8 Insane Gemini 3 Use Cases! (do anything)

In this video, I'm going to show you eight insane use cases for Google's new Gemini 3 model that will completely change how you work. This is Google's most intelligent AI model ever, and it can literally code entire applications from a single [music] prompt. So, we're breaking down use cases that range from building full stack web apps just by vibe coding, creating professional, highquality content on demand, all the way to analysis that would usually take hours now done in seconds.

And unlike what we've seen before, these use cases aren't just cool nice to haves. This model is quite literally changing how work gets done. It can independently execute 10 to 15 coherent reasoning steps, processes video, images, and code all at the same time, and integrates directly into tools you already use.

So, without further ado, let's jump into it. Starting with use case number one. And this is something that we get asked about all the time from our students and companies that we're training.

And it is insane how many people are still spending a large amount of their time on this task, and that's creating presentations. So, let's open up Gemini and I'll show you what it can do. So, here in Gemini, first thing we want to check is that we're using this Thinking with 3 Pro model.

This is that top Gemini 3 model. And I'm going to drop in the following prompt. I'll make this one a little bit bigger so it's easier to see.

Create a Google Slides presentation, 10 slides that tells the story of our quarter. What worked, what didn't, and where should we double down in Q1 2026? I've attached our Q4 marketing performance data.

So, let me do that real quick. I'm going to click add files, upload files. And here we have two CSV.

One which contains our marketing data broken down by channel and another which covers our topline month-on-month numbers. So I'll upload both of these. Now whenever you're getting Gemini to create presentations, you want to include two things.

First of all, the content. What should actually be included in the presentation? And then secondly, the style.

How should the presentation look like? So you can see here I've also included a visual star guide covering the fonts, the primary, secondary, and the text colors, the background color, what shapes, images, and charts we should use. And then finally, a very important statement.

Make the presentation visually compelling and executive ready. Use the data provided. Do not make up any numbers because we all know how much these large language models love to hallucinate.

And by the way, if you're interested, you can grab this entire prompt and any of the prompts that I'll be using for all of the use cases in the description below. But let's send this one off. And actually before I do that, one very important thing to make sure that it's actually creating the Google slides presentation.

I want to click here on tools and select canvas. This is necessary otherwise it will not work. And let me send off this prompt.

And we can see what's going to happen now is Gemini is going to start coming up with a plan. So here it's already crafting the presentation outline, defining the data categories. It's going to go and do that data crunching under the hood.

And now it's actually getting underway and creating those slides. is letting us know this will be ready in a few minutes. And now we can see our slides are ready.

Quarterly review and Q1 2026 strategy slides one of 12. We've got the little Q4 tag. And all of this is matching the exact colors that I mentioned in the prompt.

If we skip through here, first of all, we get our executive summary, total revenue, net MR added, the overall return on ad spend. This is absolutely insane. We've got our monthly growth trajectory with a nice uh chart that's perfectly fitting into this presentation.

We've got our channel performance overview. So here it's even put in an exact table with the channel by channel data from that CSV. It's decided to split up the return on ad spend between owned media as well as paid media including all of the charts and some little commentary uh underneath the key wins areas for optimization and then finally the strategy of doubling down on what works.

And here we have our Q1 2026 budget relocation which a with a nice little graphic saying in this case based on the data that that we should move towards search and webinars which is pretty insane. But the most impressive part and this is what makes this actually useful for work is that up here in one click I can click export to slides. And now if you're like me you'll know whenever we are using AI to generate anything most likely it will not get it perfect the first time.

And obviously here we can now go back and forth and chat with Gemini and improve the presentation. But sometimes we just want to take over and make some changes ourself. And there we go.

Our new slides are created in one click. I can just simply open this one. And now we have that full presentation in Google Slides that we can share with our team, collaborate, and everything is completely editable.

All right, onto our next use case, and that's creating your very own personal travel companion. So here in Gemini, I've got the prompt. I've put myself in the shoes of a photographer.

I'm a photographer visiting Tokyo for 48 hours and I want to capture the most impressive and iconic scenes in the city. Create an interactive Google maps with the top 10 to 12 locations pinned organized into day one and day two. And here I've given some specific instructions on what information it should include.

So let's send this one off to Gemini. And now in real time, Gemini is pulling in data from Google Maps. And there we go.

