Why Aren't You Using Gemini to Its Full Potential? 21 Unexpected Ways That Will Change Your Life

Did you know that Google's Gemini can analyze videos?

That is, you literally upload a video there, no matter what you are doing, playing sports or cooking something, learning something, it analyzes everything you do, tells you if you are doing it right or wrong, and corrects you. Or did you know that Gemini sees space through the camera in real time? And no, it doesn't just see it, it can literally point your finger at where to click, what to pick up, and how to set it up.
Gemini can also find files in your Google Drive and throughout your Google ecosystem, even if you've forgotten the file's name or where you put it. It can schedule tasks and meetings in your calendar, turn two-hour lectures into podcasts, edit photos, analyze reels... In short, the new Gemini from Google can do a lot, a whole lot. They've made such great strides that it even has more practical use cases than ChatGPT.
Therefore, this article will focus on practical use cases for how you can use Gemini in your daily and professional life. It is also worth reading this article simply because you most likely have Google, which means you have documents, drives, email, calendars, spreadsheets, and I think it's worth learning how to work with an assistant that is now connected to all of this. In short, 21 ways to use Gemini that you would hardly have tried. These are really cool cases, guys, I'm really inspired.
It's important to note that most of what I'm going to show you will be available in the free version, so that you can test it right away. Still, some applications will be available in the paid subscription; just be aware of that and take it into account.

Methods 1-7: Live mode, video analysis, and converting content into instructions

The first application is Live mode, which is available in many neural networks. Still, Gemini has one cool additional feature, and it's probably easier to show how it works right away than to explain it. In voice mode, you can turn on the camera, and Gemini will see everything you see in real time, but not just see it like ChatGPT, for example. It can highlight objects right on the screen if it wants to point something out. In other words, AI can literally point its finger and be with you, and this little feature seems to have a sea of applications, which we'll talk about.
For example, to figure out how to use a piece of equipment or a device. You've moved into a new apartment and need to take meter readings, but everything looks very different, with all kinds of buttons, and you don't know which ones to press. Turn on Live mode, say, I need to take the water or electricity readings and it will show you exactly where to press this button. Then it highlights the exact reading you need on the screen with this frame, and you don't even have to remember the numbers; you can get them from the chat later, you know?
Now the same thing, set up, for example, a coffee machine, or some new appliance, modes on the stove, if you don't know how to switch them, or where to press, Gemini shows you exactly where to press, or if you need to turn a tap. You don't know which wrench to use, so show your toolkit, and it will highlight the right one right on the screen. You can even ask for a missing Lego piece, it can help decorate the Christmas tree, guys, tell you exactly where it's best to hang a toy so that it looks more harmonious, it will do it.
In general, I think that now, in this mode, your imagination has already taken flight, thinking about how this can be applied: rearranging your home, what to wear with what, what to buy in the store when you're standing in front of the shelf, and not understanding what to choose. By the way, men, this is a life hack for you now, and for women too, by the way. For men, there are two: your wife or girlfriend asks you to buy a certain product, like a certain shampoo; turn on the life hack; Gemini literally points to what to take; and imagine how cool it will be when Google releases normal glasses with AI inside. By the way, write in the comments how you would use this mode, it's very interesting to see, and if you've tested it, be sure to share.
The second application is video analysis. Gemini can now not only translate video into text, as many other models do, but it also reviews each of your recordings frame by frame, so you can upload very complex videos where something is happening, something dynamic is happening, and you can get a detailed breakdown and analysis. For example, exercise technique in the gym. I think many people were afraid to exercise on their own because they didn't know if they were doing it correctly. They recorded themselves on video doing squats or running, uploaded it to Gemini, and asked for an analysis. It will look at it and say, Here your back is rounding, here your knees are going forward, and so on, everything that needs to be corrected. Of course, I'm not saying that this is a replacement for a trainer, no, but it's definitely better than just doing it without supervision.

