Perplexity and Notebook LM recently rolled out some killer new updates, and combining these tools can help you build some powerful strategies. I ran a marketing agency for the last 10 years, building strategies for solopreneurs all the way up to the world's largest software companies, and I've recently distilled my workflow down into three steps using Perplexity and Notebook LM. My team and I used to charge thousands of dollars for this; I'm going to show you exactly how to do it today for free.
First off, why do we need separate tools for this? Can't we do it all just inside of ChatGPT? The short answer is yes, you can do all this inside of ChatGPT, but certain tools excel at certain things.
When you chain them together appropriately, your results can be dramatically better, and that's what I'm excited to show you today. Here is our roadmap: First, we're going to tackle research, making sure that you have some solid methodologies and frameworks for how you do your research. Then we're going to get into analysis, ensuring you dive deep into the research that you've collected.
Finally, we're going to wrap it all into a cohesive plan. If you haven't noticed, the world is changing really fast, and having a solid strategy for how you adapt to those changes is possibly the number one AI use case. As Warren Buffett said, "There's no use running if you're on the wrong road," and there are plenty of people on the wrong road today.
Let's make sure that you're not one of them. Look, everybody loves research. It can be one of the most exciting parts of any project.
It's like that thrill of walking into Home Depot and seeing all of the new tools and fresh materials, before you get home and realize you forgot ten things and the project's going to be a huge pain. So what do you do before you go to Home Depot? You plan.
You have a shopping list, and that's exactly what we need to do when we are doing our research. We need to make sure that we're covering specific categories of our research for a marketing plan. These categories, or these silos, include your audience for sure, your own company and your own offers, your competitors, your industry at large, and the trends and different forces at work there.
You want to go out and do your research on all of those different silos separately. Trying to do research on all of these different silos in one session, or let alone one prompt, is a recipe for disaster. This is where most people go wrong with their research—they just start doing a little bit on their competitors, a little bit about their trends.
They don't have these silos clearly defined, and they're certainly not focusing on them one at a time as they're doing their research. This is where Perplexity really shines; it outperforms any other AI search tool out there. For today's example, we're going to be building a marketing strategy for a new food truck business.
This has been a topic of discussion in my family, so I thought I'd use this video to dig a little deeper and see what we can find. Hopping into the cheat sheet—I make one of these for every single video I create; they are all instantly available to anybody who supports my channel on Patreon. There's a link in the description if that is of interest to you—but we're going to jump right into page four here, where I have some prompts for doing research for your specific audience, your specific customers.
What we want to focus on here is finding real, legitimate research that has been done around these specific silos. We're not really looking for opinions about what one blogger thought about another blogger or one AI thought about another AI these days; we're looking for Pew Research Center reports or studies, etc. So I've got a ton of prompts in the cheat sheet related to these types of queries.
I'm going to just grab one of these now. This is a simple one: Just identifying some demographic and psychographic data for food truck customers in the mid-Coast main area. We will let that run; it has pulled in some great stuff all about what we've asked, and we can either take this response or some of these different sources that it has cited and begin to store them in a separate file.
We want to make sure our silos are stored separately; we don't want to jumble them all together. But I want to go even deeper here—let's grab another prompt. Here's another good one: Find publications from food truck industry associations about the evolving food truck customer.
Awesome! This is already pulling some cool stuff where they're looking for Instagrammable meals. I hadn't even thought about that, but making sure that your food is photogenic so that they can easily share it online—they're thinking of these as experiences, not just meals.
Next, I want to do a little research on the major competitors in the space. It has listed these, which is awesome, and it has also listed—I nudged it for the strengths and weaknesses of each of those competitors—and got some really good information here. I went back and asked it to search Reddit specifically for complaints about food trucks in this area, and this really started to yield some amazing things: high costs, long wait lines, small portions, inconsistencies, etc.
Now I'm going to go out and look at some industry reports. Have there been any studies done on the industry in general? As a part of every single strategy, you're going to want to do some kind of self-reflection.
Whether that's looking at yourself or your company to try to figure out how you fit into the big picture. Now, unless you're a massive company, there's not going to be a whole lot in Perplexity or online about these things. So, one quick and easy way to do this is to open ChatGPT and, in the voice mode, just ask it to interview you.
Or maybe you're sitting around the table with your team; have it interview you on all of these different aspects so that you can get that information documented quickly, easily, and accurately. There's a bunch in the cheat sheet all about this. I could do a whole video just on this stuff itself; let me know if that's something you're interested in, drop me a comment, and I'll be happy to do that.
So now you have a straightforward method for doing your research, clearly defining your silos, and researching each one in isolation. This ensures that nothing gets missed and puts you far ahead of people that are just doing some random Googling. This is really setting up the foundation for a solid strategy.
Once you've gathered all your materials and stored them somewhere—in Notion or in a Google Doc—you may even want to download PDFs of different reports that you found. You may think you're ready to jump right in and build a strategic plan, but there is one crucial step that most people miss. We're going to cover that now.
