The setup

This is placeholder content showing the shape a good teardown takes. Swap in your own analysis — the structure is the useful part. Start with what the product is trying to get the user to do, in one sentence. For Duolingo, it's simple: open the app again tomorrow, not just today.

What's actually working

Name the specific mechanic, not the vague concept. It's not "gamification" — it's the streak counter, the loss-averse framing ("don't lose your streak"), and the push notification timed right before the day resets. Explain why each piece does its job, using a concrete moment: the freeze-streak item, the mascot's slightly guilt-tripping copy, whatever detail you actually noticed.

Screenshot of the streak counter screen
Drop a real screenshot at assets/images/project-01-screenshot.jpg — this figure hides itself until one exists.
A good critique names the exact screen, the exact word, the exact moment — not the general idea.

Where it breaks down

Every mechanic has a cost. Maybe it's the users who keep a streak alive by doing the laziest possible lesson, learning nothing. Maybe it's the anxiety a broken streak creates. Say what you'd measure to know if this was actually a problem, and what you'd try instead.

What I'd do differently

End with a point of view, not just observations. One or two concrete changes you'd propose, and the trade-off each one makes. This is the part hiring managers actually remember.