude co-work

or whether that is searchable, you can give that to searchable to create. And the last 10% is you owning the outcome. So reading it line by line and ensuring the quality and that's how you should use um AI not just um not just leave it to AI specifically. Um so what can you do um Uh in the next period of time? So like I say find where you're missing find who AI sites instead of you and try and fix one thing and then track if it helps to drive performance. Um so the other half and

this is a lot of what most people are not thinking about these days is um the other half of optimization outside of your own website is optimizing these different third party websites that you don't control that you don't control. So obviously this isn't just you know your YouTube account or um you know your Instagram account or or whatever. this is a a multitude of um surfaces across the internet that you don't control. And so it's really important to understand and you can do this so easily um within uh within searchable. You can like look at

the the breakdown of sources and how to use them. So on the left there, this is a customer. you've got um sources broken down by individual website and you can see um you know uh Tom's guide Or cycling news or whatever they are being used across three and 4% of the entirety of the prompts that that brand specializes tracking. On the right hand side I've grouped it together by domains and so you can see that you know Reddit is answering 15.7% of their questions. And so this is how you quickly work out your sources um

and how to use them because this is going to this is going to direct um your your focus. Now before I introduce the speakers, so like I say, We have like the most insane speakers today, you know, we have like a VP of marketing from from LinkedIn as well. It's pretty sick. And then we have like head of GTM from 11 Labs, one of the fastest growing companies in the world right now. Um so it's going to be an amazing talk and I'm going to do the the handovers myself. But before we do, I just

want to mention um the uh AI search accelerator is rekicking off um uh in uh the next period. And I want to just give You an update as to um uh what what it is. So we have two tracks. One is for brands, one is for agencies. We have quite literally had the biggest brands in the world using um uh searchable and using our accelerator to train their teams from like big agencies like Merkel through to big brands like Dagio or H&M. And so we will train you um over uh the um the fiveweek period

on live in per live online training um where we'll give assignments. We'll give like expert-led Classes. We'll go from foundation through to um uh outcome. We will provide on demand modules that you can watch that we've created for you. Uh and so you can go through how to implement and execute on AI search. Um the next cohort kicks off on the 22nd of September. And so if you're interested in joining, uh make sure that you sign up. Um obviously that's only for searchable customers. So you have to be a searchable customer to join the Accelerators.

Um and um we've already had 700 brands go through and these are just some of the standout quotes I just pulled for you um for this talk. So if you are interested in joining, make sure that you apply. Um uh the guys on this slide effectively are talking about how the accelerator help them improve either their AEO service to their customers um or um uh improve the actual AEO performance of their um business. And every single day people are posting Their success within Circle um outcomes. They're seeing increases in mentions, increases in citations. Um, and

so without further ado, have I kept my time? I have kept my time. That is amazing. Um, so these are our upcoming speakers today. Like I say, if you have questions, put them in the Q&A. If you, um, uh, if anyone spams the chat with anything at all and it's anything untoward, we will remove you. Um, so uh, without further ado, let me just stop Sharing here. Without further ado, can I please hand over to Dan um who uh is going to give us um an incredible talk. Uh and I'll let Dan formally introduce himself

uh so he can uh talk through what he does because your your role has changed um a little bit which is cool. Talk you through what he does and then go into his presentation. So um welcome to Dan and thank you for joining us. >> Yeah, thanks for being here. at least Get my uh sharing all set and uh thank you for having me. I'm uh normally based in Boston, but coming to you from the the woods of New Hampshire. Um we're big big searchable fans, so I wanted to make sure that we could we

could be here. I just want to make sure everyone can see my screen. All right. >> Uh sounds good. >> Yeah, that's cool. >> Sounds good. Great. Um okay, great. So I want to talk today really about how AI Is changing uh modern marketing teams and the role of AI transformation at Zapier. Uh I've been with Zapier just under a year. Uh the chief marketing officer and chief AI transformation officer. So they're keeping me busy over there. Um but you know just to start out you know I always try try and step back and just

think what what what are we doing? There is so much going on right now. Uh if you go back to a couple years ago when this AI boom really hit you Know the marketing world. um where are we ultimately trying to get to from AI chat to what the destination might be? And so for us, the way we think about it is, you know, a full-blown AI operating system ultimately is the destination. Um for for us, it's appear and for a lot of the people that and companies that that we work with and sort of

wherever you think you are along this journey, it it's okay. Um, it's just good to have a map and to understand sort of what the Key things to focus on at each stage are as you move from, you know, where we were a couple years ago to a full-blown AI operating system across your organization. And so today, I'll briefly touch on sort of how we see AI changing day-to-day work. uh what AI search means for content and how we think about that, which of our workflows are changing particularly fast. Um how leaders should prepare for

the next wave and then just some practical lessons from uh the AI Transformation work going on at at Zapier. So starting around this notion of of search and content, you know, AI search is essentially assembling your brand from pieces uh across the web. And as you just heard, you know, different different LLMs, different models, they all specialize in different things. is they they pull from different sources. And so what's interesting is that a customer is likely to land on your homepage after already seeing some form Of the assembled answers that an LLM chooses to put

out about your brand. And you don't necessarily have control over the specific prompts. The level of detail people can prompt now, as we all know, is through the roof. And so the the outputs that people see are nothing like we're necessarily used to in the traditional SEO sense. And so that kind of changes the content that you choose to make. You might do more comparative pieces. You more focus you might focus More on proprietary data or things like that. But since each model is different depending on where you're trying to rank for and what given

your target audience, you're going to need to tailor the content around those types of things. And so we see it at Zapier. we have our automation bench where we're able to go in and look at all the different LLMs and we can see you know everything from on the left which specific model does the best um whether you're working on Something related to sales or marketing or ops we also can see on a cost basis um which one is costing the most or the least. So, whatever you're trying to optimize for, wherever you're trying to

show up, there's a specific path that you can you can build towards. And that's what's really critical about thinking about your content plan, your search discovery plan in a very custom way and not like you've done in the past. You're going to hear me talk a lot Throughout this this uh this presentation around changing the way you work. If I can drill home one takeaway in this session, it would just be now is the time to change the way that you work more than anything else right now. And so as we really have become an

LLM first content organization, you know, we think about it across our our team and ops and first it's it's a behavioral change saying that we're committing to geo first content. You know, this sounds Small and it sounds obvious, but needing to see certain types of reports and needing to have different types of conversations than we've had in the past and really building that muscle up and getting it to a point where now this feels feels normal to us was a really important transformation that that we went to. And we still do a lot of writing.

We love writing. Um, but we've also thought about this this shift from just writing to more creator-based Because if you have LLM pulling from so many different sources across the web, um, you want to make sure that you're being found in all those different areas. And so we have more efforts going going on outside of, you know, just our just our website. Um, and that really speaks about, you know, where to invest in this shift from on-site to off-site. This is something that every brand out there is going to have to make their own decision

on. And it's a really hard one Because if you spend too much of your time building offsite and then an algorithm change happens, you lose, right? It's like you've rented space and all of a sudden they jacked up the rent and now now you're in trouble. And so we've had to really think about what sources do we trust? Where do we want to spend our time outside of, you know, our our own our own um assets like our website and such. Um, but that's become a really critical conversation and for Us ultimately there is so

much noise out there now that cutting volume I think for us has been a good thing to really understand where we have the right to win and where we want to win and really tightening our our focus both off-site but also on-site quite a bit. You know, we always look at our content and we say, you know, where where could our competitors copy us? And you know, it's it's pretty interesting that automation benchmarks report I showed you is just One good example where, you know, we see competition able to copy us in the headline of

anything or in the sub headline of anything. But where Zapier wins is when we can really really go deep and we have the audience to go deep, we're going to win almost every single time. And so we know that and we we feel pretty good about that. And that informs how we think about the content that we actually go create and and wear. And you know, we think about driving Change at the organization as well. again, different way of work, but as we're building out these new content engines, you have to redefine success. And so,

we had gotten really, really good, just like a lot of people on this call, at optimizing our content, probably down to revenue at this point because we spent so so much time doing it. And now, we've sort of had to re hit the reset button a little bit and say, "All right, let's first make sure that We're showing up where we need to show up, and let's first make sure that we're getting traffic, and let's first make sure we're getting engagement." And yes, we're going to get down to the point of being able to know,

okay, if we do this, we expect to see this many leads, we'll create this much pipeline, this much revenue, but we've also had to admit to ourselves that going back to some of the higher level metrics for now um is is is actually a good thing because we need to Be able to start seeing the funnel um from the top down and it's something that's been kind of re recreated. How that's changing our day-to-day marketing, uh it's it's quite a bit. You know, one thing I'll say is that we have um about an 80 person

marketing team at Zapier. You know, AI changes each person's role differently and it changes each team within the marketing organization differently. And so the way that I'm asking the demand genen team to Adopt is different than the way I'm expecting or thinking about the brand team or the product marketing team. And so it can be very easy to do this top down. We need to make a change. You all need to use AI more. we need to get more agentic. But when you really dig in with each individual team and each ind individual person, that's

where it gets a little bit trickier. And so I'll talk later in the presentation about how how we've done that. Um, but really just Admitting that this is different to each person in each team. Um, and in some teams, you're going to have to push more or less or focus on this or that. That has actually been been really really good. I'm also seeing it's people spending less time around manual translation. Um, just meaning we've had to just nail getting our source of truth, building what I'll talk about later, but it's this AI brain where

we just feel good about the truth and the Sources that's influencing and informing everyone on the marketing team about making decisions. And now it becomes much more about your judgment and your taste and what is actually being shipped out and making sure that we have a unified uh truth and point of view for what our customers and prospects are actually seeing out in the market. And so that's been one big shift across the team. There's also this concept that someone on on my team uh name is Joe. He Runs our identic AI marketing came up

with this concept of lossless marketing which I love. Um for anyone out there that remembers VHS tapes, if you made a copy of it and then another copy and another copy, eventually it would just go to fuzz. And we see that a lot in marketing. uh we we see like a generational loss uh as the product truth gets translated across your teams and your docs and your decks, campaigns, channels, all that type of stuff. And so We've really had to rethink, okay, of all the stuff going on kind of behind the curtain at Zapier. How

can we make sure that what actually goes out is really the the single source of truth and the best thing that our prospects and customers need to see and that the marketing team is aligned, the product team is aligned, the sales team is is aligned, our entire go to market organization is aligned. And so it's this new way of operating with AI kind Of raising the stakes. One is everything is shared. Uh I'll get again more into it around when I get into the brain concept, but it's our context, it's our content, everything is shared

across the team and everyone has access to everything and can see everything. That role of AI versus human is critical. I know everyone's seeing now, you know, probably some type of a pullback on like AI writing because there's just been so much slop. I've seen, you know, a bunch Of companies essentially outlaw uh, you know, AI writing for their internal communications. Zap year, we have our own policy and point of view. Uh, you know, we we don't expect anyone to just create something with AI and throw it over the fence and tell someone to read

it. Um, we you wouldn't do that in other areas of your life. And so, you know, that's becoming more more of an important focus because people can identify slop and low quality And things that don't make sense a lot easier. And so, that operating shift has been really really important for us so that we can get an endstate clarity. And again, that's a single point of truth that we feel really good about. One thing that's been really really helpful that we've implemented over the last couple months, I would call this an absolutely critical meeting for

every single marketing team, and I'm sure people did some form of this in the Past, but for us, we've kind of tailored it now to to the AI world. Once a month now, our product marketing team essentially leads a look back meeting for the last 30 days and sets the narrative for the next 30 days. And in that meeting, we try and have as few people as possible, but the most important stakeholders, our CEOs in that meeting, folks from the executive team and some select product leaders from from around the organization. And again, We go

back and look at look at what happened in the last 30 days, what happened in the market, what happened with competitors, what happened with customers, prospects, the good, the bad, what happened outside Zapier, what happened inside Zapier. And we've set an agent that we've honed an agent to basically pull all the things that we care about. And then we start to say, okay, for the next 30 days, is anything going to change? Our positioning, the Way we go to market for sales, the marketing activity, um how we're going to go to market, what are the

campaigns we're going to go out out out to? And for some things, it's no, right? Your brand, your brand promise, your mission statement is not going to change every month, I hope. Uh but other things might, right? You might have ad campaigns that were quarterly that now need to be monthly or even faster than that. Uh it also has really helped us Identify when something happened in the market, were we were we able to respond as quickly as we wanted to. And so this is one meeting I absolutely recommend everyone do. Um it's been really

really critical for us. Um on the workflows, you know, one thing that's been interesting to see the changes is that AI has just flattened execution really across the team. Um you know, PMMs essentially now, right? We can create one single source of truth that everyone Can see. um you know our go to market team and our demand genen team, right? They're able to pull what the most relevant uh campaigns or talking points or insights might be and be able to take those out to market. And so that's been something that's been really really great to

see. Um again for driving product truth content campaigns, demand genen, go to market. Um we're going to see a lot more of this type of stuff. My recommendation for everyone is that as You start to feel and you're going to feel it more than see it first. when you feel that something just isn't quite right or something just feels like it's being done in in an outdated way. One of the hardest things I've seen is for people to just raise their hand and say, "Hey, something doesn't feel right." Um, it feels like we should be

working differently. You know, when my product marketing team, they'd say like, it seems like we just have a lot of Confusion right now around what the core narrative is for the next generation of Zapier's products. I'm seeing confusion over here. It starts in product, but then it goes over to sales, then it comes to marketing. like we need to rethink this whole workflow. And the reason I think people at Zapier have been willing to do that is because they feel safe and because we've created the environment for them where we sort of said, "Look, there

are no sacred cows Right now. Everything is changing. We need to make sure that we're we're adapting in real time as well." And so just think about that. It's it's as much of of an organizational change as is as as it is a technology change. Uh and that's something that that we've noticed quite a bit. Um and we've also seen this externally too. It's not just within the walls of Zapier. It's also what we're seeing our customers and everything do. We have an AI workflow index which our Head of AI orchestration strategy Andre recently launched.