I think you will find many applications for yourself here as well. For example, if you play tennis or another sport, watch this video of me playing, analyze my technique, tell me what I am doing right and what needs to be corrected, and compare it, for example, with the technique of professional athletes. What about makeup for girls? learn how to do hairstyles, cook, in fact, I think anything, anything where you can analyze something visually to give you some advice. Here you go, please. The only limitation is that in the free version, you can upload videos up to five minutes long, but if you upgrade to Google AI Pro or Ultra, you can upload videos up to an hour long. The maximum file size is 2 gigabytes, which I think is more than enough.
The next feature is to turn a video into step-by-step text instructions. Cool, right? For example, you set up a coffee machine, repaired something, or replaced something in your car. Now you want to save it for yourself or your partner, or if you're renting out, I don't know, an apartment for tenants, or accompany your product on the website with text instructions, film what you are doing on your phone, upload the video to Gemini and write, make step-by-step text instructions from this video, and that's it. It will watch the video frame by frame and write down step one, click here, step two, wait until it lights up, step three, select the mode, everything in order.
This is exactly what multimodality is about: transferring meaning between content formats. I think you can come up with a lot of uses for this, like internal instructions for the team, for onboarding new employees, shooting a video, a screencast of how you work in a certain program, setting up a complex integration, throwing it into Gemini, getting a ready-made text guide, and using it to your heart's content.
The next application is content analysis or professional analysis of reels and how they work. For example, take the reels of a millionaire blogger whose format you like, or just a video of a brand you like whose success you want to replicate, and upload the video directly to Gemini. You can ask for a detailed analysis of why these reels became so popular, what keeps viewers' attention, how the author works with visuals, information, and frames, and which editing techniques are used. Gemini will analyze it frame by frame and write everything down in detail.
The second step is to upload your own reels and ask for a professional analysis, compare them with the benchmark you uploaded, and get specific advice on what to improve. You can ask for a detailed plan on how to get closer to the level you want, and then continue the conversation in the same chat. If you have an idea for a new video, post it here, and Gemini will understand the context. It has seen your examples, knows what you are striving for, and will help not only with the script but also with the presentation, what frame size to use, what text to display on the screen, and how to present the information visually.
An important point, of course, I understand that only personal experience makes an author an author, you shouldn't just dump everything on Gemini and wait for it to come up with something for you, otherwise it will just turn into rubbish, but as a critic it can do a great job, that is, it's more like a tool for learning, for studying, for correcting, if you have some important thought. You don't know how to convey it; in such cases, Gemini will help you strengthen your presentation. Alternatively, of course, it can be not only video analysis, but also the shooting itself, the lighting, and the arrangement of objects in the frame. You submit your version and the professional's version, and ask for an analysis of what needs improvement.

Another application is analyzing a room via video. In this situation, for example, you are a real estate agent renting out an apartment and need to make a list of furniture for the contract, or you are renting a place yourself and want to record what was there at the time of moving in general, we shoot a video on our phone, upload it to Gemini, and ask it to make a list of all the furniture you see in the video, assess the condition of the repairs, note if there are any defects. It will watch the video frame by frame and give you something like: living room, sofa, gray coffee table, TV on the wall, kitchen table, four chairs. In short, I think you understand, and you can immediately insert it into the contract.
But that's not all, there are tons of options. For example, you can ask it to make a floor plan. The next step is to ask it to help you draw up a detailed plan of this apartment based on the video, and make a detailed prompt for the image generator. I want to have a top-down view as well, to describe where everything is located or to provide a text description for the layout generator. This, by the way, is the next way to use it.
Another way to use it is to visualize a room in 3D. You already have a text description of the room from the previous point. Now, right in the chat and on Gemini, you can immediately turn on Ideogram, one of the most advanced neural networks for generating images, and, as an option, make a 3D render of this room. How it works is you take the description you got from Gemini, for example, and throw it into the chat. You can, for example, supplement it with a schematic drawing by hand, where everything is located so as not to confuse anything. You can even add specific photos of furniture if you want, with exact shades or features, everything, and you get a render like this, without a designer, without special programs.
Then you can play around with it. If you want to see what rearranging the furniture will look like, ask Ideogram to move the sofa to another wall, change the light curtains, or add armchairs in the corner. Again, the advantage of Ideogram is that you can upload specific references, specific images of the sofa you choose. Previously, you had to hire an interior designer to see what it would look like, but now you can do it yourself. Of course, it's worth noting that this is not a professional render, so you can't use it. The room's exact measurements are not taken into account, but AI is not yet very good at navigating space. Still, it's more for you, for peace of mind, to play around, to look, to experiment, I don't know, maybe you want to furnish a cottage, an office, a room, rearrange things, let your imagination run wild.