Gemini has now created our itinerary for Tokyo. You can see day one split up into east and central Tokyo and day two in West Tokyo. So, here it's listed out all of the locations and the best time to actually go and visit them.

We've got day one. Here's day two. Let's see.

Oh, and here we go. And here it is. It created our very own interactive Google maps.

So, we can go and check out all of the locations in one click. We can just go up and open this uh location here and check it out further. This is super impressive.

There's a full interactive map that we can actually zoom in and check out all of the locations it has suggested. But let's take this one step further. So now I'm asking it create two sharable links with the route of each day that I can directly import into my Google Maps.

Let's send this one off. All right. And again here it uses Google Maps integration.

Now we have these two links. Let's open the day one map and see how it went. And there we go.

Every single location it suggested it is now automatically listed here um as a stop of our route that we want to take. And this is a link we can just open up on day one of our travels. Now, this is going to come in so handy for anyone traveling to a new city.

You basically have the benefit of just chatting with AI, telling it what you're interested in, and then it has all of that data from Google Maps. So, this can be used for hotels, for finding restaurants. You can just give it a list of all of your food preferences and get it to create an interactive map of the best restaurants in town.

Now, this next use case is rapid prototyping. And if you've ever had an idea for an app or a tool that your business needs, normally you'd have to talk to a product manager, they'd write up some specs and pass it on to an engineer. And this takes weeks of back and forth.

With Gemini, you can now spin up a working prototype in minutes. So to get this working in Gemini, we're going to drop in the following prompt. And this is sort of a template that we created.

Again, this is available in the description below. So you can just copy this template and adapt it to your use case. Here we're giving some requirements of the prototype.

what should be the output format. So here a single page app with all the features accessible. Include mock data that demonstrates each feature working and make all interactive elements clickable and functional.

And then here you have your briefing and here's where you can add the specific details of your app. So I've just gone ahead and filled out this briefing for our prototype use case. So what are we building?

An admin portal for an online learning platform called 9x learning. Who is the user? It's L & D managers and company CEOs.

And then I've listed out the core features. So we want dashboard and reporting. So this is so that the companies that we're training can see how their employees are progressing through our training program.

We want some student management, some engagement and communication. So being able to identify at risk employees that are maybe falling be behind or less than expected progress. And we have some settings.

And then we have this very important last sentence, build this prototype. So I can click through and test each feature. And again to make this work, what we need to do is click on tools here and enable the canvas mode.

So, I'm just going to select this and send this one off to Gemini. And we can see now Gemini is planning. It's considering the scope of this prototype, designing the prototype structure, outlining the core features.

And now Gemini is getting to work. And before our eyes, it is actually writing the code for this prototype. It's absolutely nuts.

And here we go. Our prototype is ready. You can see now we've switched to this preview mode.

You can always go back uh to check out the code that Gemini wrote. This whole thing literally took a couple of minutes to create. And now we have a fully interactive dashboard.

So here we have our 9x learning platform. We get a quick overview of uh the stats of our students, 20% completion rate, 64% uh active progress. We have the enrollment by department.

Obviously using mock data. So here it just created five students per department. And we have our top performers.

Let's check out. to all of these tabs. You can see here we can completely click through this app.

We have our student section where you can also filter by the different departments. So if I filter to engineering, this is absolutely nuts here. What is amazing is you can now just in plain language describe an app idea, a prototype, and have back and forth with Gemini until you're happy with it.

And then you could actually pass it over to someone to actually bring this to life. And they even have all the code already written. This is absolutely nuts.

Let's take a look at what else it created. So we have here an engagement section where we see here the at risk students. We have some different messaging templates.

So friendly nudge or a deadline warning. We can nudge all. I guess here it's giving us a confirmation that this nudge has been sent.

Let's have a look at the settings uh section. This is uh absolutely insane. Now if you want to go iterate on this prototype that you've built either you can just continue chatting here with Gemini and ask it to make improvements or even more impressive if let's say here in the uh engagement section I don't like something what I can do is I can click select and ask and let's say I want to change something in this button I can just hover it and then give Gemini some instructions here so it knows exactly what elements in my prototype I want to change and I don't need to explain it here in plain language.