And it's been really interesting to see sort of what workflows AI is starting to take over. Um where we still see a lot more human involved and and it'd be interesting to see where some of these numbers start to to balance out. You know, on the left, no no surprise that we see AI writing for people, extracting information. Uh we call those roles a communicator and a clerk. on the right side when it comes To analyst and correlation uh coordination. We're not seeing as as much human involvement there right now. We're of course in the

earlier stages of that, but it'll be really interesting to see sort of where those numbers start to go, but this is sort of at a high level how we think about where the workflows are are are going. Um and it's really interesting too when you actually get in by the department you sort of see relative consistent consistency and you Know within marketing what's so interesting is on the right side we still see the majority of you know analytical decisions coordination decisions being made primarily by human and so you know whatever you're seeing being talked about

in the media it might be a bit of a hyperbole out there and it's very very much early early stages of humans making sure that they have their hand on the wheel because we want to make sure that what actually goes out To a customer, to a prospect in the market, to a campaign um is at the level that we would expect and we would want to be great just like we would have wanted in in the past as well. But I do we're going to be watching these numbers. We'll keep releasing them on a

quarterly basis, but they're really really interesting to to watch. Um and so how leaders should prepare for kind of the next wave of of AI. Um this is one I spend a lot of time thinking About. You know, if I just go back to the phase one of AI, uh, for us anyway and a couple years ago when it when it really got big, you think about like midjourney hitting and the first iteration of chat GPT and we were essentially doing content creation and the level of entry to be an expert was pretty low. You

know, by the end of the day, you could be a pretty good user of of chatbt. Then we sort of went into this phase two and you had a couple Companies popping up. You had Claude coming around, you had Deep Research. Um, we started thinking about how can we influence roles and functions. Entry level there, medium I would say. You know, it might have taken you a week or two to to get the get get the handle of things, but not not impossible for someone that was really trying to push the limits. Phase three, which

we're in now, is this notion of like agentic systems, automating everything. And to Me, this is a lot harder for people. I'm seeing the the level of entry being much higher. You know, I look at some of these these these coding tools, cursor, cloud code, codecs, and as we started to see, I'd say in Q4 of last year, non-devs starting to use these tools, we called the code red at Zapier. And I would, if you haven't done this, I would I would strongly suggest that you call it code red as well, where we sort of

said something just doesn't feel right. It feels like the market could pass us by if we don't change the way that we think about those tools. And so what we did at Zapier was we did a buildathon on the marketing team. And what I did was I told everyone on the team, look, you have three weeks to go to go learn any of these new coding tools out there. I don't care if it's codeex, I don't care if it's cloud code, whatever. Cursor, doesn't matter. But you need to go out and learn it. I gave

the team access to Five coaches that were already experts, and I said, "You guys can meet with these people as much as you want. I will never know who met with who, for how long, what was talked about. I don't care. The only thing I ask is that in three weeks, everyone shows up to this buildathon ready to go. And so, for the most part, everyone did. And they showed up to the buildathon. And the ask was to create something, something related to your role, something related to your Department that you think would have impact.

And the first few days were a little bit worrying because I gave people about 2 days to or sorry, two weeks to actually build. And you know, we're tracking our token cost and not seeing the spike in in usage that you would have hoped. Uh but like any great marketing team, of course, all the work happens the the night before, which is fine. Uh and so what we ended up doing is at the end of the day, we had about 50 plus builds that came in from nine different themes. It was really really great to

see. People built reporting dashboards all the way to what I would call a full-blown software pro uh product that if they wanted to quit Zapier and go start a company around, they they certainly could have. Um, but it was really really great to actually see the amount of creativity, the amount of depth. And the biggest thing was just this mental shift from I'm really scared To now, hey, I really understand the start of this and where it's going and I'm on a journey that I'm actually excited about. And so [snorts] people had fun. We went

deep. Things got a little bit weird. You know, on on the left, someone's presentation of what they built. They showcased it in a video game that they built. on the right. It's like a green room with a gray hallway. Someone actually built a digital twin for themselves and they were trying to Think about what the twin actually sees when it's trying to think. Um, so as I said, some things got weird, but it was great because we gave people the comfort and the freedom to actually build what they were excited about. And so, as I

mentioned, you know, some people built just dashboards, but again, we had some people build complete software products. For instance, we host an annual conference uh called Zap Connect, which is happening later this year, but 70 Plus speakers. The system that we're going to be using to manage the entire event for our our speakers is now going to be something that we built, one person on the team built over the course of about two weeks and then probably honed it and perfected it after another two weeks. Um but that's that's the extent of the things that

we're building. So that was really really cool to see. Um I'll say, you know, celebrating all achievements was really Important, right? You got to remind people why we're marketers, why we're here. Have fun, be creative. AI is such a constant source of fear and anxiety for everyone. You know, we even had as much fun. We try to give superlatives to every single presentation that came in. Um, we made a joke that one person presentation actually was better at 2.5 than than normal. Um, I will say as far as, you know, leaders digging in and showing

by example, absolutely critical. I did my own build. I also watched every single presentation and gave handwritten feedback to every single build. Um, and ultimately what came out of this was three different tiers of people. Tier one, tier two, tier three. Tier one kind kind of the earliest builders probably building some type of interactive, you know, software. And the tier three is sorry, interactive dashboard. And then, uh, tier three is building like full-blown, you know, products. Now, I Actually have though three groups that I can really focus on. And that's critical because as we think

about what's next within these tiered groups, I can now go to each group and say, "Okay, I have a couple business challenges that I really need you to focus on." And the focus is on impact, not just effort, because you need to actually be solving something that's ROI critical for the company. And so then within each of those tiers, I can have a plan for each team manager. And that allows them to really tailor what an AI journey looks like for everyone. Um, out outside of that, it's just more sharing. We're going to we've continued

to do the buildathons. they've been really really cool for for people to see. Um, as I said, I'll keep saying it, the next wave is all about changing the way you work. You know, when I started that red alert, what I actually did was I brought an external speaker in to talk to the Zapier marketing team and It was someone that was really, really ahead of the curve on Agentic and they showed the team how they thought about marketing, where it was going, how they were building. It was extremely technical. It was extremely overwhelming. You

know, at the end of that session, I sort of asked the marketing team, how many of you guys are feeling overwhelmed? And every single person for the most part raised their hand. And my comment to them was, you Know, this morning you all woke up and you didn't even know that this person existed. Our odds are you're going to be interviewing against this person in the future for a job at Zapier or a job that you want outside of Zapier. You need to know that this is where things could be going and you need to

make sure that you at least have the skills to do this. And each person is going to decide where AI goes too far and where it doesn't go. And it's going to be this constant balance of of technology and taste, which I think is great for people. And again, kind of leveling that playing field of opportunity and success. Um, but after we kind of did that shock and awe moment for the team, we immediately shifted to comfort and support and hey, I have the coaches here. You guys are going to learn these tools. We're going

to do it together. We're going to share. And now the team is in such an exciting Place. Um so just again remember that these are real people. Everyone is has has anxiety around building this stuff and you know we're all learning together um in in in this journey. The other thing that we've done that's been really really important for us and and the way we've worked is to build this infinity brain. Um you know AI is obviously table stakes right people kind of they learn it on their own but then you need to get it

working across a team And across a department and across an organization. And the failure mode is that AI is giving people these positive feedback uh remarks essentially like saying great job this is right what else can I do for you and so when everyone's doing that individually they're essentially getting coached by like a genie type AI that's just you know pointing them in thousands of different direction to scale. Yep, that sounds great. Go do more of that. Yep, that's a Good idea. Go do more of that. But if it's not connected to your organization and

you don't have sort of a shared knowledge and shared understanding around how you want people to be coached, then you're going to end up with a complete mess. And so that brain is really layer one of of what we call AI transformation and how we think about AI transformation at Zapier. And it starts with the brain. Then it gets into things like the tools and the Automation, the governance and and risk control that you do and the harness that you actually work in or the harnesses that you actually work in. Um, but what actually lives

inside of that brain, that's where it gets really really important. And so for our team, right, we have the company layer. Uh we also then have it by function and by team. So it's at the most basic level. You know, it's what your brand guidelines are, your mission, all those types of things. But you also really want it to be uh sucking up all of the things happening on a day-to-day and and and week-toeek basis. And then as I mentioned, we have this notion of coaching with closed loops for people. So, you know, every week

we almost have like a weekly staff meeting that an agent's going to come in and start to inform people on what's working, what's not, uh, what follow-ups we need to do. Um, [snorts] and we do that both the company level and and the Department level. So, getting really critical what's in the brain and then how that's being used to actually coach uh is something that I I recommend because it's been really really um helpful. I will say building the brain is not an easy task, but it's absolutely critical right now and it's only as useful

as what it can see. And that's been something that's really interesting. If you have a company that's working in private, the brain Doesn't actually have access to those things. And so we've had to really transform the way that we work. Uh we're primarily working in Slack. Um you know, we're uh just under a thousand people fully remote. And so Slack is our primary form of communication. And so if all of our conversations are happening in in private or or direct messages, the brain necessarily can't necessarily see that. And so you know at the exact level

we we start by trying to work as Transparently imposs as as possible. And you can see each month uh our CEO posts the percentage of Slacks that were done privately versus channels. And you know when I joined the company I was about 50% in public versus private. Uh and now over on the right you can see last week I went from or last month I went from 51% to 98% public. That's been such a shift in the way that I've had to think about working. Um, I told the team, "Hey, I'm changing the way I

work. We Need to get deliberate about how we how we communicate." I put in a daily tracker to make sure I know what percentage of my Slacks are public versus private. I would push people to public set uh public groups as much as possible, have more uncomfortable conversations where I might have had that in a DM, but now I'm doing in groups and ultimately drive to something where I felt like it was sustainable. I I try and aim for around 90% uh public Communications. Last thing, then I'll talk. I was just going to say uh

we've got a couple minutes left before the handover just as a a warning. >> Yeah, no worries. I'll be on two minutes. Uh so last thing, practical lessons for AI transformation Zappier, right? We're all trying to get to this transformation transformational state. Um really important to move from just engagement and effort driven over to impact driven. Um everything that we Build, everything we think about having the team focus on we want to make sure is focused around ROI. Um we've gone from sort of V1 to to V2 on AI fluency. So for us, we have

an updated uh minimum bar for new hires, right? We always like to think people that, you know, got hired six months ago might not get hired now because of how the bar gets raised. Um we've added accountability in AI fluency. We make sure that our managers when they're interviewing have can show Us how they've actually led AI adoption or how they think about leading AI adoption. Uh and again kind of redesil designing some of our skills tests really to observe how people are using AI because we know they can do it um in some way

shape or form but observing the how is where we really start to see either the the red flags and candidates are actually things that that get us really excited. Um we think about it across you know what what are people's Mindset how they think strategically how they think about building and how they're accountable to it. So again all this goes into the interviewing process at Zapier. This is how we're constantly evaluating whether or not we think someone might be a a a good fit. And we've actually have a full-blown um AI rubric that we have

at the company level and then we also have the department level. So this is on the marketing side. You know, on on the left we of course Have unacceptable, on the right we have transformative. Uh but we want to make sure that people fall into the bucket and we're actually finding people and and helping training people uh based on where where we need to see them. So thank you all. Um really appreciate it. If you have any questions for me, you can find me. It's zappier in LinkedIn. Uh, and again, if you take one thing

away from this presentation, the time to change the way you work is uh is Absolutely now. So, Chris, appreciate youall having me. >> Thank you so much, man. That was epic. Um, and I feel like I saw um I saw a note in the in the chat where someone said uh like how relevant is this to, you know, marketing and and AEO? And I think what's so interesting is that in traditional uh SEO, we were all marketing towards Google. Like we spent all of our time and our energy and our effort going towards Google. Whereas

now We have so many more surfaces that we have to contend with and and so many of those new surfaces i.e. chat GBT, Gemini, Claude, Copilot, etc. all have different sources that they rely on. And so the work of marketeers has got so much higher. It's got so much harder. It's got so much broader. And so if you're not able to use AI to execute at a greater scale than you were before, you are going to get left behind. And so a lot of what Dan spoke about today for Me is at the the the

very heart of how marketing and how um marketing professionals have to think moving forwards, which is you have to train yourself to be able to work at a greater scale than before because the problem is greater. you know, if you want your business, which is a um, you know, e-commerce consumer brand to perform, you need to be able to work at scale across ChatBT, Gemini, AI overviews, AI mode. If you're a B2B business, it's Going to be like Perplexity, Co-Pilot, Claude, etc. So, you're going to need to create more content. You're going to need more

analytics, more automation. And so, I thought that was absolutely epic. And as with many other people in the in the chat, I love the fact that you talked about people's people. and great leadership within AI. So, thank you very much for joining. Um, and if everyone wants to um give Dan a follow on LinkedIn, he creates incredible Content and he's a great leader. So, thank you for coming, dude. >> Thanks, Gracie. Appreciate it. >> Um, amazing. So, um, I hate interrupting people because, um, uh, you know, I'm so in rapture as to what they're saying.

However, I have to keep things on schedule because my team told me to. So without further ado, I want to hand over to Jonathan who can introduce himself. Um has had the most fascinating career. Absolute you know subject matter expert On go to market um a friend of mine and so I'll let Jonathan introduce himself introduce his talk and I'll let you get into it and I will as with Dan interrupt you if you're close to going over. >> Okay. I'll keep my time on Chris. Thank you so much for having me. I'm a big

fan of the searchable team. This is my second time using Searchable. really excited to be here and just want to say thank you for everyone for hanging out. Hopefully I can see you. So, okay. Um I'm going to put up my slides. >> That's the second brand you've used it for. The second time in total. >> Second time. Second brand, two companies. Yeah. Yeah. Sales. Yeah. Um long story short, I was at which I'll talk about a second. I was at a company called Momentum. We were acquired by Salesforce and so I was using Searchable

a ton there and then obviously the acquisition happened and then um went on To one bind which I was like I got to use Searchable again. So, one of the first tools I bought was searchable when I got to one mind so we can start tracking stuff. So, it was really cool. All right. So, you see my screen? Okay. Everyone's wonderful. Hopefully happy. There we go. All right. So, this is going to be kind of a I guess you could say a case study. Um, and just kind of walk you through what I've seen in

both momentum and in um one mind to kind of Give you a couple things. And thank you. Thank you, Gary. It's actually I do it purely for style. I might do a little bit because of my eyes and I have screens everywhere. So, it's helpful for my eyes not to go crazy during the day. Um, okay, let's get into it. So, I'm going to go over what we're going to go through in the next uh let's see here 20 25 minutes. Um, you can see here the short list forms without you, the first six weeks

for me and it's kind of what We went through, how the model site structure, the two views of the same buyer and what's next. So this is kind of sort of a case study and hopefully you can walk away with some some nuggets you that I could give from my own learnings of using obviously searchable and my own AEO journey and trying to figure out what what's next. So, real quick before I go into some things to give you some context and and my relationship with with Searchable. Um, there was gosh, last year, so AEO

was an erupting thing back in 25 and there's a lot of different companies out there, which I won't name any of them, but I used literally everything you can think of. All the big names out there, all the hottest brands. I was testing and playing and I was able to do a bunch of different things with with AEO technology to kind of figure out what was good, what was not good, and helpful and not helpful. So I found a lot of the Technologies did a lot of analysis or kind of showed me where we were

showing up but it was missing key context around like for example Google Google Analytics which is a very simple thing now to connect but back then it wasn't it wasn't as easy to do that and um I was trying to find some way that I could like connect number one connect everything together and then not only analyze but then to create content that would be further create this loop of of Uh progression and making sure things were working right. I didn't have that. So I used like five systems to make sure everything was working right.