Okay, the next application is Gemini plus calendar plus tasks. Let's continue with the topic. Gemini can now manage your schedule and tasks, and it has become more convenient because it has access to both the calendar and the task app. For example, if you use a GTD planning system or something similar, you can discuss your plans for the next week in the chat, tell it what is important to you and what you would like to focus on, and ask it to add it to your schedule. Gemini will arrange everything on the calendar, set specific slots, and assign tasks with your deadlines. Just write that you have these plans, you want to get this, this, and this done, you have this meeting, and that's it. It will write it down, taking into account what you already have.
Another application of the previous one is that scheduled actions in the chat finally work properly. The function itself is not new; many people already have it, but when the assistant has the context of your life, your calendar, your tasks, and your email, it's a whole new level. For example, you can write directly in the chat. Every weekday at 8 a.m., send me a summary of my calendar, unread emails, and the top three important tasks for today. That's it, and it will not only remind you, but also plan it.
Or, for example, you can say,  Every Friday at 6 p.m., review my projects and tasks for the week, note where I fell short, and plan how to make up for it next week. That is, the function is the same, scheduled actions, but when the assistant has access to your ecosystem and understands the context, then it unfolds in a completely different way. It is no longer just a reminder, but a real working personal assistant.
The only thing is that I've just started testing it specifically in this context, so I'll probably gain more experience with it over time and then write a separate article. Perhaps even a comparison of different assistants, since ChatGPT also integrates with Google Calendar, but no matter how much I tried to set it up to test it, it turned out very clunky. It didn't see my meetings or take everything into account, so I just created my own personal assistant with n8n.
The next application is searching your own files. Gemini finally started to work really well with the entire ecosystem. When I talk about the Google ecosystem, I'm talking specifically about this: Gemini can be given access to your email, your documents, your photos, your spreadsheets, everything.
There's no need to worry about this because Google already has access to all your accounts and passwords. On the contrary, I think it's beneficial for you that you can now synchronize all of this using AI. I know that Brin had a conversation with the co-founder, and they said, "Listen, we have absolutely all the information in the world about every person." If we wanted to, we could manipulate the market, but we don't do that because we would lose trust. I'm sure they use some data, but my point is that if, for example, you don't want to trust third-party services with your access, why not give it to Google? You literally connect Google to Google.
It's all obvious here. So, you connect everything, and now you can ask it to find, for example, a particular document where we discussed the budget with the team. That's it, and it searches for it. The only thing you need to do to make it search for it is to put this little dog right in the chat. That's it, and then a list of services that you can look into appears. We can choose, for example, Google Drive, or choose Workspace as a whole and say what we want to find.

There are so many uses for it. For example, you can ask it to find all the receipts for the services you subscribe to and say Hey, how much am I paying each month for all these services? It will find them, search for them, and tell you. Cool. You can search text documents, look for contracts, and find contracts related to a particular customer or service. Listen, you can even summarize correspondence, and that's the next use.
And here, by the way, there are even two options. For example, the first is to request a summary of all emails for a given period. For example, if you have accumulated a lot of emails, you can ask to divide them into groups and give you a summary of everything. And the coolest thing here is not just the summary, but that you can click the link directly from the chat to open a specific email and reply to it. This is very convenient.
The second option is if, for example, you have accumulated a huge amount of correspondence with someone. Email is probably not the most convenient place to navigate through this correspondence later. That is, it is convenient to correspond, but navigating it is not. All these replies, replies to replies, huge lists. So, you can ask for a summary of this huge correspondence. Or find specific data, a specific message, for example, some details, or the email of the person who wrote to you. Please find me the email that the customer gave me. In general, I think you get the idea.
The next application is connecting NotebookLM to Gemini. In general, combining Gemini with other Google services addresses many fundamental problems. For example, in this case, you can significantly reduce hallucinations. This is probably the main reason why we rarely turn to neural networks for serious questions. When we ask AI something or even upload our files, manuals, books, and tables to the chat, it does not always rely on them. It may read a small part and take the rest from the internet, or it may simply want to answer itself. Accordingly, you are not always sure of the answer. You don't know if it was based on a knowledge base or if it was made up, because it responds so well, sometimes very convincingly. Accordingly, this gives rise to hallucinations.
Google has a separate tool that solves this problem. It's called NotebookLM. To explain how it works in simple terms, you upload your files, links, and documents there, and it will respond strictly based on them, only on your data, and also show you where a particular phrase was taken from. So, NotebookLM has been integrated with Gemini, and everything you create in NotebookLM now, all your knowledge bases, which are called notebooks there, can now be used directly in communication with Gemini. That is, you literally connect them to your assistant.

