Designing an App People Keep Using
Last updated:
The first session decides most of it
Typical business apps lose a large share of installs within a week, and most of that loss happens in the first session. Someone installed it, opened it, and did not find the thing they came for quickly enough.
Retention work therefore belongs at the start of the funnel, not in re-engagement campaigns aimed at people who already left.
Get to value in under a minute
- No account required to see anything useful, where that is possible
- If sign-in is required, make it one tap — platform sign-in or an email link
- Skip the tour; show the thing
- Pre-fill everything you can already know
- Ask for permissions later, at the moment they are needed
Every screen between install and value costs you users, and the registration form costs you more than any other single screen.
Design for the thumb
Phones are used one-handed, often while doing something else. The top of a large screen is genuinely hard to reach.
- Primary actions in the lower half of the screen
- Touch targets at least 44 points; nothing important near a screen edge
- Destructive actions away from where the thumb rests naturally
- Bottom navigation rather than a top drawer
- Test one-handed, standing up, in bright light
Never make people wait without explanation
Perceived speed matters more than measured speed. An app that shows a skeleton layout immediately feels faster than one that shows a spinner for the same duration.
- Skeleton screens instead of spinners
- Optimistic updates — show the result, reconcile behind
- Anything over about ten seconds gets progress and a cancel
- Cache the last known data and show it while refreshing
- Errors that say what to do, not what went wrong internally
Give a reason to come back
The apps that survive have a recurring hook: new work assigned, an order status change, a weekly summary, something to check. Without one, even a well-designed app fades from the home screen.
- Something changes that the user cares about
- A notification tells them, without being noise
- The app opens directly to the thing that changed — not the home screen
- The task is completable in under a minute
That third point is neglected constantly. A notification that dumps the user on a generic home screen wastes the attention it just bought.
What we watch after launch
| Metric | Healthy | What it tells you |
|---|---|---|
| Day 1 retention | >40% | Was the first session worth it? |
| Day 7 retention | >20% | Is there a reason to return? |
| Day 30 retention | >10% | Has it become a habit? |
| Time to first value | <60s | Onboarding friction |
| Crash-free rate | >99.5% | Technical quality |
These are rough benchmarks for business apps rather than laws. The trend on your own numbers after each release matters far more than comparison with anyone else's.
Frequently asked questions
How do we improve retention after launch?
Should we copy the big apps' patterns?
Does visual design affect retention?
How many screens should an app have?
Considering an app for your business?
Tell us how often a customer would open it. That one answer usually settles whether you need an app or a much cheaper mobile site.
Related services
What we build for problems like this one