It was a pain in the puot. And when I finally found search ball, I was like, "This is awesome." And it helped me dramatically in my team going from using five different tools to to just one, which I'll give you an example here in a second. But that's kind of my context is I started using Searchable back in early Middle 2025. And so I've been customer now for a year and a few months and it's been awesome. So a couple things about me just so you know, my name is Jonathan Carford. Um, uh, oh, look

at these. I'm gonna see all these things. This is really cool. John Wood's cool. Okay, make sure I don't miss your comments. Um, my name is Jonathan Cartford. Some people call me coach. Long story short, I was in a startup once upon a time several years Ago. I was in a room of like 200 square feet or I maybe it's larger. I can't remember. Um, and there's 40 people. 20 of us were named John. We all got nicknames. Mine was coach because I was the sales coach of the team and then it just stuck. Currently

the VP of marketing at One Mind. Um I was again at at uh go to market strategy and marketing at momentum which was acquired by Salesforce back in February. And then I'm owned the GTMai academy. So for the Last five six years I have taught go to market professionals in marketing sales customer success revops enablement of how to use AI both individually and organizationally of AI strategy. And I host the GTMAI podcast which speaking of which Chris I got to have you on and have your team come on and hang out which would be great.

Um, and I've I've advised a ton of companies both large and small on go to market AI strategy. I teach at Bryant University AI. I'm the Co-author of the Ignite Your GTM with AI. And so I live, breathe, eat, teach, sleep, talk, sell, market AI all day. I am not an expert. I am a nerd as I like to tell people. Like I know experts in the field. I just love this stuff and geek out. So a lot of fun. I want to talk about the story around Momentum because I think it gives you an

understanding of why I'm so passionate about Searchable and what what they're doing there. Um, hey, what's up? What's Up, Coach Matt? Love your name. Love it. Um, okay. So, a time I did a split test back before I had the technology where I could do everything in one place. I used several different places or several different technologies in order to see how I could optimize content. And again, it was kind of an erupting field then where we're figuring out like what content, what format, um how to position things like it was it was a lot

of it was a lot of different content. And Actually Amelia, who was who is one of the searchable team before she joined it, I used a lot of her inspirational content to kind of help me with AEO, which was a lot of fun. Um so here's my process. Before I had to I I wrote this advanced mega prompt. So, if you know anything about me, I'm I'm a prompting nerd and now it's now skills, but back then it was prompting before skills came out. And I had this massive prompt that I gave to my my

human writer. Gave that Prompt. We put it into chatblad and they did a deep research project on that subject based on um key terms. So, we used Google search terms. I put those into a prompt. It did a deep research project on that thing. And then we took that, we wrote an article. So the human uh my human writer would take that deep research, edit it so it becomes more formatted to like a blog thing. Then we put it back into clear scope which optimized it for SEO keywords and then My human went back through

it again and then we designed a blog so was nice and pretty and that was it. So a six-step process probably took the writing for the human side. This probably took a half hour at least if not an hour for each article and um the design part obviously took a little bit more time with my designers. So that was the first step. So that's what I did when I find searchable. This is back in last gosh like a year ago back in August. Um I was Curious to see how the agent inside of searchable would

write and how it optimized. So what I did was I was like okay I'm going to take the same thing and say hey searchable I have this key term. Number one, can you confirm it's the right one? And number two can you write me an article about this key term? And that was it. I didn't have on purpose. I didn't have my human edit it. So I was like, I just want to see what can be done by just using searchable Just to see just to test it. And then we designed the blog. It was

200% I'm not kidding with this 200% more traffic to the pages and the blogs that were written by searchable than the other side of the process. And my human's not u not a bad writer. She's very very talented. Um so the difference between the two is like drastic. And I was like holy Hannah. When I saw that, like we did it for a month test and I saw the traffic between the two. It was kind of Insane. So that's when I was like, "Okay, we're doubling down on on searchable so we can get some more

more things here." And we've done the same thing at at at one mine, which has been fun. Okay, so I like to um again, you all know this probably already just make sure we're on the same page for this the next 20 25 minutes. Um buyers build a short list inside of a model and we we at one mind we have instrumented both sides. So, I Know a lot of people have like use tools like searchable obviously to see what's going on inside of AI. The other side of it is um sorry the other side

of it is like what happens when they when they um actually attend. Oh, sorry Keith. Um sorry I with with one mind we talk in my human. So my apologies for not being trying to be offensive. Just want to make sure that when I say writer people know it's it's it's a human being. All right. So Anyways the um uh with one mind we have a technology where when someone comes to the website that um uh we have an advanced technology that can go deep in conversation. So um we're able to see like how people

are interacting and talking to to um uh the technology and and we can see what's going on inside of searchable with the AI search. So um what I'm just saying Julia is that we use both sides. So in other words, we used searchable to see What was going on inside of the model and then we used our own tech with one mind to see what was going on on the website to see like how are the conversations mapping together like what's the similarities between those two things because for example when people come to your website

unless you have an AI that can have advanced conversations you have kind of shallow results with like a regular chatbot. So our technology can go deep in Conversation and answer a lot of different questions. And so we did this research study where we found from a half a million conversations like what's the difference between just a regular visitor and those who engage with our AI tech and we found this this pattern. So which I'll get into in a second. So with with searchable when we started again you can see the dates here from May 11th

to I pulled the data 2 days ago I think um was 18 we went from 11.5% to 18.4%. So, this right here, by the way, if you're inside of Searchable, you have the ability to break this down by branded or unbranded search terms. This uh metric is for both unbranded and branded. So, our branded is obviously way higher. It's like in the 85%. And unbranded is lower, but between the two, you can see like this this scaling up, which is interesting. And we just so you know to give some context, we did a total redesign

and re-indexing of the Website in the middle of June. So, in the middle of this, we had a re-indexing of the website, which you can see like the the shift as we right about here. It dipped down a little bit and it's come back up as a result, which was a lot of fun. So, you can see like how we're performing and I can um I may show the back end of searchable. You can see a couple different things as far as our own stats, but very interesting to see how this has been shifting and

changing. And I've also seen an uptick in um uh referral traffic from AI searches to our website based on you know obviously tracking. So I did this thing where I I don't know if you know how to do this but inside of searchable you're able to take your website put into searchable and it does like this analysis of your web pages and how healthy or not the each page is. So with us we looked at a 100 top pages and we found things like this like there's no canonical tag on a 100 of the of

the pages that we analyzed on 59 and 59. there's no article schema and then 97 of a 100 there's no breadcrumb schema. So the cool thing about searchable and again I'm not trying to be positioning searchable. This is why I'm geeking out is that not only can it analyze that these are issues that I need to fix but it can also give me the schema or the code so that I who I'm not I we use web flow and I'm not a web flow developer. I can give Um coding from searchable and then give it

to my web flow developer and say hey on these let's start with the first 10 pages on these first 10 pages has no canonical tag can you fix it can you put this in there so then it becomes a very easy copy paste thing for the web flow developer and it makes it much easier on her because otherwise she's trying to describe to me things that I don't understand cuz I'm not a web flow developer and then um I'm trying to Describe to her the problems and it just becomes this this challenge and I'm really

looking forward to when I have a little bit more advanced which Chris I don't know if it's already you're live so please forgive me [snorts] I don't know if um uh they have this already but there'll be a day where searchable is is um connected into web flow so that this becomes I don't have to go through a developer I can just start doing things inside of searchable which would be Really really cool so here's um a couple thing ah there it is Chris I knew it was coming so the average score by page type

so we looked at like service pages we looked at um generic pages kind of like the blogs or search terms And um [laughter] I know I know I do too. Um and then we had the blog. So we could see a score and uh the homepage right now is 85. So I look at this kind of stuff all the Time to try and figure out, okay, what is it that I need to do to optimize this? Because I'm not a web flow developer. I have a lot of questions as a marketing leader. I'm trying to

figure out, okay, how do I get this so that the website can perform better than it is now? And I can go deep with with searchable with the agent and be like, "Hey, tell me what's going on with the homepage and how do I make this better and what's the coding and all this other Stuff which is just like so helpful for me." Here you can see like citation volume. So like we started back in May 11th with down here with I don't know what's something something like that. And I can see like where um

these are by the way sources of where we get cited the most. So we have LinkedIn, we have Reddit, we have Salesforce, we have YouTube which I'll show you a graph of that in a second. But I wanted to focus on on one bind because back when we Started in May it was like 100ish or so and then um right now as you can see it went up to 8 800 900 right here. So we went from um Arjun I'm I'm I'm with you on that one. You know if I had I only have budget

for so many things. So there you go. Um anyways the we had 100 to 900 within a few months. So like the the change in how well our website is optimized. And by the way, you can see here the change in how we designed it was right here. It Just started to skyrocket, which is awesome. Um, and then here it's it shows you the types. So like what type of content is being used the mo and thought leadership like it's not like specifically here, but I would say like the topic guides. I do a lot

of thought leadership kind of woven throughout all of these, but it's just interesting how thought leadership was not one of the main categories. So you can see here like I don't mind the dip because this Is obviously from the data but but this is so helpful for me to see okay YouTube's being is like growing like crazy here which is great and then I can see one mine going from here up to here. So our website's performing better and better and then um then you see all these other sources so like you have Salesforce like

some of the blogs in Salesforce we have LinkedIn we do a lot of LinkedIn activity from um our own team posting and influencer posting etc Etc and then Reddit. So, it helps me know, okay, Reddit here, you can see is like actually going down, which is interesting. You have LinkedIn that's going up. You have one mine going up. YouTube's all over but doing good traffic. And then when I go to each source, I can look at, okay, what is happening inside of each um category. So, I can say, oh, YouTube video. And if I click

on this inside of searchable, it'll show me the specific YouTube Videos that are being cited. So, I can see how I can either optimize for it, how I can mimic it, how I can um have someone like, you know, go to a um there's several people who do like tech reviews on YouTube, so I can send it to them and see like, hey, can you I'm going to submit this tear apart my tech. Give me a review and kind of see what's going on. Um Chris, [laughter] Chris, I'm not a bot. I'm here for you.

Um but this it's Just fascinating to see like what you can do with this. So from a strategic point of view, it helps me know as a marketing leader where do I go for AEO results and how can I create more content. So again as a result of this thing this this metric I started to create a lot more YouTube content through the one mind uh channel so we could actually have you know more traffic hopefully is the hope. So okay so I want to point out some things we Found from research. So kind of

going back to one mind for a second, we looked at a half a million conversations of anonymized aggregated uh data between humans having conversations with AI and um what I found interesting is that in marketing they talk about how classically 5% of people in your market are in market to buy the other 95% are not um because they're not you know in the process of buying whatever. So the what we found as we looked at the Human to AI conversations is it's a little bit different in the metrics. So it's more like 73% of people

who are engaged and 27% of people based on the conversations we analyzed on on our website and other people's websites who opted into this study are engaged or showing buying signals. So instead of being 5% it's 27%. I bring that up because um Forester put out a report saying how 94% of people use uh AI search inside of their B2B buying Motion. And so that means that a lot more people are coming to your website when they actually do come to your website looking for different answers going deeper. So, it just helps me as a

marketer to think about, okay, I need to think about AEO content that's not just awareness focused, but also like more advanced um research focused or more advanced content so that people who are in different parts of the buying process can be optimized. And I'm not just Focused on the awareness people, which there's nothing wrong with that. I'm also focused on the people who need more depth. So, I was able to use searchable to say, okay, look at the conversations I'm having inside of my own one my technology on our website. here's the patterns based on

our own analysis. I need some more content based on this. And so I was able to create content really quickly with it, which is really cool. So again, this is um Julie, I Think you asked earlier, what is it mean by having both sides for us, it was about saying, okay, what's inside of the models? What what source they going through? And then I could use our own technology and figure out what are bios asking in their own words and figure out the pattern match between those two and see how many are asking awareness stage

questions versus research state questions and then be able to optimize for both which is again fascinating. Um, okay. So, the u I'm going to give you example. Andrew from Zoom Info, he's the CEO of GTM over at Zoom Info uses One Mind. And what's interesting, I want to make sure I give you some takeaways, is that they they obviously use One Mind. And based on the AI human conversations, they moved, I'm not going to say how many figures, a lot of money from their AEO SEO search results based on the conversations they had with AI

and humans to then move that Budget to where it needed to be because they had data that was um uh trend trend analysis based on what's going on inside of their own uh conversations to move the budget. So they're optimizing for what people are actually looking, asking, talking about to their own AI. So one of my takeaways would be is if you have a website and if you're optimizing for AI search using searchable, my question would be is that you have to think about what is the Journey of someone when they go from AI search

to um your website? How are you optimizing for that experience? And once you do optimize for it, usually I'd love to use one mind, but whatever technology you use, take the patterns of conversations you have from your own website and start looking at and saying, "Okay, what what's the pattern analysis besides someone asked about pricing or the basics? What's outside of that that you can use to then optimize inside of AEO and SEO so you can have better results?" So for them, one of the things they had um one of the things they had u

a big difference was moving the budget and seeing massive results as a result of use moving the budget which is interesting. So again we again kind of give you an idea like again I'm going to use some things to help help you with some things hopefully. Um in AI search query there's a study I have to find where this came from. Average search Query is like 5.48 words and you can use your own searchable data to kind of figure this out. We found with this what our product is called one mind superhuman and the average

term is around six six words six to 10 words and 23% direct questions. So there's there's this correlation or this match between how people are prompting and then how they talk to AI on your website. In fact I have this crazy study about the similarities between the two. Um but you Can take those two things if you look at both sides saying okay what's searchable? What's happening on my website? How do I match those things? Then you're able to see okay how do buyers talk? What are the words they're using? What's the what's the phrases

they're using? What's the competitors they're asking about? What's the pricing things they have? Like all the basics. There's patterns in there that you can use to further optimize for obviously Search. And so what we did is that we looked at obviously how buyers are talking. We looked at the prompts um of what was what was performing inside of searchable. I tune both of them to make sure obviously the AI has answers to what people are asking about and the AE search AI search is being optimized and I can win on both sides. So for us,

we've seen a lot of uptick in traffic as a result of actually listening to what people are saying to AI and then using It inside of AEO. So here you I'm going to show you a quick thing right here which is right there is so um we found inside of search bowl in 90 days we went from as you can see here from 3,000ish to way up to here to in July. It's come down a bit because obviously we're in August and so I don't have the data but we went up to 14,000. So you

look at the difference between three and 14 that we scaled like crazy. we had this big bump and then same with citations we had this Spike as well and partly because we had some YouTube stuff go live in July and some other things but that's it's interesting to see these patterns and then inside of searchable again to to tout you guys which I love each one of these mentions or citations I can dig into and figure out okay where did this come from what's the conversations going on in that blog in that YouTube video in

whatever that is and how do I use that information to create more content then Optim information more so I can just create this loop and wheel of of information so I can make sure I'm constantly um on board which again I don't know if anybody's been in marketing for a while but for me this is awesome because I have this instant loop of be able to optimize and be able to see results really quickly which is just fascinating so here's kind of like our um our mode is like okay what what's the attribution what what's

