Users go through the same onboarding twice when they start a new project
Your product should learn from how each user works and apply those preferences automatically.
When to use
Users repeat configuration steps across projects or sessions. Your product doesn't learn their preferences.
Hypothesis template
If we remember and apply user [preferences/settings] automatically, satisfaction will increase by [target]% because the product feels smarter over time.
Method
The problem: Users set their preferences once, then have to set them again for every new project, workspace, or context. The product doesn't learn.
How to do it:
- Track user preferences: default views, sorting, filters, theme, last-used settings
- Apply them automatically in new contexts
- Show that you remembered: "We set this to your preferred [setting] — change anytime"
- Learn from behavior: if they always sort by date, make that the default for them
- Don't be creepy — be transparent about what you remember
Key insight: The product that remembers how you work feels personal. The one that makes you reconfigure everything feels like a chore.
Success metrics
- •Repeat configuration time
- •User satisfaction scores
- •Feature adoption of smart defaults
- •Session efficiency
Prerequisites
- A preference storage system per user
- Settings that can be inferred from behavior
- A way to surface that preferences were applied
Common pitfalls
- •Being too aggressive with inferences
- •Not letting users override smart defaults
- •Not being transparent about what you track
- •Applying preferences that don't transfer well across contexts
Source: Figma, VS Code. Smart preference memory makes products feel personal and reduces friction.
Suggested ICE scores
Ready to run this experiment?
Sign up free to use this playbook with step-by-step guidance and track your results.