So, onto our next use case, and that's data visualization. Now, for probably the past year, AI models have gotten pretty good at crunching data by writing Python scripts. But what's been seriously lacking is actually visualizing that data in a way that's useful and interactive.

And Gemini 3 takes this to a whole another level. So, over in Gemini, I just gave the simple prompt to create an interactive dashboard where I can play with the marketing performance data from the last quarter to prepare me for our quarterly review with the CEO. and I attached two CSVs of the channel performance and the topline numbers.

Gemini then got to work crunching those numbers and created our quarterly marketing review dashboard. Let's open that one up. Very important, you do need to have canvas mode on as well for this to work.

So, let's open up our quarterly review dashboard. And here we have our full interactive marketing quarterly review dashboard. We've got the total revenue numbers, the net MRR growth, blended uh return on ad spend, blended CIC, some nice graphs that sort of combine different metrics from those uh spreadsheets, the MR, the churn, the net growth, funnel velocity, so how much uh deals are we winning month-on-month.

We can even have a few different tabs here. So, let's see what this has created. And like I didn't prompt any of this.

I just said build an interactive dashboard. Gemini then went looked at the data, thought what would made sense and created a very useful dashboard for us to actually understand the numbers. So here we have in the channel view we've got this efficiency matrix.

Super super nice. It's even gone ahead and separated the owned from the paid media. This is again not something that I promed that is in the dashboard.

It just understood hey email campaigns, referral programs and webinars that's owned media and then we have paid media over here and decided to color code them accordingly. That is absolutely insane. And let's have a look now at the planner.

This is a really, really nice touch. Of course, maybe Gemini doesn't nail everything or not everything is going to be useful, but I'm sure it will give you some ideas that you might not have had if you were just looking at those CSVs. All right, for our next use case, we're actually going to head over to another place where you can access this new Gemini 3 model, and that is in Google AI Studio.

So, you want to head to a studio. google. com, google.

com and then we're going to head to the build section. So, this next use case is interactive lead magnets. If you're running a business, you want to capture leads on your website, one of the most tried and tested ways to do this is by creating interactive tools like pricing calculators or quote generators.

And I just want to show you how quickly we can spin one of these up inside AI Studio. So, here in AI Studio, I am in the build section. I've got the Gemini 3 Pro model selected.

Let's drop in this prompt. And so here it's create a standalone embeddible investment growth calculator widget. So this is something you would put on your website if you are offering some sort of financial services where your users can just calculate their investment returns.

You see a lot of these uh out on the web here. We want to display the results with large bold final value numbers, a simple line chart. So we're giving uh Gemini very clear instructions about how we want our widget to look.

And most importantly below the results add a lead capture. want a personalized plan with name and email field. So, let's send this prompt off to uh Gemini inside AI Studio.

And now Gemini is going to get to work and actually code out this lead magnet capturing tool. This little widget that you can then eventually put over onto your website. And just like that, our pricing calculator or here investment calculator is ready.

We can see this is a fully like let's say interactive widget that we can use. So, let's say we want to start with $1. we haven't got too many savings yet and we want to do a,000 bucks a month and let's say we're optimistic.

We want aggressive investment strategy. And you can see here even it's got a super nice graph that as we play around with the different years or if we change um the amounts we can see this graph uh is changing. This is something that you could literally just drop uh onto your website and here want a personalized plan uh get a detailed breakdown sent to your inbox and this is like looks very very high quality.

You can see here little copyright wealth widget uh incorporated. The way this works now inside AI studio obviously this is a fully uh coded application a fully coded widget and it gives you a few different options. Now, this is something that you would need to have some help from a developer for or maybe follow along with another more in-depth uh tutorial.

You can basically save this to GitHub and eventually then push that over to your website or even deploy directly into your Google Cloud project. Now, I've tested creating a few different lead magnets here in AI Studio. This one I tried creating a quote calculator for a home cleaning business.

The calculator should ask for the type of home, the number of bedrooms and bathrooms, the type of cleaning, and how often. And like Gemini literally spun this up in less than 2 or 3 minutes and now you have a full almost Airbnb style um checkout system that you could put on your website if you're a local home cleaning business. So let's say okay we want an apartment.

I want three bedrooms, two bathrooms. We can see the price being adjusted here. Then we have this one's going to be a deep clean and we want this bi-weekly.