the gaps in our Mentions how do I do a model side by side so I can switch between how I perform in co-pilot versus chatbt versus perplexity Where does my ICP, which most of mine come from co-pilot because we work with enterprises a lot. So I focus a lot on co-pilot to see like what's the patterns in co-pilot versus chatbt. What's the sentiment? Like another thing that's awesome about um Searchable is you can see like what's the good things that AI mentions about the brand versus What's the perceived negative things and how how can I

as a marketer combat any negative narrative that's going on as a result inside of content or inside of my YouTube videos or inside of like when my when my CEO goes out and does interviews on podcasts I can I can tell her saying hey I see this pattern of negative narrative based on searchable that has this thing which is incorrect So make sure when you talk on this podcast to say these five things, whatever the case Is, which is really cool. And then I can use voice and then causation. I start to see this model

in this loop, which is pretty cool. So this is just kind of very very quickly what I've seen before. Like a lot of times in established plan, domain authority, head terms, large library, all these things are things you probably know about before. But where we sat was like, okay, how do we use our novel first party data inside of the human AI conversations? How do we use That to narrow our own questions into into focus? Then we can publish. We can see like where the blind spots are and then publish information that we didn't know

we needed because you don't know what you don't know. And so having a tool like searchable where you can figure that out is awesome. And then I can look at okay, what are the podcasts? What are the founders? What's the Reddit posts? What's the conversations happening? and how can I how can I Influence those different places in a positive way that we can bring value and then make sure that we're not invisible on a category term. [clears throat] So, pretty cool. So, hopefully this makes sense with you. Like a 90-day instrument search, we we did

not have a lot going on when I first started. I'm able to see like what pages that sell score the worst. So, we can make sure, okay, does anyone own this? We know what's going on. How our Model setting or structure narrative earns nothing. So, it's trying to figure out, okay, where is what we're trying to deliver delivering or not on AEO. And then lastly, the the conversation logs from your AI to human conversations on your website are like golden data sets that you should be using to optimize, which we have like crazy. So, it

it helps you remove guesswork from you and obviously using searchable to see what people are searching for, which is Insanely helpful. So, that is it. Chris, what did I miss? So, sorry. I'm happy to take questions. >> Sorry, I literally just grabbed a glass of water. >> It's all right. >> Um, thank you very much. Uh I suppose the um the the thing that I think was um like most relevant for everyone is that if you don't if you don't know how You're performing, if you don't know where the mentions are coming from, if you

don't know, you know, what is driving the citations and and you don't clearly have an idea of the sources for each model, you can't optimize towards your search. and and you've seen that across two companies now >> is that >> yes >> the first thing you need to do is establish that map you know that that Road map of of changes and then whether it's you know feeding that back as someone said in the chat to you know in PRs to the team or whether that's feeding content or whether that's feeding you know technical changes

that is like the most crucial thing um for everyone before they take this on because as I said in my own presentation optimizing for chat GPT isn't the same thing as optimizing for you know Claude or Gemini and so it's >> right right >> critical to understand the differences between those two things. >> Yeah, I totally agree. It's it's been what I love about a AEO is that you can I mean at least with me using search bowl I can see results from different strategy changes within like a week. It's pretty quick. So, like when

you start to move things, you can see this dial. Like to me, it's kind of a game. So, I'm like, "What can I do today to Get AI search to pop up and then you can obviously see results inside of your own uh search traffic and and website demo requests and all other good stuff, you know, which is fun. >> Is it It's really interesting. I um on Searchable, you can see your exact AI traffic. So, you know, the the the bot traffic, um which pages are performing best, etc., etc." And a lot of the

kickback that you get on AI search is because you can't attribute it Perfectly. And you know, I've been in marketing for a very long time. For 12 years in SEO, we drove traffic and clicks towards websites and that's how we tracked it. Whereas I want to give people a frame of thinking here for a second, which is um previously when I was using Google to book a holiday or buy a product or whatever, I would open loads of links and I would go between loads of things, therefore creating traffic. Whereas now, if I'm asking uh

Claude or Chbt a question, I'm getting the answer to the thing that I'm interested in there and then within the chat. And so it creates less traffic. It's not that the demand has gone anywhere. And so to Jonathan's point, um there was a study from Adobe recently on um uh like retrospective uh attribution. And so they were saying that um you know one to two% of their traffic was AI which felt really unmeaning you know like a not helpful however when they Asked people who made demos for their tool it was 20% had come from

AI search and so you can see that there is a massive difference right now in um the uh like attribution and I can tell you now that I've been in marketing digital marketing attribution etc for such a long time that at the very beginning of Facebook ads, at the very beginning of Google ads, like the attribution was not perfect. Um, and yet it was it was the companies that that innovated that, you Know, took on the new frontiers that made changes that did the work that when attribution became clear, they were so far ahead of

their competitors. Um, and so yeah, I want to thank you very much for coming on and for speaking. Um, thank you for being a early advocate of our tool. Um, and yeah, thank you very much for presenting. Of >> course, my my pleasure. I was honored to be involved. So hopefully hopefully it's Helpful. I can just say again, I'm going to pitch one more time, but like if you're looking to have a tool that does results fast, not just analyze, but actually produce results, you can actually get more traffic like Searchable is my opinion by

far the best on the market. So, which is why I've done it twice. >> There you go. >> Thanks very much. Um, yeah. Cool. Awesome. Um, so just as a reminder to Everyone on the call, if you have questions, please put them into the Q&A and I will do a Q&A at the end myself which um I will ask answer all and anything to do with um AEO. Um to give you some context, I work with everyone from small, you know, 8 to 10 person startups through to massive global enterprises. And even at the enterprise

level, and I'll give you a good example. We worked with like one of the biggest um out ofome furniture and electronic Companies in the world, a multi-billion dollar company. Um and working with them over 60 days, we saw citations for their product pages increase by 100%. And that comes through really intelligent um identification of um like content freshness um adding FAQs, understanding what prompts you care about and then having an off-site strategy. Um and so let me just um uh do this. Um and uh I've just turned the chat back On so everyone can chat

again. I just want to keep the chat off during the main parts of presentations just because there are thousands of people on the call and if everyone's messaging it's incredibly distracting for everyone else on the presentation. Um I I suppose to to that question to that point there though from the chat um I really mean what I said before like over the last year I felt like almost quite bad in a way for particularly SEO Marketers because this new thing has happened you know AEO has happened and it is so big and it's so significant

it's complete change in how marketing has been done um and will be done moving forwards um and SEO sort of get it in the neck for, you know, for not being incredible at AEO out the gate. However, it's it's such a harder problem. It's such a more complex problem for people to solve, you know? Um there's so many more surfaces. Like I'm talking to enterprises who care about AI overviews, Gemini, AI mode, perplexity, chat, GBT, like they want to do well on five platforms. That's five different surfaces. And the sources that go into that aren't

just your website. It's many other websites. And so you really need to understand the um the backdrop on what is being used. Um and just to remind people, you know, the the the searchables growth has predominantly come through we were I think the last of The market. So I think we put our payment plans live in January or something. Um and it's the main reason for our growth I think has been in which prompts you actually select. And so um pulling uh clickstream data so like actual prompt volume on how people are searching these prompts

and we're collecting that from different clickstream providers um and plugins that record keystrokes. So we get enormous amount of information as to um What prompts being searched. We then use GSC to model a prompt universe. And so if you're a brand out there and you work with another provider and you're tracking 300 to 500 prompts and it's costing you 30 to 50k a year, book a demo with the searchable team and one of our consultants will create you a prompt universe and it will show you that what you're tracking is like this small in comparison

to the demand that exists in the market for you. So yeah, so during The presentation I will pause the chat for a little bit. Um, and so I want to welcome Luke if he's there. Hello. How's it going? >> What is up, dude? Um, how are you? >> Very good. Great to see you again. Very excited for this and I've already seen I caught the end of the last presentation. Looks like you've been diving in lots. Lots of great questions coming up. So, it should be a lot of fun. >> Yeah, absolutely. Well, look, I

I'll Hand over to you. I'd love you to introduce yourself. Um, most people will know what 11 Labs is, but it'd be great to have a oneliner so everyone has the contacts. Um, one of the fastest growing companies around. Um, and therefore we are incredibly privileged to have you here. I can't imagine how busy you are. Um, and so without further ado, I'll hand over to you to introduce yourself. Um, and can't wait to watch. >> Perfect. And can you see my slides? >> Yeah, absolutely. >> Perfect. Cool. And it's full screen right before we

get >> perfect before we get get too far in and realize we've uh knocked out. I have this joke which is I think somehow AGI is going to solve everything apart from audiovisisual setup that seems AGI resistant but uh great to meet you all. I'm Luke. I lead growth at 11 Labs. Our uh give you a quick introduction to 11 Labs and then we'll jump into some of Our lessons learned from our SEO and website teams which are functions we take really seriously here at 11. So the first one is I just want you to

imagine reading this phrase in the voice of Morgan Freeman. So hopefully you can all imagine that it's kind of that deep booming sound. It's that very dramatic and cinema cinema experience that we we all experience and enjoy. And what I really want to illustrate with that is just Every voice has this incredible depth of emotions, these pauses. It has different uh dialects, languages, cadence, mistakes, pitch, pronunciation. So it's every voice is so different and it's almost like its own fingerprint. And 11 Labs actually got started with trying to crack this problem of can you make

humanlike voices and this isn't an old problem. It's sorry isn't a new problem. It's a very old problem. We've been trying to do it since the 1700s. We had These wind powered instruments at first. We then had these digital synthesizers uh which were fantastic for accessibility but it doesn't sound like the same Matthew McConnA bedtime story on the car map. Uh then we had the Siri like voices where it starts to get interactive but you can still tell it's not human. So this was this big unsolved research problem and 11 Labs and Pota our CTO

he was the first one to crack how do you apply transformers the same Technology behind LLMs to voice and make these humanlike voices and since then we branched out into a number of different platforms. So we have 11 agents for building interactive voice agents and chat agents either for support or you can actually build it into your own app and website. We have 11 creative which is um a marketing suite uh being able to use to create lots of images and videos and voices for different ads or brand marketing videos and we have 11 API

so You can actually use all our text to speech or speech to text or music models uh directly via our API and we're working for working with lots of incredible companies maybe a few of our customers are already in the call and so it's been going going well so far And and what's amazing about SEO is I actually argue in terms of all the AI labs, we're actually probably the lab that cares about it the most. This is our landing page and right at the start, This was when we first launched. You can see we've

got the SEO phrases we were already optimizing for like text to speech, AI, voice generator, text reader. So this is something we we really care about as a company. We have a dedicated SEO team uh right from the early days and want to share a few of the the learnings that we had and probably a little different to some of the other talks today. So um I'll definitely talk about some of the AI Parts but our overall philosophy is that lots of the foundations what makes an incredible high conversion and a hightraic site in the

current age is still the same like strong foundations. and I'll share a few uh funky tips and tricks we've learned along the way and then um also talk about what happens in the much more AI AI native world. So we'll start cover a few of the lessons learned. So one's treating your website as a product. Then we're going to how to Use launches to maximize your SEO, the importance of agent DX or agent uh developer experience and finally uh building the growth or and what does the team look like to actually put this all into

practice. And so starting with teaching your website as a product. So overall I have this philosophy that most companies underinvest in their website and the website. This is something I learned when I was back at Post Hog, which has this incredible um web page Here, which if you haven't been is completely wacky and out there, but the overall thing is your website is actually your best salesperson. And so you're doing all this work to drive traffic to it or um you're doing outreaches on LinkedIn DMs, you're everyone's coming back to your website. And your website's

not only your highest performing salesperson who touches every single lead, but it's also um how it builds your brand for trust, for Partnerships, it helps you recruit new hires. So most people, I think, chronically underinvest. So we have a full website team which is staffed to our website to make it um both high conversion but also uh designed in a way which is high trust and we're really proud of. So this is the 11 lab site on the left here. And when we actually think about the website, this is something I learned from our head

of design, which I really Like this phrase, which is your website is a billboard. Your product is a tool. And when you're thinking about it as a as a billboard, it's about the design and copy. It should grab attention. It's this kind of large visuals. It's kind of the eye-catching fun copy. Whereas the tool is much more kind of stripped back. It's opinionated. It's uh getting out of the way to enable the users to do the work and actually get it done. Um, another big bit we think about is once You've driven traffic to the

site is how do you actually bring this wow moment as soon as possible and particularly in this age of AI where new models bring new capabilities, you can actually have this wow moment be something which the users can experience right on the site. So if you're 11 labs, how do you get that voice there that's something they can interact with? if you're searchable, how can you give people that uh that sprinkle or spark of just what the tool Uh can do for them to help them rank in a so to give you two examples. One

is we do lots of these very interactive demos and our websites in general follow this pattern of it's kind of like the hero has this interactive component to really get people to start to engage and have the wow moment and then below is often like more SEO optimized. it's going into more detail. Another really fun example I want to mention is our AISR. And so as you go Through our contact sales flow, uh actually as soon as you enter your details into the contact sales flow, you can then speak to one of our voice agents.

And this voice agent, it will ask the normal STR screening calls. So do you have a bud budget authority need timing? It will then actually grab all that data, send it to Clay to enrich it and then work out, okay, are you actually a lead which we can skip and go straight to NA to book into the Calendar, show you a cow.com link. Do we actually need to still do an extra step with the STR or do we reject? And one of the very cool things about having voice agents on the site is it being

able to be multilingual follow up on questions. We famously don't really have titles at 11 Labs. How do you then actually work out does this person have authority? it can kind of ask those follow-up questions. Um, and then as we're thinking about Bringing the wow moment and creating these these uh high engagement landing pages, which ultimately is what's going to serve the user's intent for AEO or SEO, and here's a few other things I would suggest thinking about. This one I think is often underrated is is there actually a self-s serve version of your product

that you could use um or like a new self-s served version you can release. The other one is interactive demos. These demo builders now are Really good for actually being able to uh record your different UI have the user talk through and the other one is AI agentled on boarding and so particularly for more like complex or technical products yes you can staff an FTE team but is there also an agent you could build which could analyze people's codebase or setup and help work out how to inter in integrate it and finally there's this fun

goal we always set ourselves when we build any web pages Which is how do we make each individual landing page so good that it itself can go viral. And so this is one for our music where we invested the time to actually have this really fun component where you can try out different songs when we released it. So people are going around clicking on the site increasing the the time they stay and helping train up the different models that yes, this is the the site which should rank at the top of Google. Nice. I'll jump

to the the second lesson which I'll chat through which is about launches. And so um at 11 Labs we really care about launches and in particular this is a great way that we drive those initial back links for our SEO pages to help get them to rank in the indexes and again kind of linking back to what I said at the start. So I still really think lots of the SEO fundamentals really do matter because it's the same indexes which are used by people as well As by agents. Um so the why now alpha launches

so we've we've had software and SAS for a while and it was never really uh interesting enough to go properly viral but in this age of AI that we live in different new models unlock new capabilities which give you these new viral moments and so to give you an example um if we if we release a new speech model the first speech model we released was actually the first AI voice Which could off and so how could we actually capture that into one of these moments which can go viral and that's something we think about

every time we launch one of these products and when you think about virality it can help get you the social mentions as well as the back links and the press um so why do these launches matters well I think they're really fantastic motion within a company where you can drive urgency and so you can say okay you as a As a as a leader marketer can say this is the date that we're going to launch this product and you can rally the whole company around it. There's also the clarity side and so that you can

get both sales, marketing, engineering product to all be looking at the same launch at the same time. And then there's the revenue piece. So hopefully if you do it right, you'll get your first first revenue, your first leads or subscribers coming through. And really Importantly SEO. So when you actually first launch, as lots of you uh I'm sure know already um or or starting in your SEO journey, domain authority uh does really matter. And so being able to actually get those initial initial backlinks, initial users to the site is critical and people searching for your

site. So I'll go into a little bit more detail on how we think about launches. So there's a few steps. So there's first of All there's the messaging. So how do you really define what exactly you're launching? Then you work out the different assets and then the distribution. And so broadly we think about three tiers of launches. We have tier one which are the new product lines, the big fundraising announcements. Tier two is often major features and and again new features may unlock new types of SEO searches or SEO volume which you didn't Have before.

