The check-in, and what it learns.
No account, no setup beyond picking apps. Here's exactly what happens after that.
The check-in itself
Open a managed app and, before it appears, Rudder shows a short screen: a plain prompt ("What are you here for?", "Did you mean to open this?"), a brief enforced pause, then two ways forward — continue into the app, or go back. That's the whole interaction. It doesn't ask you to justify yourself, and it doesn't get longer or more judgmental the more times you see it.
Pick a mode, per app
When you add an app, you choose how Rudder should treat it:
Static
The same check-in, every open, forever. Nothing adjusts. Useful when you already know exactly what you want from that app and just need the pause.
Adaptive
Rudder starts the same way, then watches how you actually respond over real use — and can adjust the check-in itself if a clear pattern shows up. See below for exactly what it looks for.
How adaptive mode actually changes
Not a general "AI decides" black box. Rudder looks for real, repeated patterns in how you respond over actual use — never a snap judgment from one or two opens — and adjusts the check-in itself when there's enough evidence: sometimes making it matter more, sometimes trying a different approach, sometimes asking you directly rather than guessing. It's deterministic and explainable, not a model quietly making its own call.
You can always see it, and undo it
Every adjustment Rudder makes is visible on the home screen — what changed, why, and the real before/after numbers behind it, not a vague "we improved this." If a change doesn't feel right, switching back is one tap.
What doesn't change
Nothing about the check-in's core shape changes: still one short prompt, still one pause, still your call to continue or go back.
More on what Rudder does and doesn't do with your data: Privacy.