So we can see we get some bi-weekly discounts. Even when I click this book a cleaning button, it goes to the next step where we're capturing the information. This is absolutely nuts.

Another one here, create a standalone embeddible project quote calculator. Um, this is going to be for some sort of like website uh freelancer. So whether you want a landing page, full website, uh, e-commerce, we can see the cost uh adjusted here.

How many pages do you need? Do you want any uh add-ons? Like this is uh absolutely insane.

What's going to happen is maybe it doesn't quite nail the instructions the first time. You can simply hear chat at the bottom. See, make changes, add new features, ask for anything.

You just need to have a good way of telling Gemini what you want in your application and it will build it for you. All right, onto our next use case and that's creating your very own personal sales coach. Again, here in AI Studio, I've created my own custom app from a very, very simple prompt.

Build a sales coaching intelligence platform. The user uploads an audio file of a sales call. The output needs to be a dashboard showing a uh diorized transcript, so speaker A versus speaker B, a sentiment graph showing engagement levels over the duration of the call, and an AI generated coaching card that lists three things the salesperson did well and three missed opportunities.

And why this use case is so impressive is the fact that it highlights one of Gemini's key capabilities, and that is the fact that it is multimodal. And here it has no issues with understanding audio. So, in our AI app that we literally just created in seconds, I'm going to upload an audio file from a real sales call that I had.

And now Gemini is getting to work analyzing the call, literally listening to the audio, transcribing it word for word, and then analyzing the sentiment as the call happened. And here we go. Gemini has just finished analyzing that sales call.

And we can see here this engagement and sentiment flow with basically different scores as the conversation happened. So, this conversation looked like it went slightly well. I obviously need to blur out this part of the uh transcript because it might give some some things away.

But here we have the AI coaching insights. What went well? Effective qualification and expectation setting.

The salesperson skillfully qualified the prospect by asking about their background, motivation, and time commitment, deep product knowledge and value articulation, and proactive objection handling and clarification. However, there were some missed opportunities, lack of specific painoint reiterations, limited exploration of business impact, and no clear call to action for immediate next steps. Now, obviously, this whole app was created from a very, very simple prompt.

If you are a sales team leader, you've probably laid out very specific things that your sales team should cover in their calls. You add them here to your prompt and your sales coach, basically your AI sales assistant, is making sure that all of your team's calls are following the exact structure that you've laid out. This app, I can even put it here in full screen.

This is something that can then be actually shared with the team. So here you have share app. You basically have the default Google settings.

So only people in your organization can view it. So you basically create your personal sales coach app according to how you think your company's sales should be done. You can then share this app with the rest of your team and maybe when you're having then one-on-one feedback with the different team members, you can basically use this AI sales coach to go over some of their calls.

So, onto our next use case and that's turning YouTube videos into blog articles in a matter of minutes. So, here we are in Gemini and I've got quite an in-depth prompt here. Again, this one will be available in the description below.

So, you're a blog writer specializing in AI automation and workflow content. So you can just adapt this to whatever niche you are working in. You transform YouTube video transcripts into comprehensive standalone blog articles.

And what we're also going to upload here is a blog style guide to make sure that the generated content doesn't sound like the typical AI slop that we see all over the internet. So I'm just going to upload my style guide. And this is a markdown file that just basically gives Gemini all the information it needs in terms of the language it should use and how it should structure the blog article.

This one will also be available in uh that same link in the description below. Then I can simply send this one off. I'm going to also make sure that I have the canvas mode enabled.

And now Gemini has basically confirmed that it's understood everything in the blog reference guide. And it has followed up asking which YouTube video it should analyze. exactly like the instructions that I laid out.

So to test this one out, I'm actually going to use one of our more popular videos, which was me covering some of the best AI tools for businesses. So now in Gemini, I'm simply going to say create a document for and then paste the link to that YouTube video. This create a document.

This is very, very important because what we want is we want Gemini to just not respond with a blog article, but actually create it in a document that we can use. Plus, we've enabled canvas mode, so that is everything it needs. Let me send this one off.

And what makes this all possible? As we can see now, Gemini has decided it needs to use its YouTube integration. And this is something that is unique to Gemini thanks to the fact that it is owned by Google, which also obviously runs YouTube.