There's case studies. Uh there's KPI milestones as well which are a fun one to launch. And then tier three is the more incremental bits where yes we have like a stripped down case study and but and we're often do a refresh of our existing pages but we're not not something we're putting loudly on socials. And so here's an example for Scribe which is uh the most accurate speechtoext or transcription API that we launched. Um, and here we're like Actually defining what the name of it is, what's the audience we're going for, the different value props,

the primary, secondary, and the goal. And when we're defining and doing this like product marketing work, we're really often thinking about SEO in mind. So, we're like, okay, well, I remember one big discussion here was, what's the term we actually want to go after and want to try and win with Scribe? And there's actually two we could have tried to go After. One was speech to text or speech to text API. The other one was transcription. And so we're doing the digging. We're like, okay, what's the different volumes for these different search terms? Which is

searching more? Turns out transcription has much higher volume, but it's often people searching for transcription in the context of DNA, RNA, biology style transcription. And so we were going through like, okay, what are the actual top results of each of These things? But then like, okay, we really need to focus on speech to text because that's the one we want to search and like actually seem to translate more often to to people's intent and where we should be showing up. Another thing we realized when digging into this is there's actually real time speech to text

API and non-real time and they also have their normally non-over overlapping set of search pages and results. And so again, we decided actually two separate Launches with different landing pages. So you can see how like the launches and the SEO strategy are really tied in together. Then we build the landing page and again using this format that we spoke about before where we try and have a really engaging high performance hero. This is normally a custom block and then more established blocks that we use below. And we you can see some of the different keywords

we're trying to trying to map To uh for the SEO page design. And on a on a side note is I think with AEO there will be certain types of pages which end up getting more traffic and certain types of pages which will be getting less traffic. And I do think for now some of the like tools pages are ones which will be like more traffic now. Maybe in future as you get artifacts and the MTP marketplaces that will change, but we're we're really bullish on tool pages still being incredibly important. Uh then we do

the the social post. We try and get that to go viral. That's a great way to get the the mentions linking at the bottom to our landing page. Uh and then we do the video which I think the time are skipped today but would encourage you to to go check out in the uh on our YouTube channel. And for the videos we really find this is a great way to package often very complex products into a high performing and engaging way. Maybe if there's time we Can always jump to this at the end. Um and

then there's distribution. And so everyone's checklist will look different, but once you've created your messaging, you've taken into account what are the different SEO terms, uh then you look for okay, what's the different assets and you build this high conversion landing page and do the appropriate backlinks. Then there's the actual like distribution and getting out into the world and I just highlighted a Few of them here which are like particularly important for SEO. So, if you can uh engage journalists or you have a press agency, that's a great way to get high domain authority backlinks.

There's also the social ones again, which is often a signal being used, particularly on something like Reddit. Um, and broadly, yeah, just making sure you've done your homework of which are the actual keywords you want to rank for. Often there's a few uh which you Could do. Cool. That's the second lesson. So, how we use uh launches to help kickstart our SEO for new new keywords. It's also just a really valuable exercise for when you're launching new features, what is the keyword set or traffic set that it can potentially unlock. The next one I want

to chat to you about is agent experience. And so, broadly at 11 Labs, we and I'm sure everyone on this call agrees uh search is changing. And There's search, I like to think of it as like the search 1.0, which is your 10 blue links on Google. There's search 2.0, which is um often what we see with chat GPT and claw. This is agents recommending to humans where it's really important to use uh great AEO tools like searchable in order to actually track where your um where you're showing up, maximize showing up. But a big

bit we're often thinking about at 11 Labs as well is search 3.0. what happens when the Agent actually does the search, but it selects and it buys it autonomously. And so if you imagine you're a developer and you're building an app um and you're wanting to add a voice agent to it. So let's say you're a Dualingo trying to build an interactive course um or you're a creative tool wanting to add in speech. But what we want when the agent starts to search, we want it to see I should use 11 Labs, make the API

key, integrate that and actually buy on the User's behalf when they just say this is something I want to do. And so we really think about how do we design for this agent experience? And lots of this, if we don't write in the 3.0 world, also very much applies to the search 2.0 world when it's recommending to humans. So there's a few bits. A big one for us is making it API first and so agents can actually do everything that the UI wants to do. A big bit and where we've seen a good increase in

ranking is also Enabling serving of markdown for our pages such that when there's all the different crawling and searching it can responses respond as the markdown way. Another is we've done this llm.txt txt. So you can click on that and it can actually give a much more agent navigable site uh which is increasingly becoming a standard. We've released MTPs as well so that uh it's much easier for agents or humans through these uh tools like claw to be able to actually Interact um and then we measure it. And so just like everything else we treat

as a system we try and track what are the different inputs and outputs. We do a bunch of experimentation and but in particular we're looking at like the crawler level logs of uh are people actually finding these changes and improvements. One which I'll just show you was this like markdown page request we've really seen has start to increase um and I think this is a good sign that Some of our uh AEO type search is doing and AEO is in fact our one of our fastest growing channels. And one fun one I just wanted to

mention is that we're really thinking about the future of this search 3.0 and it's not only will agents maybe it's coding agents or a claw co-work not only will it go and kind of discover it but it will actually hopefully do the like payment and then integration on the user's behalf. And so a really fun one We're thinking about is how how do you actually enable this agent payment kind of two ones to put on people's radar. One is the X42 which is this actually old uh code about a payment being required um which has

now been like taken over by Coinbase and a few others to taken over in a good way to actually change it. So it's a crypto protocol where you can say hey I want to use this and it goes okay it's a certain amount of crypto per API request. There's a similar one where stripe agentic uh commerce is doing. So we've wor with stripe projects where you can actually procure an API key through that and so these were like in early expiration for these but we think this is a wave which will come early and we

really want to ride. Cool. Uh the final lesson I want to talk about is building the growth org. And I've had quite a few lessons of uh of setting up the SEO team here, SEO and Website teams that I want to share. So I'll start with the general general advice and then we'll jump into the SEO specific. But in general for the growth teams at 11 Labs, I think about there being generalists and specialists and they require different types of uh hiring processes. And so we actually only have a small number of specialists. SEO like

particularly the person who's obsessed with the data the Kr logs you need the specialist who who knows what They're doing and once you get to the stage where you're like okay this is a channel where we really want to invest in and so uh in general what I do for specialist hiring is one work out the companies that are really great at the thing and often I'll be like so sleuththing on uh what were their SEO results in Seamrush how are they actually growing is the company great at that two are then work out who

specifically was responsible for like The key inflection points in SEO on the company's journey. And then when we do the call, I like to screen for enthusiasm and for work efforts. For generalist hiring, it's quite different. It's more I think about as describing and creating obstacle courses, uh calibrating it to make sure we're getting great people through and then grow the pipeline. So uh again practicing our our marketing muscles of how do we get the candidates in from Tofu mofu bofu and if you look on our careers page we have like a nice video which

we made with a6 which gives people a feel for the company and here's just a few of the like for the generalist parts that I really think about are particularly important kind of general traits across the growth team that we hire uh so copywriting quantis and some of the backgrounds are often like exfounders PMs, consultants are the more generalists that we hire for growth Marketers. And then in terms of the org structure, uh what we generally think about is each person owns one KPI and then that kind of cascades throughout the org of how you

uh break down the KPIs, but they build back up to that northstar for us, which is um self-s serve revenue and a B2B pipeline. And uh so in general we then grow the team in proportion to the revenue or the size of that KPI um and then Actually increase the size of the team as that grows and and allocate people to each individual bit. And so often we're start with a fraction of a person. And the mistake I've made time and time again is actually as quick as you can, you want to get to a

full person who really owns and obsesses with that channel. So if you're in a a company which is maybe venturebacked, you you have conviction SEO and SEO will work for your company, maybe there's other Competitors doing it in the space, then it's like I would argue as soon as you can, you want a full-time person who's just obsessed with SEO day in day out because hopefully quite quickly you can prove the ROI and then once it goes from one person when the the impact of that channel and that revenue is big enough then that grows

into a full We then have a matrix of these. So the different KPIs kind of break down by products, by channels, by geography. Um, And when we then think about the SEO team and how that works is we have this head of SEO, which is one of the first ones we hired. But then under them, we have different SEO uh folks who are just focused for the different product lines. Um, and in future, we may start adding in the the geography parts as well. And so to give you a sense of what our team looks

like on SEO at the moment, um we have a SEO data lead who's that that specialist type. We also have an SEO Content lead. We have three full stack engineers who are really trying to make these really fantastic pages which are high conversion. Uh we have a website designer. We then also have a more general website team which is a more brand focused team and work with the product marketers through the launches. So website designer to website engineers. And then we have a CRO team. So that's three growth engineers and a growth PM who actually

work every we've Just changed this up recently, but they actually work that full journey from the website all the way through to onboarding subscription payments. And hopefully the main thing uh some of you can take away with for this slide is arguing for some extra headcount um and being able to invest in SEO more. I think we How are we doing, Chris? Should we skip this one or go straight for it? >> A couple of minutes. >> Couple of minutes. Perfect. Um and yeah, Final one I just want to talk about is um the turning

every marketer into a marketing engineer. And so, uh there's this lots of hype at the moment around vibe coding. And I really think the potential is huge, but also in a in a company context, it's really important to like do it properly with high security for anyone who's at larger companies. One thing that we found works really well is setting up this new function called Automations engineering which actually enables everyone else to uh all the other marketers who aren't engineers to be able to actually maximally use their tools. And so a few we found helpful,

Love Bull, NHM, and then taking a friendly engineer in the company and turning them into one of your automations engineer where they actually like set up the platform, help review other people's NHS and lovables before it goes live as well as proactively Building. We've also set up a claude skill which can help review a lot of this and help the non-engineers to design stuff correctly. And then we've built a little approval workflow on Slap which seems to work well. But um nice. So that's the summary of some of the lessons learned we've had around website

SEO SEO. We've still found lots of the um foundations of SEO and and ranking uh still really matter. SEO is one of our top performing channels and they are one Of the fastest growing. Um, so to recap, the first one is treating your website as a product, really investing in it, making something you're proud of and high conversion. The second is the importance of launches, particularly for going after new sets of new SEO keywords. The third is this important of agent DX uh, and I'm excited for when agents will be autonomously buying as well as

just searching and consideration. and then how some of the Learnings we've had at building the growth or and hopefully giving you some ammunition to get some get some more headcount to go after SEO. So, thank you. >> Thank you so much uh for for doing this for us today. I think that um I think that people for for some context like it's quite hard to get people uh on calls who lead GTM for companies that are worth you know whatever it is 11 billion Something like that you know 500 million ARR um so thank you

so much for coming on and giving us your view um that was fascinating so thanks for joining us dude >> yeah no problem thank you so much >> take it easy um okay everyone so uh our Next talk um is uh like very uh exciting because uh as a lot of people know I have been LinkedIn obsessed uh for many many years and I would say my LinkedIn audience has uh been the The reason that my businesses have succeeded so well over the years. Um you know building such a strong audience and and having such

a a great community of business and marketing people. Um so I want to welcome um Andrew uh to to talk today and introduce himself and talk about um how you know LinkedIn um you know coincides with with AEO. I I just before I wanted to say like from a you know this is this is anecdotal the data backs it up but just from a personal Anecdotal point of view. We did a bit of research on my own LinkedIn in the last few weeks and the amount of my posts that are now used as sources for

AI answers around AEO is quite literally staggering. And so um you know for years now for many years I've told people that um you know as a founder as an entrepreneur to not build your personal brand is a detriment to you because you know it converts so much more than your um than your business brand. But now we Are seeing chat GBT and other channels uh using the posts we create on a daily basis to answer key questions within LLM. I I find that to be um like the coming together of two worlds that I'm

you know completely obsessed with. So, I want to welcome uh Andrew to the stage to introduce himself, VP of marketing um for LinkedIn and tell us a little bit about um his world. >> Hey Chris, can you hear me? Can you hear me? Okay, >> I can hear you fine. Yeah. >> Amazing. Thank you so much for having me. Hi to the Searchable crew and to everyone on the call. Um it's my first time being on Searchable, so I am delighted that you folks have had me today. Um so, as Chris said, I'm VP of

marketing at LinkedIn. I'm based in London and I guess I in this conversation today I want to pull back the curtain and try and help in any way that I can um from what we've learned in The over the past year. Um this isn't like a theoretical conversation. Everything I'm going to share comes directly from our own testing, our own learnings and the wins that we've had on the LinkedIn platform. We've built playbooks and actually in fact we just published a guide yesterday which is called unlocking AI search visibility. It's the B2B marketers guide to