Most people, even the pros, end up smooshing their research, analysis, and plan creation all into one messy heap. We want to separate your research from your analysis from the plan creation. So the next step here is to really focus on the analysis of what we've researched.
Not only that, we want to do our analysis on each silo individually before we combine everything into our grand strategic plan. When you lump all of your analysis together, you miss some crucial details and some crucial opportunities. It's in this analysis section where Notebook LM truly shines above and beyond any other AI tool.
It can quickly and accurately draw stunning connections between the different sources that you have found during your research. Since it's very specifically focused on only the sources that you have found in your research, it is the perfect tool for doing this analysis step. So here I am inside Notebook LM, and I've created a separate notebook for each silo—one for my food truck business and another for the customers, competitors, and the industry.
Inside of each of these, I've added some sources that I found inside of Perplexity. Now, back in the cheat sheet, I'm grabbing this prompt for analysis, which basically looks at the psychographic characteristics, demographic characteristics, and specifically asks it to highlight any unexpected or noteworthy findings. This analysis step may be the most important part of the entire process.
You want to dig deep here and really try to understand what all this information that you gathered means. There are a lot of ways to do it here in the cheat sheet. You can look for identifying different patterns, comparing and contrasting the needs of different segments, and so forth.
You want to do that exact same thing when it comes to your competitors and yourself—your own strengths and weaknesses. In my food truck competitors' notebook, I'm just going to ask it to identify any gaps in the market based on these different offerings and the information I pulled around the competitors. Then we can analyze all that industry information with these Notebook LM prompts, asking it to please identify the top trends in this industry most likely to impact our growth this year and a brief analysis of each.
Here's a great one: evaluating the potential impact of emerging technologies in this industry. See what Notebook LM can pull about that. By doing this, you'll uncover some incredible insights, such as what's really going on inside of your customer's head and spotting weaknesses that your competitors are not covering.
It'll also help ensure that you don't get hung up on one idea that catches your eye during the research phase, resulting in you spending your whole time building a strategy around that when there were ten other things you should have considered. Now that we've methodically done our research and analysis, combining all of this stuff into a solid strategy is so much easier. You can see I've gathered the best of what Notebook LM was able to pull together around our customers, our competitors, our industry, and our company.
Now I'm going to go to another large language model to begin pulling this strategy together, and the large language model of choice for this one would be ChatGPT Pro. I do pay for this Pro plan at $200 a month. I know that is pricey, but I can tell you it's been worth it in the last month or two.
If you have access to ChatGPT, you already have access to the Pro version, so that might be the one that you want to choose. Claude could also be a great fit for this strategy creation. The reason for this is that these models are really the most powerful at combing through a lot of information, which is what we're going to throw at it here in a second.
Here’s the prompt I’m going to use for that: simply giving it the role of a business strategist and letting it know that its goal is to create a winning positioning and marketing strategy, reminding it that it has access to these documents. So, let's copy this into ChatGPT, and then we're going to grab all of those resources that we pulled together— all of those different silos. Analysis: We're going to copy those right in here as well.
We're going to let it rip, and it thought for about three minutes and came up with this positioning and marketing strategy that looks pretty awesome. Frankly, it gets into all sorts of brand personality and messaging, target audience and their motivations, strategic pillars with storytelling, and fair pricing. This is all pulled from everything that we saw our customers were struggling with, so this feels like an awesome plan.
I think it's a good idea to just give it a very general prompt at first, saying, “Hey, come up with a marketing strategy” or “Come up with XYZ,” whatever strategy you're working on. It's interesting to see what it pulls back in just without you giving it too much info because that way you're getting the best of its brain, and you're not giving it any sort of bias based on your prompt. From here, you can dive a lot deeper into asking it for a step-by-step process or generating twenty other angles or ideas.
You can always go down those rabbit holes later on, but it's great to get just an unbiased opinion from the large language model about what you should do. I think this is just awesome. I really love it when a plan like this works out.
Now you have a full end-to-end workflow for doing your research in these various silos in Perplexity, pulling those sources into different Notebook LM notebooks so you can do your analysis there, and then pulling it all into one of these powerful models for creating the final plan. Being in marketing for all these years, when I think of strategy, I almost always go to thinking about a marketing plan. However, in the course of making this video, I realized you could create strategies for so many other things— from hiring to personal development, you name it.
In fact, I came up with a hundred different things you can build strategies for. I've got that all in the cheat sheet, and that's available to my Patreon supporters. There's also a ton in there about creating silos and doing those self-reflection exercises, as well as tons of prompts for analysis inside of Notebook LM or research inside of Perplexity.
That's all there if you want to support this channel and grab that, along with over a hundred other resources available in my Patreon. There are also some coaching options in there as well. But if you got something out of this, I've got another video that goes even deeper into this strategy specifically for product-market fit and go-to-market strategy.
Here's that video. I will see you over there!