For example, you have a notebook with your expenses, health analyses, or career plans. And now, right in the chat, in the drop-down list, you can click NotebookLM, select the appropriate database and notebook, and it will refer to them. And, for example, ask: "Listen, are my financial goals that we just talked about realistic, considering the expenses in my notebook A and the work deadlines in notebook B?" It will rely specifically on your data. It won't make things up or hallucinate. That's the beauty of NotebookLM.
You can now create separate spaces for your different areas of life: health, finances, and various projects. And through Gemini, you can do a wonderful analysis of all these areas together. And if you also have a calendar connected, Gemini can work with Google Calendar, so you get a whole system. You can literally ask Listen, considering my plans in my notebook, my expenses, and my calendar, will I be able to achieve this goal? What could get in my way? How can I best organize my schedule? An ecosystem — that's what Google is trying to do. They already had it and connected it all to make it easier for you to work with.
Of course, it's worth mentioning that if you use Gemini once a week, you won't notice much of a difference. Naturally, for example, if you don't use calendars or documents, you have a couple of files there, and you expect it to do something unique, cool, and really useful, it won't. It is suitable if you use Google Calendar, or if you switch to Google Calendar now and use it to record your meetings. If you use notes and documents, then everything will be synchronized.
The next application is Gemini, which we now have in documents and spreadsheets. When you purchase a Pro subscription, you can use Gemini not only as Gemini itself, but also connect its services and use them directly in documents or spreadsheets, for example. There is a small button you can use to call up Gemini. That's it, and it sees the entire context of the page; you don't just connect it for no reason.
For example, I work this way with drafts. I jot down the times of meetings, some thoughts that are important to me, and then I ask Gemini to structure all this text. It's convenient, super convenient. If you have, for example, a contract in Google Docs, you don't have to copy it, paste it, ask ChatGPT, or drop it into Perplexity, as I sometimes do, to analyze or correct it. That's it, you can call Gemini right here and ask it to analyze it right here, correct it right here, highlight its weaknesses and strengths. How can it be improved? And then, with just one click, make changes to the document itself.
The same works with tables. No need to search for formulas anymore, you can make a basic table and then ask Gemini to make it look nice, add metrics, add some formulas, supplement the data, and analyze it. There are tons of applications. The only thing is that this particular application only works with a Pro subscription. But I think many people will switch to a Pro subscription after reading this article.

Methods 15-21: YouTube, podcasts, visuals, and unexpected applications

The next application is a podcast about hot summaries. I saw this trick on someone's Twitter, I don't remember whose, but I really liked the idea. It's called a podcast about hot summaries. But again, you can use it not only for resumes, but also for your project's professional experience, presentations, websites, anything you want to highlight, or even your personality. Well, I'll show you an example with a resume. It's very interesting to listen to. And you can do it either in NotebookLM or Gemini, whichever you prefer.
Upload your resume there, for example, your LinkedIn profile, or upload your portfolio file. If you're doing it in NotebookLM, select the audio podcast format, select the critique function, and add the following prompt. Critique this resume, identify its strengths and weaknesses, and give it a real grilling. Something like that, right? If you're doing this through Gemini, upload your resume here, select Canvas mode, and ask for criticism.
That's it, then create a podcast and listen to two AIs discuss your project or website. In general, there are various interesting applications. It's not particularly useful, but why not? Why not? The next application is debating with yourself, that is, working with your ideas in debate mode.
When you have a business idea or an idea for a product, an idea for some content, an idea for a lecture you want to give, a lesson, anything, anything you have, you can also listen to a podcast debate in audio retelling mode, where there will be different points of view on your topic. Today, we have on the table, let's say, a very interesting document — a business plan for opening an automatic car wash. I am sure that you will definitely find something useful for yourself.
For example, in my experience, I often listen to debates to get a broader perspective on a particular issue. And with Gemini, this opportunity increases exponentially, because where else can you hear debates on a specific topic that interests you? Where else can you find that? But here you can listen to it, it's very useful, listen to it, write your opinion.
Next, we have timecodes on YouTube. Working with YouTube through Gemini, it's integrated into our platform, opening up a sea of possibilities. My personal case is writing timecodes for videos. Yes, guys, we have them now. Now, finally, you can send a link to a newly uploaded video and ask for timecodes to be written. The only thing I've noticed is that sometimes it puts timecodes 3-4 seconds earlier than I would. Again, if I notice this, I ask to move all the timecodes forward by 2-3 seconds.
That's my personal case. In general, you can, for example, use specific moments in huge videos. For example, there is a two-hour lecture on proper nutrition, and you only need one part about intermittent fasting. Send the link and ask to find the timecode where intermittent fasting is mentioned. That's it, Gemini will indicate the exact moments. Please, this is how you can save time.