LinkedIn. So, it's designed for B2B marketers who are actually new to Optimizing their content. Um, and so it's going to share a lot of examples of what we've learned over the last 18 months. And as Chris said, you know, there's a lot that you of leverage you can get from LinkedIn. Obviously, Chris has a brilliant and amazing profile. So, obviously, he's going to look um he'll get cited a lot, but um LinkedIn was it can give you a lot in terms of your discoverability. So, I want to share that with you today. So, I am

going to Start to share my screen like hopefully. Can you folks see my screen? >> Yep. All good. >> Yes. Success. All right, Chris. You're going to I'm going to ask you to give me like uh when I'm down to my last 10 minutes in case I go too far because I get very excited. >> Yeah. I'll give you I'll give you I'll give everyone a fiveminute warning. >> You'll play the Oscar music. Yeah, I'll Just I'll just come back on. [laughter] >> Sounds good. All right, so let me get to it. Okay, so I

guess so many of us are on this call because we all know that one thing has changed and is like certainly true. The B2B discovery process has moved. Your buyers aren't starting their research on your websites anymore. They're using ChatGpt, Gemini, Copilot, you name it. They're building short lists before they ever reach your Website. And if you're not showing up in those AI generated answers, you are missing the beginning of the buying conversation. You are not in the B2B conversation if you are not there. And that's a that is a huge kind of starting point

for this because this affects anyone in B2B um and anyone who's in marketing. Um so what I want to talk to you about is over the next few minutes these four things. The new era of AI discovery, how to build a content system That's ready for it, how to optimize your LinkedIn presence for visibility, and how to measure whether any of it's actually working. Um, and I can tell you I have plenty to share with you, plenty of war stories to share with you within this. Um, so this is the beginning of where we are

in the AI discovery era. The numbers are pretty dramatic and you may have seen some of these already, but 94% of B2B buyers are using generative AI for for Research. Um, Gartner predicts traditional search volume is going to drop by 50% in just a couple of years. And here at LinkedIn, we've actually seen our own non-branded search traffic decline by up to 60%. Not six, like 60. That last stat felt very real for us. This isn't something that is happening to other brands. It is happening to us. It's happening to everybody. Um, so if I

take you into how we think about AI discovery at LinkedIn, There's a shift that I think is important for people to understand. AI search isn't just changing where discovery happens. It's changing what it takes to be discovered full stop. Traditional SEO rewarded you for keyword words and backlinks. AI search optimization or AEO um rewards something different. consistent thought leadership, genuine expertise, content that's structured so AI models can actually extract and cite it. As my Colleague Mimi Turner said, the shift to AI search ushers in an era where reputation can be more intentionally built, not just

bought. It's not just about if you have the biggest budgets that's going to get you discovered. That framing really shaped how we thought about tackling this problem. So the seauite is certainly paying attention. AI search is a top priority for 90% of CEOs according to Forester And that pressure is trickling down the organization. 80% of VPs and CMOs are already asking their teams about AI visibility. But there's a big gap because only 30-ish% of B2B tech marketing actually feels prepared to execute a strategy. And 51% of them say that a lot of it is because

they just don't have enough internal knowledge. So the urgency is definitely there, but the knowledge often isn't. And that's exactly what I want to try and help to Close today. So one concept that's really shaped our thinking is something at LinkedIn that we call viability. It's it's how you get bought in B2B as a brand. In B2B, purchasing decisions require committee approval. As we all know, and the average B2B buying group has 13 stakeholders, nine external influencers, and that number often doubles when you're actually uh thinking about AI being involved in the process. AI acts

As a research tool learning from the content signals available online. And here's what most brands miss. Buying decisions are not purely rational. We all we know this like in our brains, but when we're doing our work, it doesn't always follow follow through. I I've been guilty of this myself in my team. LinkedIn and Bane research shows that 34% of enterprise B2B buyers rank self-p protection over price or features of a product. What that basically means is I Need to know as a buy as a someone who's buying, can I defend this decision if it goes

wrong? And that's the number one emotional priority that B2B buyers actually have. So what does that mean for marketers? Being viable means reaching an emotional threshold, not just a rational one. And so that means building trust. So let's talk about building trust. Credibility isn't built by one voice. It's built by a system of voices that Reinforce each other. Your brand sets the point of view. Employees humanize it. We saw sixfold engagement when our employees activated as creators. Customer advocates validate what you actually have to say. Customer advocacy is four times more influential than a product

claims their influence is supposed to be. And creators scale all of that. So 84% of consumers trust brands more than more when creators demonstrate and review products. Um, so When these voices align and reinforce the same themes, credibility compounds. And that's the stack, the credibility stack that we're building at LinkedIn. And it's the same stack that I'd encourage any of you to build. So LinkedIn's the number one most cited domain for professional queries and AI search, according to Profound's uh report this year. Beyond citation volume, LinkedIn content scores high in semantic similarity. What that means

is Basically so when LLMs like ChatGpt or Copilot site LinkedIn content, they tend to quote or paraphrase really closely. So it's your words, your framing and your ideas stay intact. Now also 75% of LinkedIn citations come from individual member profiles. Just like Chris just talked about just now and let me just give you some context on that. It's not the majority is actually not from company pages. CEOs only account For 8% of cited content. Founders similar number. So job title actually matters less than expertise. So that means individual contributors can earn visibility on a pile

of executives if they have a niche where they have a level of expertise. So the takeaway here is you you need to empower your experts, not just senior leaders, to post consistently on topics that you want to rank against. So let's talk about building AI ready Content systems. Now my my experience at LinkedIn when we built a cross functional task force, it was certainly not easy. Um when AI search arrived at LinkedIn we had to stop thinking in channels we had to start thinking in systems. So that's actually a direct quote from in Meccllin our

um senior director of digital marketing and it captures exactly what our mindset shift involved. AI visibility isn't a Single team's job. It actually draws on signals from owned earned social services. So we had to build a cross functional team. Um it was basically an AI search task force that we brought together across SEO, PR, product marketing, um social, you name it. But we brought everyone into a single aligned strategy. That task force was responsible for three things. Accelerating content activation, correcting mis misinformation surfacing In AI responses, and launching new content optimized for AI search visibility.

So was certainly not an easy piece but there was there was a lot that we learned. So we certainly learned very early on that when teams operate in silos our signals scatter and we were we did begin in silos. So we saw that happening and when they coordinate you build consistent reinforcing signals the AI actually rewards and I'll talk more Um about that. So, LLMs, as we all know, work differently from search engines. Well, traditional search engines. Google ranks one best piece of content per query. LLMs aggregate information across many sources and synthesize an answer.

That means a model is more likely to site your brand when several credible sources reinforce that same idea. So, one strong article is not enough. The solution is topic clusters. Build content plans around subjects that you want to own. Then create multiple pieces supporting each theme across LinkedIn and your other owned platforms. So using consistent naming, conventions, product terminology, messaging everywhere is really a foundational thing that you need to do. A practical tip is to replace marketing jargon with language your customers actually use. Clear, direct language is what you really need. Um, So think about

how AI models actually understand a stronger, more consistent signal. We'd recommend building your content around 7 to 10 prompts per topic. These should reflect what people actually type when they look for your services. Traditional SEO keyword research is a good starting point, but you can also use AI to actually help to refine that list. So think of LinkedIn as an amplifier for Your credibility which I was just talking about earlier on. Your company page establishes your brand as an entity for that AI models can recognize and trust. So when you have that as your entity,

there are two key things that can actually really help you. So one is articles. Articles are long form pieces of content that give you depth and authority. Our internal data shows that articles generally they generate around 60% of LinkedIn um content citations. Posts are similar but they're actually a shorter more frequent touch point. So you could have multiple posts around an article and they make up the remaining 40% of citations. So these are not competing. They actually are complimentary. They are a system and we'll get into how they work together in the next section. But

one insider note for you to be aware of is so our newsletter it's called this week in marketing and it reaches over a million People um and well certainly reached overund a million in its first year and it's consistently ranked as our top performing piece of content um each week. A newsletter is basically something where it notifies subscribers every new addition that you have. And AI models treat that newsletter as if it's a standalone article. So if you write regularly on a single theme, a newsletter format can actually be a really powerful way for You

to generate a long-term audience because AI will think about that as if it's actually a single really strong article. So your content is only as visible as the voices that are behind it. So when multiple authors contribute to your topic clusters using shared keyword themes, you create depth and variety that a single company page just cannot match. So this is the thing you really need to be thinking about in your Organizations. Encourage executives, ex product experts, everyday employees to post consistently around your key topics. The more you can generate consistency around the things you want

to be known for as an organization, you're not just amplifying reach, you're building a network of credible sources that all point back to that same expertise. So look beyond your own's channels as well. Placements in external Publications corroborate your expertise in a way that self-published content can't. So when a respected third party source reflects the same themes that you're building on LinkedIn, it validates your authority. So share those ex external placements on LinkedIn as well um either by posting them about um by yourself or reposting content from an external publisher. So, leveraging visibility on LinkedIn.

So, we've talked about articles and Posts. Before we get into optimization specifics, I think it's important to really understand these two formats because they're pretty core to your presence on LinkedIn. Articles are long form pieces, generally between 800 and,200 words. They're built for thought leadership, for frameworks, and for deeper analysis. They're more likely to be cited as a primary source by AI models. Posts, as I mentioned, are shorter, so they're typically 200 to 300 Words. They're built for timely insights and they're conversation starters. They often act as a secondary pathway pointing towards that longer form

asset. And articles basically earn authority and posts essentially distribute that authority. So, our research into the top cited content revealed four pillars that consistently separate what gets cited from what gets ignored. Pillar one is really about structured content for retrieval. LM reward content that they can easily extract and quote. So, you need to frame your content around a real question that your audience is asking. Lead with your most important keywords in a title or in an opening line. Use direct questions and subheadings. Then answer each one in one or two clear sentences. Use a short

summary or a checklist. Models often pull from these um and they basically Kind of take actionable lists directly into their answers that they provide. So that's the that's the first bit. But these are things that we have tried. So I want you to be aware that we've done this and this has definitely worked. So you can see the importance of structure when looking at the top 24 most cited LinkedIn articles. 100% of them use bullet lists and numbered items. 75% name specific companies and tools and 67% include hard Numbers and data. So please think about

that as like your structural musthaves if you want to kind of build that content recipe to be cited. So, here's pillar two, establishing credibility. LLM's favor sources that they trust. Staying topically consistent, timely, and relevant content on a focused subject is far more likely to be cited than scattering posting. Uh, I know that can be a temptation sometimes to just Pick up on the newest thing, but what is it that you want to be known for? What are the things you want to be known for? So share original insights, proprietary data, firsthand experiences. Authors with,

let's say, 2,000 or more followers tend to have content surfaced more frequently. And that's because follow account tends to signal some level of expertise, but you know, two thou that doesn't have to be a huge amount of followers, per se. So pillar three, drive engagement. Engagement is both a relationship builder and a trust signal to AI models. Highly cited content typically earns 10 or more thoughtful comments. Reply to comments to keep the conversation going. And remember, 95% of citations on LinkedIn come from original posts, not re-shares. So, your original voice really matters. Pillar four, maintain

freshness and discoverability. Publish articles weekly and post at least two or three times a week. Refresh older content with inline edits or new comments. Doesn't mean you have to ditch them. You can build on them. Include specific dates where relevant things like top AI tools for 2026 signals recency or check your profile's public or your content is visible and comments are enabled. If LLMs can't access your content, they can't surface it. So please, everyone on this call, please Promise you will go and check that um your content is actually visible and your pro your profile

is public if you are actually wanting to be cited. Um Melt Waters data backs this up. 48% of cited content is less than 3 months old compared to just 12% for the um of content. So answer block formatting is really important. This is really the like do not bury the lead. If articles are your Depth play and posts are your distribution play. Answer blocks are the formatting layer that makes both easy makes it basically easier for AI models to site. So like start with a concise answer block. Provide deeper context as well after that and

use optional labels. I know this sounds very um regimented but it really will help you to build what you need. Um the idea is basically simple. Start each major section with a 30 to 80word answer block that addresses the topic that's named in your heading. then follow it with deeper context, examples and sources. You can even add a label like quick answer um before your summary that what that does is it it signals to AI that actually this is a direct response. This follows the same logic as anyone who's done traditional SEO pillar and waterfall

approach. Start with the definitive answer then let the rest of the piece break it down in a structured Way which you can see in this example here. It's one of the ones that we've done at LinkedIn. Um, and like we we have had to learn and test. And this is something that what I'm sharing with you right now is something that we've done ourselves and we continue to do in my team. And here's more information to kind of think about how you use a mix of bullets and definitions that actually speak to your audience. So,

you can see there's Different examples we got here of who initiated the buying process, who evaluated options. you know you those things you're adding in there are all being signaled to AI as well. So your article checklist really really important. So you've got two separate things articles and posts. Please please I'm sure we'll send these to you um these slides to you after but your article checklist you must start with a questiondriven title that doubles As your URL. Add a TLDDR summary near the top. Like that might feel like it's overengineered. This really, really works.

Um, with your posts, write a clean opening line that's built around your core keywords and frontload your um, main takeaway. Keep to one idea. There's a temptation to just throw in things. Please keep to one idea and publish two to three times a week. So, excuse me One sec. So, if we look at how articles and posts work together, here's the workflow that our team uses at LinkedIn. Publish one strong article as your like main citation worthy anchor on a topic. Break it into three to five posts and that's where each post takes one idea

or a statistic or a framework and gives it room to breathe in the feed on LinkedIn. Use those posts to test angles. See what works. Like try which and see which Ideas actually spark the most comments. Look at the engagement. See what's actually resonating. And then fold what you learn back into your next article. So articles remember earn authority. posts distributed. So I'll do the final section because I know I have like eight minutes left. So I'm going to jump to it. Um so measurement, I'm sure folks might have questions about this. Um so I

think the first the the number one thing I would Say was a learning for us with measurement was we were very keen to see results incredibly quickly. Um and so it's really about setting realistic timeline expectations. There's a truth that protects programs from being cut too early. AI visibility shifts rarely happen overnight. And I think it's important for people just to be realistic about that. According to Profound's research, the median time for Claude and Chat GPT to site a new page Is almost 7 days. And for 90% it's actually 37 days. So a significant proportion

of new content takes over a month to get cited. So, if you're not that 7-day example, you could more than likely be the 30-day one. And that's just you may see how that changes for you, but be realistic about how quickly that happens. But there are three reasons why retrieval signals can lag. So, indexes refresh on their own schedule within chat GPT and other Places. Um, AI models need multiple corroborating sources before registering consensus and AI models update independently. So visibility actually shifts um not gradually, it shifts in steps. So please just like regardless of

what the reasons may be, just know that you need to set realistic timelines. So the most common mistake is actually declaring a program a failure at week three. Um so you've got flat cit flat citation numbers in the first couple of Weeks. um and then you make a verdict and you say it's a failure. Uh my advice to all of you is to please allow 30 days before evaluating AI visibility impact and judge outcomes against your baseline at day 30 and actually maybe even beyond day 30. So three category measurement framework most teams monitoring AI

search pick one headline metric. uh it's usually citation count that people tend to focus on and that means that if you do that You miss the structural patterns that reveal whether a program is actually working. We'd recommend grouping your metrics into three categories. So leading indicators move quickly and predict outcomes early. So on LinkedIn, these are your engagement signals. Are you getting post impressions, reactions, comments, and shares? Strong engagement suggests that your content is more likely to appear in AI generated answers. But that's not like 100% proof. So don't treat these as the final scorecard.

