A common activation problem
Every new user gets the same generic start
The short answer: ask one question, shape the start. Here is how to recognise this problem in your funnel, why it happens, and how the best consumer apps design it away.
How to recognise it
- Activation varying wildly by audience segment
- Early content engagement close to random
- Users digging through settings before their first win
Why does a generic first experience hurt activation?
A generic start is optimised for the average user, and almost nobody is the average user.
A runner and a weightlifter opening the same fitness app want different first screens. Serve both the same one and each concludes, correctly, that the app was not really built for them.
The irony: most products already ask questions during signup, then do nothing visible with the answers. Asking without acting is worse than not asking, because it teaches users their input changes nothing.
How do you personalise onboarding without friction? Ask one question
Find the one question whose answer changes the most about what the user should see first. Goal, level, use case: whatever splits your audience widest.
Ask it, then make the answer visibly matter within seconds: the feed reshapes, the plan adjusts, the examples switch. The point is not the data. The point is the user watching the product bend around them.
One question that transforms the start beats ten that fill a profile. Every additional question before value pays rent in drop-off, so each must earn its place with a visible difference.
And if you already collect answers today: wire them up before adding anything new.
The building blocks that solve it
- Personalisation: Shape what each user sees around what they told you.
- Investment: Every answer a user gives should sharpen their product.
Who does this well
Pinterest: Pick a few topics and the entire product rebuilds around them before your eyes.
Headspace: One question about what brought you here, and the first session matches the answer.
Questions founders ask
How many onboarding questions should we ask?
As few as change something visible, which for most consumer apps is one to three.
The honest test per question: could the user point at the difference their answer made? If not, the question is a survey, and surveys belong after value, not before it.
What if the product can't personalise much yet?
Even shallow personalisation reads as care when it is visible: reordering the same content by stated goal, naming their goal on the home screen, tuning the examples.
The rule that survives every stage: never ask what you won't visibly use.