“Engagement” is one of the most quietly corrosive words in design. It sounds like a measure of value — people are engaged, so the thing must be good. But engagement measures time and taps, not benefit. A slot machine is extraordinarily engaging. The question worth asking, before optimizing anything, is whether you’re designing to help someone accomplish a goal and leave, or to keep them from leaving at all. Those are opposite designs, and most teams never admit they’ve chosen.
The two goals pull in opposite directions
Once you see the fork, you can’t unsee it. Almost every interaction pattern serves one master or the other. The infinite scroll, the autoplay, the streak counter, the “you have 3 notifications” — none of those help you finish. They’re friction against leaving, dressed as features.
| Designed for engagement | Designed for calm |
|---|---|
| Success = time on screen | Success = task done, user gone |
| Infinite scroll, autoplay next | A natural end, a clear stopping point |
| Notifications that manufacture urgency | Notifications only when something truly needs you |
| Streaks and guilt to pull you back | No penalty for living your life |
| Measures taps | Measures whether you got what you came for |
Calm is measurable too — you just have to choose the metric
The usual objection is “calm is fuzzy, engagement is a number.” But calm has numbers too: task completion time, how quickly someone accomplishes what they came for and leaves, how rarely they need to come back. A tool people use for two focused minutes and then forget about all day can be wildly successful — it’s just successful on a metric most dashboards aren’t set up to celebrate. The reason engagement wins isn’t that it’s more real. It’s that it’s easier to grow and easier to sell.
The studio’s whole posture is a bet on the other metric: a surface that stays still until you need it, then gets out of your way. (The principle, applied to motion specifically, is in Motion Born From Stillness.) That’s not a stylistic preference. It’s a decision about whose interests the design actually serves.
How to Proceed
- Write down what your interface is actually optimizing for. If the honest answer is "time on screen," name it — you can't change a goal you won't admit.
- Find every pattern whose job is to prevent leaving (infinite scroll, autoplay, streaks). For each, ask if it helps the user or only the metric.
- Give your product a real ending — a point where it says "you're done" instead of "here's more." Endings are a feature.
- Pick a calm metric to track alongside the loud one: task completion, time-to-done, how seldom people need to return.
- Audit your notifications. Silence every one that manufactures urgency rather than reporting something that genuinely needs a human.