Outcome metrics reflect actual results like citation counts. These move slowly and they lag behind the work. So you should track brand mention, citation, share of voice, but don't report these weekly like I've mentioned already like you know that those are outcomes metrics that can take time. Diagnostic patterns reveal whether your program is structurally sound. These are patterns in your data that Show what to change when results are not in line with what you're expecting. So it could be, you know, top content formats, high high performing prompt uh prompt types. These are things that you learn

like we learned on this. It's not different like we went we were in a very similar situation and we still are. We've not figured it out. >> Just to give a a little uh reminder of time. >> Okay. Three minutes. >> Yes, sir. >> All right. Thank you for the prompt. Um, all right. So, our new play our new playbook goes into a lot more depth about what you can measure. So, actually I'll actually not go into a lot of information here, but just a signal to you to check out our playbook. It I it

really will help you. Um, and so what to prioritize? Don't fall for universal best practices. What works for one brand doesn't necessarily work for yours. It's Crucial to test and just look at your own data. Um, our checklist for getting measurement right is pretty straightforward. Get a cross functional team on board AI search citation tracking platform. Most importantly measure sentiment and downstream business impact. So I guess my last slide to you folks is these are the things I would say and my takeaways to take with you. So you know discoveries moved look at the fact

that um articles Do heavy lifting. They are a thing that can be a core asset for you on LinkedIn. LinkedIn is a big amplifier for you if you use it in the right way. Um, short form posts can keep you fresh. Systems beat one off posts. And the last point is measurement looks different now folks. So that is it from me. Thank you so much for having me. It has been lovely to meet you folks and please feel free to ping me on LinkedIn if you have any questions. >> Very much uh appreciate you uh

coming on and talking to us. Very useful. Um, and as I said in the in the chat a moment ago, it's like my two favorite worlds coming together, AEO and uh, LinkedIn. So, I'm very glad to have had you and thank you very much for for speaking for us today. >> Thank you folks. Enjoy the rest of the day. >> Take it easy. Um so ladies and gents I mean the the You know something I've been talking about for the last um year or so now which is you know whether it's um you know obviously

uh specific to LinkedIn in that exact presentation LinkedIn is a huge source um and and it really does facilitate two uh two core priorities which is I can I can grow my business by growing my personal brand by getting my name my face in front of my business but at the same time I have the ability of being um um mentioned within within AI. I think that the um the advice uh as we go on through the presentation I will try and tie the advice through to how this is applicable to small businesses, how this

is applicable to small teams. I know we've had bit big speakers on today. I think it's like a privilege to be able to draw the type of speakers we've had on the call today. I don't know where else you could go um to just tune in to your laptop for a for for that type of quality of speakers. So, I'm really proud of that. But as we go into the Q&A, I will personally try and um deliver insights that are relevant at all stages of business because I can tell you now, you know, um as

much as like any speaker talks about how to optimize something, I think the way to show that you know how to do it is to do it. And so I've always told people about optimizing their LinkedIn for the last few years. Um, and I continuously focus and grow my own LinkedIn to show that I Know how it's done. Um, I think there's so many people who talk about um, how it's done. They talk about, you know, data and whatever, but I genuinely think that you should look at the results and work backwards from there. Um,

I think the uh the the next speaker is incredibly personal for me, which is really cool. Um, so uh Joe and I used to work together many moons ago. Uh, we did SEO together uh for predominantly like um sort of luxury fashion, hospitality Type clients. Um, and now Joe is here today to speak to you. Um, he runs his own incredible agency. They do unbelievable SEO and AEO work. We we work together on a a bunch of different clients. Um and so Joe, if you could introduce yourself, uh your business, what you do, and then

um yeah, I'll hand over to you to take it away as our last talk before the the Q&A. >> Amazing. Yeah, thank you, Chris. I mean, I'm not quite sure how I follow Andrew. Um so that's that's not ideal, but hopefully I can do a good job. There's a little bit about me on the presentation. So, what I'll do is I will get that shared now. Um, I won't be able to see you. So, if there's anything that you need to say, just shout over the top of me. Um, we will go from there. Um,

[snorts] but essentially like I I'll talk a little about myself, but what I'm going to talk to you a bit about is how we're Aligning AI search with SEO as an SEO agency. I'm Joe. Hello. very short and sweet. Um, I have been in the agency space since I was 17. So, I'm now 32. Um, I started after my last A level. Went straight in the on the Monday after after that last exam. um worked at kind of like really large scale agencies working across things like Tommy Hilfiger, Calvin Klein, Urban Outfitters and then >>

[snorts] >> um moved to a luxury agency where I was working with brands like NetApporter, the Ritz and then three years ago I set up uh an agency called Verde and I wanted to specialize really around like fashion, luxury, ecom. Um, like I said, we've been going for three years now and being biased, I think we're working with some of the coolest brands in the space. Adenola, Paul Smith, Pure Sport, Super Dry, the Couture Club. Um, and we Actually announced JD Sports for the Canadian market last week. So, um, that's a very kind of like

brief intro into me. And that's actually how I met Chris as well. I can't see Chris's face, so I'm not sure if he remembers this photo. I think it was the sweatiest bar in the whole of London. But I met Chris at my last agency that we worked together at. Um, and that's kind of I've been SEO side literally since the day that I Started in digital marketing, including when I was working with Chris. But anyway, that's that's kind of then this is now. Let's actually start with SEO. That's kind of like my bread and

butter. That's what I've been doing for a long time. And that has been the focal point of building Verde as an agency. If you spend too long on LinkedIn like I do uh over the past year or so, you would have seen many articles saying That GEO, AEO, AI search, however you want to kind of phrase it, is the death of SEO. But I actually don't believe that at all. And there's actually search data to back up why I don't think that if we want to look at how many people are searching for SEO agencies

globally, that is now higher than it's ever been before. And I think that's because, and we're going to go into this, is that there's so many skill sets that go in from SEO into AEO and GEO that people Know that there's a massive opportunity to take advantage of in search. So instead of a death, I'm actually thinking that it's a form of a rebirth. And I'll explain more why I see it more of a rebirth um than anything else. So why to me SEO has just become more important? And the reason why I believe it's

more important is because people are searching more than ever before. So, there's nearly 15 billion searches per day, which means that there is a massive amount of potential to draw those users into whatever you're trying to show them or sell them. And actually, what we're seeing is that compared to like when I started in the industry 14, 15 years ago, there's a massive opportunity to target searchers across such a broader spectrum. And like all of you guys on the call, I know that I don't need to teach you how to suck Eggs, but when I

first started it was very Google heavy, bit of Yahoo, bit of Bing, but actually now that it's it's AI search, it's co-pilot, it's chat GPT, but actually it's also like Tik Tok, it's Reddit, it's YouTube. Um, there's so many opportunities. People are searching everywhere. And this kind of plays into why I see it as a rebirth is because it's still search optimization. It's just how do we target depending on where they're searching. So where are we seeing the opportunity? And I just want to um say that this is kind of like from my experience as

an agency owner with an SEO agency. So there's some kind of like key pillars that we're seeing um that we're taking advantage of. And I've broken this into three key areas predominantly around e-commerce because that's really where our our target market is and where our clients are. The first one is sentiment and Understanding what AI search and and what LLM sorry are saying about your brand. The second is category pages. I'm going to talk a little bit more about the success we've seen from an SEO standpoint, but actually how that's then going to play into

that kind of like next wave and where we see even more opportunity there. And the third is longer form content and kind of going back to the question that that is being asked is like has AI Killed blog content? I'm going to dig a little bit more deeper into that and kind of give you my thoughts on that. So let's start with sentiment. As an agency, we cannot control the narrative around our clients. There's going to be things written on Trust Pilot review sites. Um it might even be on like publications around the brand. We

cannot control every single facet of that. But what we can start to do is help Guide it. And this is one of the this is kind of like one of the areas that I think is like super interesting um from the AI search perspective and using searchable for. So one of the things that we're really looking into at the moment is turning those sentiment pain points into content ideas. So here's an example for one of our clients. And the final negative um piece is that sizing can be inconsistent. Now there's some specifics around like the

sizing. There's a little bit more about like international sizing and sizing around jeans, but what we can see is that where we can see that there's inconsistent sizing. We're working with them to build out improved size guides. So, this can be based on product type, fit, style. I mentioned jeans. They have such an array of jeans. It's not just a case of like how do their baggy jeans compared to their kind of like more Slimmer fit. Is that actually sometimes the waist size is slightly different and how can we educate that user? Um, I might

be getting ahead of myself in regards to my slides, but what we want to do is like with the sentiment side of things is like if we notice a painoint, how can we actually address that before they even get to that? And I think this is what we can use from this sort of data. Oh, there we go. So, yeah, if we see Negative center point sentiment points, we want to know what is the best way to actually go and address that. Therefore, we can use searchable to then go, right, the way it kind of

works is it gives you like almost like a priority score. So, like in this instance, we're seeing high priority. We want to counter those occasional sizing inconsistencies. Um, it's being seen as a weakness. What can we do to to kind of go about that? And then what we can start to do is Actually build out content around that. So then again, all in platform, you can then say, look, I want to create a new size guide to help alleviate the issues. We're working on something a bit more in depth here because it's more around like

the the jeans, the international sizing. But what's great about this is it kind of ties back to my my point earlier about people are searching across different platforms. And what's great is that from our perspective at the moment, We want to have something that lives on the site, but this can also then be replicated into create some sort of like uh YouTube content, then writing a YouTube description, or is it something that lives on Tik Tok as well? And what's great is that when you go through and actually go and create this content based on

the sentiment points that it's pulling, it's then helping guide the the narrative and the conversation. Say like, look, here's Some specific complaints that we're seeing come up within those prompts. So, as I mentioned, one is around international sizing. Now, this particular client um they are UK based, but they have a very strong uh presence in the US and Japan. So, if we know that customers are are buying the wrong size and then they're complaining about it online, how do we address that ahead of schedule? So, then they have that content they can read and then

go and um Buy the correct size. But actually, this is great because one, it it stops returns, which is another business problem. But two, it might even like actually help someone go and then make the purchase in the first place because they might second guess themselves if they're not quite sure and go, you know what, I'm going to hold off and buy elsewhere. And that kind of leads into to one of the things that I want to talk about is Like this data goes way beyond just helping uh for SEO or AI search. This can

actually be used by the wider business. And I've got like a few examples, but as an agency, one of the things that you that make you stand out is how you can integrate with your clients businesses and actually offer them support above just digital marketing or above your kind of like service level. So if we can see within sentiment there's customer service Issues, how can we share those common themes with the customer experience team? For one of our clients, we we now have a direct link to their customer experience team to relay this information back

to them. They can then overlay that with anything that they're seeing in like Instagram DMs and stuff like that and create content off the back of it. Pricing discrepancies or concerns, especially for some of like I say smaller businesses, they're still Making millions of pounds. But we can then relay that back to um the the founders of of the business, the business owners, and they can then create content. Say you have like a um what's the word? You can have a FAQ section where it's like understanding why things are priced this way. Is it premium

materials that can be shared out of stock products? If I'm looking for something and and it's out of stock, I might then just Kind of scroll and not think anything of it. But if people are actively then saying like this is a frustration of theirs, can we then work with the buying team to go like look, we're seeing this consistently come up that this particular brown cardigan is always out of stock. Is it something that next season we can buy more of? Are there specific content types that are performing well? And we're going to talk

more about content types, but if we Can see that style guides are performing really well from an AI search perspective and SEO, can that be shared with the social team to recreate on Tik Tok? are competitors being cited where you're not appearing. That's something we can then share with the PR team. Um I was actually talking to my team earlier. This is this is late edition, but what we can do is then see that are anything around sentiment are people looking for a particular product type or style that They cannot find. We can then overlay

that with internal site search data. So it might be that actually someone is looking for striped clothing because they like the brand but they don't offer it. We can then take that sentiment if people are saying that as well and then go look here's a product development team here's what we're seeing people want this product is it something you can create and as an agency to me that is how you Add extra value yes we're an SEO agency but what we want to do is ingrain ourselves as many parts of the business as possible and

be as helpful as possible beyond just that monthly report of saying how much SEO is up this So the next step is category pages. In e-commerce SEO, we've seen huge success with category pages. So this is just an example of one of our clients. Um what we did is we compared peak versus peak year on year. Um this Is non-brand clicks only. Um and it was up about three and a half times year. So we know that this approach works really well. It's that non-brand approach across those category pages. But now is the time to

double down on that approach and I'll explain why. To start, you have to be the one targeting on your own site. So, I haven't thought of a catchy of a catchy catchphrase for this yet, but if you don't talk about your business as X, Then how can you expect to be discovered? An example of this could be if you are a sportsware store and you have a tops page and a shorts page and a socks page, that's great. But if you're not actively targeting running socks, pilates, leggings, etc., then that's the first step. How can

you expect to be discovered if you're not even talking about your own products and your categories as that? Here's an actual example for one of our Clients. So, say you sold men's suits, which they do. There's so many types of suits that you could sell, right? It's it's almost like unlimited. But what we wanted to do is really kind of drill down into the types of suits and the searches and the people that they wanted to target. We really wanted to go hard on weddings this year. And we really kind of wanted to focus on

that designer and luxury searcher. We didn't want to kind of target that mass Market appeal or someone looking for kind of like prom. So things like that. We we were quite specific about the audience that we wanted to go for. So we did what we normally did. This was for Paul Smith. We would then go and create the page. We would create the content. We would look at the internal linking structure. Real focus around like designer wedding suits for men. And off the back of that, we saw great results. So what were they? We saw

a 67% Increase year on year in clicks from luxury premium and designer non-brand searches. But I have to admit, this isn't just for that wedding page because that wedding page didn't exist. So, I can't give you an exact like for-like comparison, but it's an approach that we took across the site that we knew that there was this market to go for. Um, oh, I've just skipped way too far ahead. Apologies. Let me go back to where I was. Um, here we go. But that new page um since going live, we're probably looking about I think

seven or eight months ago now, maybe six months ago at the start of the year. Um has generated 100,000 impressions. So every click is is brand new compared to what they didn't have before. And when you then look at all the opportunities here, that really compounds because that didn't it didn't exist last year and you're now becoming More discoverable. That really demonstrates the SEO value. But that has really translated over into AI search as well and the value that we're seeing there. So using searchable, we're tracking multiple prompts around things like designer men's suits for

weddings, etc. And what we can start to see is like over the past week or so when this was pulled, what what's coming up when those prompts run and like how many products Are being seen within there. And this is helping us justify the yes, there's SEO benefit, but there's there's there's an additional benefit outside of that that we really kind of want to hone in on and grow. So, when we then did some kind of like checks as well, what we're seeing is that when we're going on chat GPT looking for designer men suits

for weddings, we're coming up consistently within those kind of like top three, top four positions. Um, when we ran this Recently, we was coming up first. And then if you look at something like Gemini, they kind of had um four sections. Um so like they were they were talking about retailers and specific kind of designer houses. But again like the brand was coming up against some like really well renowned brands like your Tom Fords, like Ralph Lauren. And actually that started from us creating that content and we created some like kind of more blog content

as well and Worked on that. talking about designer men's suits for weddings, really focusing on that the site without even kind of then thinking about the next steps around PR etc. And that's why I really believe it's time to double down on that on that category or collection based approach because with AI search tools you now have that other level of data that wasn't there previously and that can really help inform the Content strategy as well. that can help us identify what is the type of content that we want to create around this. So we

can create a category or collection page, but is it that we need to have a ranked list? Is it that really uh we should look at creating YouTube content around that because we know that that's what's being picked up within LLMs. This is what for me is like kind of gamechanging for um a tool like searchable. Then we go into longer form content. AI search has killed blog content. Um, that's what everyone's saying, right? Or did it? And this is something that I want to explore in the next few slides. The data is showing that

