6 design misses stopping Gemini from catching up to ChatGPT

Gemini comes with a major update that breaks the internet, everybody jumps on it to check it out, and after a day or two of hype, everybody goes back to using ChatGPT. The familiarity with ChatGPT trumps all the innovation Google has been making, but there’s another reason why Gemini still trails. The Gemini app simply isn’t as inviting as ChatGPT’s, which has been around longer and has a better way with people.

Google is already working on a big redesign for the Gemini app, and I have a wish list for what it needs to get right.

Long-term memory and continuity

Gemini only recently got the handiest feature that ChatGPT has had forever — cross-chat memory. The feature lets ChatGPT remember what you mention in a chat, and it becomes a reference for your future interactions by default. So you don’t have to tell it over and over how you prefer certain things and what you want avoided (like fewer em dashes, heh). It works like a person’s memory, just sharper and without the fading.

Gemini’s version still feels limited compared to ChatGPT, as if it’s a kid who’s forgetful and struggling to catch up with all the information thrown at it. For one, I couldn’t spot Gemini’s Saved Info feature because it’s called something entirely different on the Settings page — I had to Google it (ironically). Speak of user convenience!

Moreover, Gemini didn’t automatically add my instructions to its memory for following chats, which ChatGPT has long been doing effortlessly. Google did announce such a feature, but the implementation still feels far behind ChatGPT.

Folder organizer

I’ve started using ChatGPT for a lot of things in my life, from checking minor health issues (remember it’s not a medical tool) to keeping track of my finances. With so many varied tasks at hand, thanks to my muscle memory of firing it up at the slightest inconvenience, my ChatGPT needed some organization — and it had the feature built in.

Instead of staring at a wall of chats, I’ve siloed them under different projects, with multiple chats living under each for easier access. Gemini continues to be a plain list of chats that sits unorganized, and you have to scroll through and read a bunch of titles to figure out what you’re looking for. I’m not claiming ChatGPT’s approach is bulletproof, but it’s still a solid starting point for Gemini, which it seems to be adopting soon-ish.

General user-friendliness

The Gemini app right now feels more like an enterprise tool than a consumer one. When you open it, you’re greeted with a promotion for the latest Gemini model the app is running on — when end users are least concerned about what model is running behind the scenes. They just want their emails written quickly. What they expect is a predictable interface that invites them to talk to the chatbot instead of pushing them away slightly.

This becomes obvious only when you open the two apps side by side. On ChatGPT, you get a welcome page that isn’t packed with text, has the model name tucked away in a corner instead of being prominently advertised, and the text box gets all the focus, along with a couple of starter prompts to nudge you to begin. If Gemini picks up one thing from ChatGPT to make its app better, I want it to be this.

Third-party integrations

Google has a pretty solid app ecosystem of its own — you can get your productivity sorted with Workspace apps and come to YouTube when you’re bored. Gemini integrates across all those apps seamlessly without requiring any additional setup. It just works — as long as you’re in the Google ecosystem, that is. It cannot connect with any third-party app, and guess what, people do use apps other than Gmail and Docs.

ChatGPT has a much richer ecosystem of third-party integrations. You can talk to Spotify or Slack or even Google Drive, plus there are a bunch of custom GPTs to play with. That kind of interoperability makes it a much smarter app for anyone who isn’t strictly locked into the Google ecosystem. Opening up Gemini to more tools outside Workspace would make it a more capable assistant.

Clearer quotas and timeline

Google’s One AI plans have tons of variations these days, each with a different quota, and the app also limits Gemini usage on the latest LLM when there are too many concurrent users. So you never really know whether you have three messages left or thirty. This is more of a wishlist — since ChatGPT doesn’t do it yet either — but I wish Gemini would clearly show how much daily quota is left. A simple progress bar would be enough.

While we’re at it, what I really want Gemini (and even ChatGPT) to adopt is awareness of exactly when a message was sent. This would help it keep a precise track of when we had a particular exchange and how much time has passed since then. This is especially helpful in chats that require timeline knowledge, like my history with a client I’m working on a long-term project with.

Better cross-platform access

Since it’s a Google product, Gemini obviously works better on Android phones — you can access it from multiple locations: a corner swipe, the search bar, a voice command, and from inside apps. But you can’t say the same on an iPhone. Except for the dedicated mobile app, you can’t even open Gemini through the main Google app, adding a lot of friction. On top of that, Gemini doesn’t even have an app for Mac or Windows, which makes it feel behind the curve compared to ChatGPT. Maybe a better presence across the board could help Gemini’s case.

How has your experience with Gemini been so far? Is there anything you saw on the ChatGPT app that you wish Google had copied for Gemini long ago? Let us know in the comments below.