I used to think onboarding was unnecessary. I was wrong.
For the longest time, I avoided building onboarding for my product (chrome extension).
My thinking was:
“it’s simple enough”
“people will figure it out”
“nobody wants tutorials anyway”
But then I kept noticing the same thing:
Users were only discovering the obvious features.
Meanwhile things like focus mode, zoom mode, virtual backgrounds, customization options, advanced settings, etc. were barely being used.
The problem wasn’t complexity.
It was discoverability.
⸻
So instead of building a traditional onboarding with screenshots and arrows…
I built an interactive “copy” of the product itself.
✦ If the product IS installed:
- the onboarding syncs with the real UI in real-time
👀 If it’s NOT installed:
- the website renders a realistic version of the UI in the exact positions where it would normally appear
- controls, popups, floating elements, everything
- users can still interact with the flow and understand how the product works before installing anything
The funny part is that many people don’t even realize they’re looking at a simulated version 😂
📈 What surprised me most
People actually started discovering and using features on their own, without support messages or documentation.
And somehow...
Within 2 days of launching it, the number of users increased by ~800 🤯
Honestly still trying to process that part 😅
⸻
This whole thing completely changed how I think about onboarding.
Sometimes the product is simple.
But discoverability isn’t.
The reframe from 'complexity' to 'discoverability' is the right one, but I'd push it one level further: the problem is usually that users build a mental model of what your product does from the first session, and then stop exploring because the mental model feels complete.
The interactive onboarding you built is clever because it plants a richer mental model before they even install. The 800-user lift probably isn't from the onboarding being more helpful -- it's from users arriving with higher expectations and therefore looking for more.
One place this pattern shows up outside product UI that often gets missed: freelancers and service providers have the same problem with clients. A new client builds a mental model of what the freelancer can do from the first project or two, then stops asking for anything outside that. The features they're not using are 'off-scope' by assumption, not by decision.
The ones I've seen handle this well treat every client touchpoint as a discoverability moment -- the kickoff call structure, the first-delivery email, even the invoice copy, all include subtle signals about what else they do. It's not pitching; it's the equivalent of your interactive onboarding. You're showing, not telling, so the client updates their mental model without feeling sold to.
Curious whether you saw any data on feature adoption by the users who came in through the new onboarding vs. the users who pre-dated it. That natural experiment would tell you a lot about how durable the mental model uplift actually is.
Love this. The insight about “it’s not complexity, it’s discoverability” hits hard. The simulated UI onboarding is a super smart way to bridge that gap.
Appreciate this a lot :) yeah, that shift from “explain” → “let them experience it” changed everything for me.
"Discoverability isn't" is such a clean reframe. The problem you're describing shows up in analytics data too — feature adoption rates in event tracking often look like a power law: one or two features dominate, everything else flatlines. Most founders interpret that as "users don't want those features" when it's actually "users never found them."
The interactive onboarding approach you built is clever because it closes the loop between the product and the mental model before the user even installs. What did you track to confirm the 800-user lift was from the onboarding specifically vs. other factors? Curious whether you had any instrumentation in place to isolate it.