So it has an in-depth connection. It's basically able to see everything in the video, the thumbnail, the full transcript, the description, everything that is in the video, and it can basically use that now to create the blog article. Previously, what we'd need to do is take that video, put it into some sort of transcription tool, get the transcript, then give it to AI.

Now, here Gemini takes care of everything. And we can see just like that in a matter of minutes, Gemini has created this document with the full blog article. Uh here it has absolutely nailed every single tool that I suggested in the video.

And one like very important part in terms of how we prompted this, we basically told it, hey, the YouTube titles most likely will be very much focused for YouTube. So rather reward this. So rather than saying, I tried 325 AI tools, these are the best.

That's a title that works well for YouTube, not so well for blog articles. So here it simply said the best AI tools for 2025 from productivity to business automation. A much more suitable title.

This is following the exact structure of how we want our blog articles. But what makes this even more impressive is the fact that this has actually created a document here means I can actually go in directly and edit this or ask Gemini to edit certain points of this document. And literally in one click, it's a bit hidden, but here in export, I can export to Docs.

And now we have that full blog article perfectly formatted in a Google Doc that we can share with the rest of our team. Now, if you thought that last use case was impressive, this next one just highlights how well this new Gemini 3 model is integrated with the rest of the Google Suite and specifically YouTube. And that's using Gemini 3 as your very own YouTube AI analyst.

[music] So, over in Gemini, I'm just going to drop in this very detailed prompt. Again, as always, this prompt, you will find a link to it in the description below. So, here we've basically asking it to analyze a specific YouTube channel, which we can edit.

So this it's not that it only works for your channel. It works for every single YouTube channel. And this is also then perfect for potentially getting some ideas from competitors.

What we've laid out its analysis framework, all the different things we want it to do. And then finally, its output. We want to structure this as a full channel audit report.

And this is something people would be paying thousands of dollars for YouTube Strategist to complete for them. Let's see how Gemini goes. I'm going to test this one out with our channel.

So, all I need to do is literally drop in our handle, which is at go9x. And I'm going to also make sure that canvas mode is enabled so I can create a nice visual report at the end of this. And we can see straight away Gemini's decided, hey, I need to use my YouTube integration.

I'm initiating video discovery, reviewing channel dynamics. This is going through. If you ever want, you can actually see under the hood.

And this is using specific YouTube tools that is only available in Gemini. And here we go. Moment of truth.

Our full YouTube channel audit report is ready. It starts off with a scorecard. So, we only get a B+ for our content strategy.

We've got some strong use case format. So, I guess if you're watching this one, you probably are also a fan of the use case format, but I'm struggling to convert that interest into more technical deep dive tutorials. The consistency, it's only an A minus.

We've got a regular upload schedule, though. Recent view volatility is high. Engagement is quite high, which is nice.

and a high growth potential. So, I guess we're on to something. If you also think we're on to something, make sure to subscribe and like this video.

Let's see. It's given us a performance snapshot. And this is all absolutely correct data at the time of this recording.

That is the exact subscriber number. The total videos because Gemini has that direct connection to YouTube. We've got some view volatility extremely high.

The channel experiences massive variance which re with hits reaching 370k plus views and some videos only getting 3 to 7k views. This is absolutely accurate. The channel successfully pivoted from no code for marketing to AI agents and automation.

The pivot caused a massive spike in growth specifically driven by two viral hits in late 2025. So even down to the exact time when this happened, this is something that you would be happy paying a consultant to create for you. Now we have a full content breakdown.

What are the formats that are working? What are some of the formats that are not working? And it's actually referencing real videos.

So you can you can see the exact videos that we created. You can check them out on our channel with the exact views and why it worked. This is absolutely nuts.

And then finally, it gives some recommendations. So we've got quick wins. What is the immediate impact we can make?

What are some long-term strategic shifts for our business? And what are some content ideas to test? Now, Gemini has absolutely followed the format that we laid out here in our prompt of how it should run this analysis.

If there's another area of your YouTube strategy that you're interested in, maybe you want it to analyze your thumbnails, maybe you want it to analyze your scripts, Gemini can totally do this because it is multimodal and has direct access to YouTube. Now, one aspect of Gemini 3 that I didn't cover because it totally deserved its own video is the new Nano Banana 2 model that just came out. And if you click here to go to the next video, you can see just how powerful it is and learn how to use it for yourself.