I studied more than 200 app onboardings flows. Here's what I learned

For the past 3 months, I have been studying the onboardings of the top mobile apps.

I learned a lot about how to make them, how to convert better and applied all of this to my own mobile app to reach more than $1,700 in organic revenue in less than 3 months.

In this article, I'm sharing everything I learned so you can apply it too.

1. Mascots

The best onboardings have a face.

A mascot creates instant emotional connection. It guides users through steps, softens friction and makes the experience feel less like a form and more like a conversation.

Duolingo's owl is the clearest example but it works at every scale. Even a simple illustrated character that reacts to your choices changes the feel of an onboarding completely.

If you're not using one, test it. The personality gap between apps that do and don't is massive.

Some of my favorite examples are Duolingo, Catzy and Falou.

2. Video demos

Nobody reads features. Everyone watches them.

A short looping video inside your onboarding, showing the actual product doing actual things, does more selling than any bullet list. It answers the silent question every user has: "But what does it actually look like?"

Keep it under 15 seconds. Show the aha moment. No voiceover needed (but I have seen it also).

2 good examples doing this are DecAI and Plantum:

3. Collect user data

The majority of apps ask you 10–15 questions before you ever see the product: your goal, your current habits, your biggest obstacle. Then they spit out a personalized plan built just for you. That plan isn't magic. It's your own answers reflected back. But it works because now you feel seen, and you're invested before you've even paid.

That's the first reason to collect data: it makes the experience feel personal, which makes users far more likely to convert and stick around.

The second reason is attribution and this one is criminally underused.

Ask users how they heard about you. Early. In the onboarding.

No analytics tool gives you this accurately. Word of mouth, a friend's recommendation, other social media app. these channels are invisible to your dashboard but crystal clear to your users.

The data you collect here does three things: it personalizes the experience, it tells you where your users come from and it lets you message smarter for the rest of the user's life.

I only added this to my own app recently. One question, a handful of options and for the first time I could see exactly where my users were actually coming from. Should have done it on day one:

4. Review prompt mid-onboarding

Most apps don't ask for a review at all.

The best apps prompt for a review during onboarding, right after a moment of delight. The user just saw something that impressed them. They're warm. They haven't hit any friction yet.

That's when you ask.

The timing is everything. A review prompt mid-onboarding, placed right after a "wow" moment, has a high chance to get a positive review.

I have added it myself in my app and I managed to get 152 ratings with a 4.8 out of 5:

5. Paywalls (hard and soft)

This is where most indie developers leave money on the table.

According to RevenueCat's 2026 State of Subscription Apps report (built on data from 115,000+ apps and $16B in revenue) hard paywalls convert 5× better than freemium (10.7% vs. 2.1% download-to-paid by Day 35).

The data also shows that hard paywall apps generate 8–9× more revenue per install at both Day 14 and Day 60.

But here's the nuance: that advantage doesn't mean you wall everything off immediately. The key is timing. Show the product's value first. Build momentum in onboarding. Then show the paywall, when the user already wants what you're offering.

Soft paywalls (freemium with gated features) work better when your app benefits from word of mouth or network effects. Hard paywalls work better when time-to-value is fast and the product speaks for itself.

Know which one you are.

And once you know, experiment. The best paywall isn't the one you designed first, it's the one you iterated to. RevenueCat makes A/B testing paywalls surprisingly easy: you can set up experiments, swap designs and measure real revenue impact without touching your code. If you're not running paywall experiments, you're guessing.

This "Maximize your height" app is a great example of how to do a hard paywall:

6. Closing offer when closing the paywall

Most apps lose users the moment they close the paywall. That exit is actually a second chance.

When a user dismisses your paywall, trigger a closing offer. A discount. A shorter commitment. A free trial they didn't see before. Anything that lowers the barrier for the "almost" users.

This is one of the highest-leverage moments in your entire funnel. The user already showed intent. they saw your paywall and considered it. A well-timed offer here can recover a significant chunk of those near-conversions.

Don't let them leave without making one more ask.

For example, in this app if you close the paywall, you get a roulette with a 74% limited-time discount:

7. Weekly, monthly, lifetime + trial combinations

Don't offer one plan. Offer a choice but make the right choice obvious.

The data from RevenueCat shows that apps whose most popular plan is yearly generate the highest RPI at both Day 14 and Day 60, roughly 2× more than monthly-dominant apps. But that doesn't mean you should hide the other options.

The combination that works best in practice: weekly + yearly + (optionally) lifetime, with yearly highlighted as the recommended option. The weekly plan makes the yearly feel like a deal. The lifetime plan makes the yearly feel like the smart middle ground.

On trials: longer is better than most people think. RevenueCat's data shows trials of 17–32 days convert 70% better than trials of 4 days or less (42.5% vs. 25.5%). Yet nearly half of all apps now use trials of 4 days or less. The industry is leaving conversion on the table

Test a longer trial. You might be surprised.

The first session decides everything.

RevenueCat's data puts it plainly: 55% of all 3-day trial cancellations happen on Day 0. The battle for your subscriber is won or lost in the first few minutes.

Onboarding isn't just a must have. It's your best sales tool

With tools like Anything + RevenueCat is hasn't easier to do it. Build the thing.

If you want to follow my journey building in public, I share everything at @cesaralvarezll.

And if you're serious about growing your app. I'm building Feectory, the growth factory for app developers. It's on waitlist now. Would love to have you.

Links

• x.com/cesaralvarezll
• feectory.com/