many blogs are seeing a huge reduction in traffic. But perhaps what it's done is it's devalued meaningless content. And I think that that is something to like really kind of like think about is because I've worked in SEO. I know how It works and I think for too long SEOs have been creating content in the pursuit of clicks. It's a case of we need to look good in the next monthly report. We're going to do this blog post that's quite generic because we know that people are talking about it. Uh and that will kind of

get us where we need to be and show a lovely graph going up. [snorts] But clicks don't always equal revenue or value. So if you're not bringing revenue and you're not bringing Value in, what is the point? Uh we're working with a few clients at the moment that have really seen like um a big hit in terms of traffic because of their blog content and because of AI search but actually the blog content they were producing three four years ago it drove clicks but when we actually looked into the data nothing was happening with it

and actually what is the point of this content and continuously creating that if there's no additional benefit um and People actually going to convert. So AI search has eaten up clicks to blogs, but it's now time to focus on what actually matters. And ultimately, it always has been, always will be the user. So let's still use data points to inform content ideas. We can use traditional SEO tools. So I'm looking at widelegg jeans here. Let's have a little look at like what people are searching for. How can we then go and create content off the

back of that? We can then use some Use something like Tik Tok creator search insights to kind of validate that. Uh you can see these little gray graphs on the right are saying like actually there's been an increase in people looking around like how to style widelegg jeans. So let's amalgamate that data and create the content. But now we can actually overlay AI visibility data as well. So what we can do is we can then see what type of content is produced. So we know what People are searching for. We know what what we can

create. But actually this gives us a little bit more guidance in like should it be a ranked list, should it be a YouTube video, should it be an amalgamation of the both. Um, [snorts] and I think that's kind of shown on the right hand side, but on the left hand side as well, it shows where um, like kind of what sources being pulled for these prompts. So in this specific example, we can see There's been a decrease in in Reddit, but actually an increase in GQ magazine. So then we can then feed that information back

to the internal PR team and go look we're going to create something like a rank list here or a topic guide but actually we think there's real value in you kind of reaching out to GQ and seeing if there's something you can do there. So again on searchable it shows you the type of content that you can create and It gives you like a priority score. So what's great about this is for example if we look at the first one it says create a ranks list um now for some of our clients it really has

like their actual competitors in terms of like XY Z um are showing like a lot of um a lot of visibility here and that's great because as soon as you mention competitors to your client it's like okay go time let let's do whatever we need to do but what's cool is that in this first part It shows that in 30% of responses, this type of content is showing up. So, we if we know that this is consistently showing up, we know it's really important to go and create that. Now, um and this is what what

has been great for us is it informs the type of content rather than let's just go and create content. And I kind of already mentioned it with that GQ example, but it mentions those sources and websites that host that Content. So, can we reach out to them? Can we work alongside them? Even what type of content are they producing? Is it something that can kind of be replicated um from our side where possible? So where are we heading in terms of like SEO world, agency world? It's now not just a a nice to have alongside

SEO. It's really helping influence decisions in SEO, which is why it's vital. um it's Giving us another level of data that can help even if we just said right we we only do SEO it's giving us all this information all this data that we can go and take so for me it's just like a massive massive plus it's helping influence SEO decisions but also it's helping influence wider business decisions when you use it correctly and that goes back to the point around sentiment how are things changing for the SEO Industry [snorts] so it's making SEOs

test play and learn. It's making them think about how the customer behaves and it's creating engaging, informative content. But shouldn't we have been doing that the whole time? I'm going to give you a spoiler. Yes, it should have been doing the whole time. And that's why I say it's a rebirth of SEO because I think that it goes back to that point I was saying about chasing Clicks and vanity metrics that aren't actually doing anything. This is giving us new data that we can analyze and actually really target the user and kind of like [snorts]

getting rid of years and years of like rubbish that has been created or done. And I think this is why it's such an exciting time to be in SEO. It's also given PR a new lease of life as well because brands want to be cited or recommended. And it goes back to my point around if you don't talk about Your business as X, then how can you expect to be discovered? What? Five minute warning, dude. >> Yeah, all good. But what PR does is it adds that additional layer of trust and it adds those third

party sources. So you have to start with that content at first and then you can add that level of PR in. So to quickly summarize, it's not the death of SEO, it's a rebirth Because SEO and AI search, they're not they're not completely separate entities. It doesn't need to be you can only do one or the other. They work together. And this is how you need to use the tools to to allow that. Being discovered starts with you and you have to go and create that content. You have to go and create those pages. Test,

play, and learn with data. This space is super exciting. I think part of the reason why it's super exciting is no one Truly truly has it figured out. We're all learning and that's why it's it's great and there are actual next steps and it's time to get started now. So whenever I join something like this, I think, oh, this is great, but like what can I actually do? So I wanted to kind of give like three steps that you can do um to kind of get that ball rolling um in terms of AI search. So

audit for sentiment, Find users pain points and actually build out the content around it goes back to what I said earlier. If we can target those pain points before we get to it now, we can then possibly have increased conversion rate because people already have that content and they go that is the size that I need. I'm going to buy but can then impact the wider business as well in terms of I don't need to return this because I've already bought the right size because the Content's there. Develop deeper categories or services. So I keep

saying this but if you don't talk about yourself as X, how can you be discoverable? And I know that I've spoken about this with uh ecom SEO lens, but actually as a business like I need to be doing that as well. So like we work with lingerie brand brands, swimwear brands, sportsware brands. We need to create content around all these different elements because yes, we work With fashion but that's so broad. We need to target all these kind of like subpockets um if we really want to be discovered there and then create content. But we

don't just create content for the sake of it. we can now see what content is deemed as authoritative by LLMs. So, let's use that data. Let's take all the data points that we can get and actually build something really worthwhile. Let's create content that is actually useful, Is actually authoritative, and that's going to help you find a deeper connection to your audience because you're building something that they want rather than just creating content for the sake of creating content's sake. That's my face. Uh, feel free to add me on LinkedIn any questions. And I just

want to say massive thank you to Chris and the searchable team for having me on. >> Thank you very much. Um, that was Awesome. I think uh an incredibly um practical and realistic way of like driving traffic and um and optimizing your site and thinking about it as a real business problem, which it is. Um, so thank you very much for coming on. Um, and thank you for delivering that. And if you want to connect with Joe, make sure that you give him a follow on LinkedIn. Their agency is amazing, right? They do incredible work.

Um, and so if you're looking to optimize your Business, I would say like they're one of the few agencies that I really, really would re recommend. >> Amazing. Thank you so much, Chris. >> Thanks, dude. Take it easy. Um, okay. Awesome, ladies and gents. Thanks for for being here with us. I'm going to move on to the Q&A and answer as many questions as I possibly can. Um, I think there's a point in Joe's presentation I really want to pick up on. Uh, and I talked about it earlier, Which is you saw that slide where

you had, you know, a hundred different ways of describing the same brand. It was to do with fashion and clothing. Now, if you're only tracking a couple hundred prompts, you can see how you're going to end up with the most like limited view of how your business is performing. And so I just want to make sure that I hammer home this point that most of the businesses like the bigger businesses or The scaling businesses that have switched from the likes of Profound have done so to track more prompts because the more prompts that you track

the more information and data you get. And every speaker has mentioned this today but the sources is going to be the most important part of the optimizing of your business. And so let's say that you track 10 prompts. you can only see the sources for those 10 prompts. If you track 200 prompts, it's the same thing. Now, if you're a big business and you have like a 100 products on e-commerce and you're not tracking enough prompts, you are quite literally shooting yourself in the foot. And so I just want to make it very clear because

I know the caliber of people we have on this call that if you want to see what the prompt universe for your business looks like and you want to get a a greater understanding of how to drive your business forwards, get in touch with the Searchable team and we can do that prompt universe um exercise with you. Um so just as a reminder before I move to the Q&A uh the next um searchable AI search accelerator will be launching very sooner coaches like on the call today like myself who can show you how to optimize

your AO. These are the types of brands that have done it. The sign up list for this current cohort is absolutely insane. So, Make sure that you apply and you have to be a searchable customer obviously. I'll always say um there is live training, live coaching, um on demand modules that we've prepared in advance for you. There's private workspaces for networking community working out problems. We kick off on the 22nd of September and we'll be going deep on how to optimize your AEO. Um, and here are just a few examples of people who have done

it before and what they said um About it. So, like I say, I hope the call was very useful today. If you're um if you're tuning in afterwards, thank you for tuning in afterwards. If you're tuning in today, thank you very much. I'm now going to pull up the Q&A and I'm going to try and answer as many as I possibly can. Um, so let's get into it. Someone says, uh, someone said, "Woo, Salt Lake City." Yeah, I mean, uh, it's a it's a great place. We now have an office here, so if You're in

town, say hi. Um, someone says, "Why is Claude trained on Brave?" Well, I think that the the core thing to put across is that outside of that business, no one is going to no one is going to tell you exactly why they make their product decisions. As a marketeer though, my job is to work out uh the state of play and the rules to which um your to which you're able to optimize. And so whether you're looking at Gemini or Claude, I don't really care why they made the decisions they did. I care about how

to optimize things moving forwards. Um someone says, "How many prompts would I recommend to track?" Well, it's not like an arbitrary number of prompts. That's what I'm trying to put across to people. It will vary enormously depending on the business, right? And so, you know, for a small business, you might be talking about um, you know, a hundred prompts or 200 prompts. For a bigger business, you might be talking thousands. And for some of our enterprise businesses, you know, tens of thousands. And so, it completely varies based on your own business size. Um uh if

your organization is only now starting to optimize for AI, are you late to the game? Honest. Honestly, no. Um we are we are so early in the uh AI, you know, evolution as it were. And I think that that's really important for People to take away is that some companies are obviously much further ahead. Some companies are much further behind. But the companies that are behind, you know, you've got 6 12 months before I think it becomes standard that most people in businesses are using AI to an extent, but that would be like using a

chatbot. Um, and so I don't think you're late at all. I think there's a huge amount of opportunity. Um, when does the how much is the Cohort? The cohort is uh for brands. You have to have a pro plan on Searchable. So, it's $125 a month, $125 a month. If you're an agency to join the agency um which is our highest demand program, um it's $400 a month for an agency scale pro uh account. And once you have one of those, you qualify for the uh accelerator. Um how do we Someone says, "How do we

choose which profiles are best for our businesses?" you know, whether it's Gemini, uh, ChachiBT, etc. Well, I'm going to give you a general rule set, which is if you're a B2B company and you're looking to, uh, generate, um, inquiries for your business, you really are going to have to focus on Claude because most uh, businesses are moving towards using Claude from an enterprise perspective. The second one to think about from a B2B perspective is C-pilot because C-pilot is within the Microsoft suite and therefore all these big Companies use C-pilot and so you want to be

looking at that as an option. However, if you're like an e-commerce or consumer brand, your main focus is going to be ChatgBT Gemini and um the Google stacks AI overviews etc. Um, let me see here. A lot of these questions were answered. Will we be sending out the slides? Yes, everything is recorded and the slides will be sent out to everyone. Um, What are my top tips for optimizing for AI search? Well, look, I mean, the reality is if you track 100 to 200 prompts and you set them up correctly, you use Google search console

and our data to create them, you then look at the sources that are being used to answer those questions, that is your road map quite that's the road map. You see that in your sector uh you have, you know, most answers being answered by YouTube, then you have to make YouTube Content all work with influencers. Um, if it's editorial content, then you need a digital PR strategy that's going to go after those editorial titles, either to uh edit what's already there or to create net new content that references your brand. Um, you can't, someone's saying

you want to track how successful your content is right now on and you're with profound. Um I mean I don't know their tool well enough but I do know that um on searchable you can reverse look up the success of your content that you create and so you are able to see like how successful your Instagram posts are, your YouTube post etc etc the content you're creating. You can reverse look that up um and see how successful your existing marketing is which is a really critical feature. So just book a demo with the team and

we'll show you that. Um how can search will help sol? Well this is a uh this is an interesting point in general is that everything that we've talked about today whether it's uh Luke from a company worth billions or whether it's um uh you know LinkedIn worth hundreds of billions etc. the principles all remain the same. It's just at what scale you can do them. And so I saw someone post a Darth comment in the chat like, "Oh, automate isn't important for AEO. Where's the AEO presentation?" But the point is it's got harder. There's more

ground that you need to cover. And so learning how to automate, you know, for example, you can use the searchable agent with perfect context, uh, trained on how you write to, um, create a whole batch of upcoming 30 articles for the next month. And so a soloreneur can do that as well as a marketing manager in a business. So it's learning the tools. Um Uh so someone says how how can I track the success on adverts? We actually just released a feature on searchable the other day which shows you where your competitors are doing chat

GBT ads and it shows you their creative and on what prompts they're they're spending money on which is really useful to understand how you can strengthen your own content. Um, what are the things I would do next? I Mean, take tracking for granted. Okay, you have to track to understand this, but create content and refresh your content. Those are the two things really. I mean, of course, I think I could give you a playbook that is like a hundred um, you know, actions long from anything from creating LinkedIn posts, tracking more prompts, creating YouTube videos,

writing more articles, freshening up your content, use the searchable agent To rewrite the meta descriptions and meta tags of your posts. Like the list of SEO is so significant that we had whole teams of people like Joe was the head of SEO in my agency when I was like 29. like we've both been doing this forever and the list of things to do is so significant and so it's not just one thing. So stop looking for the magic bullet and appreciate that if you win in this game it is so valuable to your business. Um

so I'm going to try and Take uh one more question. Someone says have I changed my opinion on LinkedIn articles from this time last year? I wasn't the biggest fan last year. Well, here's a lesson about marketing that you really need to learn. What is the right thing today is not necessarily the right thing as of a year ago because these platforms have algorithmic determination as to how content is spread. And so a year ago, you could get advice on how AEO worked And it wouldn't be that relevant today. And so if I was saying

like I'm not a huge fan of um LinkedIn articles last year, I would say that at the time there wasn't a strong argument as to why you should make them. And had you focused on creating predominantly um posts, then you would grow your following, have a bigger following, and actually I have even better data than LinkedIn on LinkedIn posts recently, which is if a post gets over 500 likes, its propensity To be cited is at like 80 90%. And so if you build a strong personal brand through posts and then you start creating articles, that's

clearly a strategy that's going to work even better for you. Um, if there's any questions in the Q&A that answer that weren't answered live, um, we'll follow up in the uh in a like frequently asked questions in the in the follow-up email. Um, but ladies and gents, I just want to say Thank you for joining. Um, for us here in Salt Lake City, it was an early start. I don't know where you are in the world. Thank you for joining. Thank you for supporting Searchable. I can't believe how well Searchable has done. Um, eight months

since launch. It's crazy. I appreciate you all. Much love to everyone. And if you're interested in driving your AI search, get the team in the cohort. Make sure you apply. Much love. Take easy. Adios, amigos.