The next application is podcasts from YouTube videos. It is very convenient to create a podcast from any long video directly in Gemini. That is, you provide a link to the video and request a full transcription. Transcription is another way to use Gemini. Or you can write an article with the main thoughts or ideas from the video, or some rules. I often watch various life hacks where, for example, YouTubers or entrepreneurs share their advice.
So, you can ask Gemini to write it down and turn it into a study guide for you. The only important thing is to make sure you do this in Canvas mode. That is, you check this box so that the article appears in Canvas or the transcription appears in Canvas. And in Canvas, there is a button that says Create. That's it, and here you have the option of an audio podcast. And you can end up with a podcast in which two AI hosts discuss this topic.
By the way, I really like this format, and now you can do it in Gemini because Gemini has another product called NotebookLM. I love it even more. I've filmed a lot of material about it so that you can check it out on our website. It's a mega-cool tool for getting rid of hallucinations and creating your own knowledge base. In short, it's a great tool. Use it, it's free too. So, developers have now integrated tools such as audio podcast creation into Gemini.
Oh, and by the way, it's important to mention that you can make podcasts not only from videos, but from anything at all. You can upload an article, a PDF, or a text file, and it will turn all of that into a very interesting, exciting podcast, and you can listen to the topic.
The next application we have is generating photorealistic mockups and everything related to advertising. Gemini has an ideogram built in. It is currently the most powerful model for generating visuals, and it is intelligent. That is, you can give it not just a text prompt to turn into an image, but also a task. In addition, you can upload your actual product — a sports water bottle, a bottle of shampoo — and ask it to literally create it, for example, on a marble table, in the bathroom, or in the morning light. That's it, and you get a great advertising banner.
The only thing I will say right away is that when you look at the result in Gemini, as you can see now, sometimes the quality is not very good, and it may seem that it was generated in poor quality, but that's not the case. When you click Download, the image will be downloaded in a larger resolution. I think they did this on purpose to avoid slowing down the work. The only thing is that it will have a watermark, but there are now many services that can remove it. In general, there are many uses for it. Write down what uses you would find for yourself.

The next application is photo color correction. Ideogram can not only generate images from scratch, but also process your images. That is, you upload a photo here, and there are two ways to do this. The first is to use this prompt. It will perform color correction to highlight details and make the image look more professional. The second way is if you want something specific, you can first upload a photo frame, for example, where you like the color correction, ask it to analyze it, and then upload your frame and ask it to use the color correction from the previous file you uploaded on your frame. It does this really well.
The only thing is that if you regenerate it many times, some defects and distortions will appear. But on the first and second iterations, the photos are practically indistinguishable. That is, I mean faces, details, eyes, hands, everything remains the same, only the color correction itself changes.
The next step is to create visuals with you. This is a very cool application. For example, I use it to make covers for materials, but there are many other uses. And the cool thing here is that you can provide the input. That is, you create a template in any program, whether it's Paint or Canva, then add a photo of your face, then draw a car, and then throw this template into Ideogram and say, "Here's my template, make a full-fledged cover, poster, a photo, a trigger, a photorealistic image, keep the appearance.
The only thing that sometimes changes is the face, guys, it happens to me quite often, but there are always attempts when everything turns out well. But, again, you can do this with absolutely any visual. If you make cards for social networks, make posters using such templates, it is not always possible to find references. Now you can work with visuals in more detail and with greater precision.
By the way, there is also a handy feature: when you upload a frame to Gemini, you can highlight something with a pencil or point to something with an arrow, and then the AI will understand exactly what you mean.
The next application is auditing landing pages, websites, and books—basically, audits. As I said, Gemini has very powerful image analysis technology under the hood, but that's not all; it also navigates interfaces very well. Gemini currently places a lot of emphasis on the front end, which means it excels at landing pages, websites, and interfaces. I would say this is a fairly common story; many people are writing code these days.
I want to highlight a slightly different way of using it. Since it understands interfaces well and has been trained on a large number of pages, landing pages, and websites, you can ask it to analyze them. You can upload a screenshot of your landing page, your page, your digital product, or your social networks, and ask it to find 10 problems that kill conversion and suggest edits to the text, blocks, and visuals.
And it works really well with this, highlighting flaws and highlighting good points. Basically, good people give criticism.