Get Your Money's Worth from Gemini (2026 Guide)
Gemini is Google's equivalent to ChatGPT, and if you are already paying for the Google Suite, you probably do not need a separate ChatGPT subscription. I would go as far as saying the twenty dollars you pay for your Google Workspace or Google account are the best twenty dollars you can spend on any AI service, because they give you access to Gemini, NotebookLM, AI Studio, and more. In this walkthrough I focus on Gemini itself and show you every feature that earns its keep.
What you need
- A Google account, ideally with a paid Google Workspace or Google AI plan
- Nothing else: every tool here lives inside gemini.google.com
1. Pick your model: Fast, Thinking, or Pro (0:39)
On the Gemini landing page you choose between Fast, Thinking, and Pro. Honestly, I always keep it on Pro. From time to time Google can have problems and you will have to switch to Fast, but do not get scared: the quality of the fast responses is very good, so you are not getting less value out of it.
2. Deep research (1:26)
The standard tools work just like ChatGPT, and deep research is the first one to master. If you want to research a subject and get the sources behind that research, this is the mode. You actually get more deep research uses in Gemini than in ChatGPT, and the quality is just as good. Use it to learn a topic, do market research, or prepare an outreach: it is great at finding contact information.
3. Video generation with Veo 3 (2:00)
Video generation is something ChatGPT simply cannot do. There is no video model inside ChatGPT, and you cannot use Sora there; you have to go to another application. In Gemini I generated a test video with Veo 3, one of the best video generation models out there. A lot of the Instagram channels you see that are one hundred percent AI use Veo 3 in the background. Sora is the other very good model, but it is still not accessible everywhere. Veo 3 is.
4. Nano Banana Pro: images that reason (2:39)
This is probably my favorite Gemini feature. Nano Banana Pro is single-handedly the tool that changed image generation forever, because it does not only generate cute images, it reasons visually. In the video I show four things in a row:
- A clean image of the Pantheon, then a full subject swap into a new image.
- Combining subjects from several different images into one. You can combine up to twelve subjects without any hallucination in the image.
- Changing the perspective of an existing photo, which has a lot of very good use cases in architecture.
- Creating angles that never existed in the original shot, like finally seeing the top of an object the camera never captured.
I have a dedicated video on Nano Banana Pro and will keep making more, because this tool deserves it.
5. Canvas: prototype landing pages in minutes (4:21)
Canvas works exactly the same as in ChatGPT: it lets you create landing pages inside Gemini. In the demo I built a cat bar, a bar where there are cats, with landing pages for all the services, in a matter of probably two minutes. The UI is not the best out there, but it is perfect for prototyping quickly, checking that your contrast is right, and validating what should go on the page before you invest real design time.
6. Gems: your custom assistants (5:00)
Think of Gems as literally the same thing as custom GPTs in ChatGPT. A Gem is a very precise assistant that helps you with one thing only, like a brainstorming coach. Google ships several pre-made Gems, each with conversation starters you can click to begin.
What is great about Gems is sharing: you can send a Gem to a teammate. One thing to understand, though: a Gem can only be modified by the person who created it. If you know your team will ask you to change the Gem time and time again, the best move is to send them the instruction file and have each person create their own Gem.
7. Create your own Gem (6:10)
Click New Gem, give it a name, and give it a description. The description is only for you, so you remember what the Gem is for. Then write the instructions, which is where the magic happens:
You are going to behave as my assistant.
I want you to be doing this, this, and this.
That is the whole setup. Describe the behavior you want and iterate.
8. The underrated feature: Workspace integration (6:31)
Gemini is already connected to your entire Google Workspace: Calendar, email, Sheets, Documents, and Drive. You can read any document inside your Google Drive by typing the @ symbol and specifying Workspace or Drive. This wins you an enormous amount of time you would otherwise spend browsing databases hunting for the right document. It is very underrated, and it is the reason I tell people already on Google Workspace not to pay for a ChatGPT subscription: everything you need and more is already included, so you save money and get better performance out of your dollars.
Pitfalls and tips
- If Pro mode has issues, switch to Fast instead of waiting. The quality holds up.
- Gems cannot be edited by teammates, only by their creator. Share the instructions file when a team needs to own its assistant.
- Canvas output is for prototyping, not final design. Use it to validate structure and contrast, then rebuild properly.
Where to go next
Nano Banana Pro deserves its own deep dive: Nano Banana Pro Changed Everything About AI Images. And to compare against the competition, read Get Your Money's Worth from ChatGPT.
