The commitment you can see does the work your reminders can't.

Every product that wants someone to come back eventually reaches for the same tool: the reminder. A push, an email, a red badge. And every product learns the same lesson — reminders are cheap to send and easy to ignore, because they come from you. They’re your voice, nagging. The thing they’re trying to substitute for is something the user was supposed to want on their own.

There’s an older, quieter lever that works on the opposite principle. Instead of you pushing from the outside, you let the user’s own earlier self do the pushing. The name for it in the research is commitment and consistency, and once you see it you’ll notice how many good onboardings are secretly built on it.

The mechanism is this. People carry a strong, mostly unconscious drive to act in ways that line up with what they’ve already said and done. A stance you’ve taken becomes a small debt you owe yourself. Breaking from it costs something — a flicker of the same discomfort that shows up whenever your actions and your self-image disagree. So the brain quietly steers to avoid the bill. The interesting part for builders is that this drive is not evenly strong. A commitment binds hardest when it is active (you did something, not just nodded along), voluntary (you chose it, nobody cornered you), effortful (it cost a little), and above all visible — written down, said out loud, put somewhere it can be seen again. A thought you merely had is easy to walk back, because there’s no record to contradict. A thing you typed into a box and watched appear on a screen is now evidence.

That last property is the one you can build around. Most products collect intentions in the weakest possible form — a passive one. The user clicks “Get started,” which commits them to nothing, and then you spend the next three weeks firing reminders at a person who never actually said what they wanted.

So the build decision is small and concrete: at the moment a user feels the intention most sharply — right after a win, right when they’ve seen what the product is for — capture it as a commitment in its strongest honest form. Have them write the goal in their own words, not pick from a dropdown. Let them name the thing they’re here to do. Then do the part most products skip: show it back to them later, in their own words, at the moment they’d otherwise drift. “You said you wanted to ship a draft this week” lands differently than “You have unfinished items.” One is your voice. The other is theirs.

A warning, because this lever has a dark twin. The same drive that helps a willing person follow through can be used to trap an unwilling one — the “you already started, might as well finish” pattern, the commitment manufactured by a confirm-shaming button. The difference isn’t the mechanism; it’s whose goal it serves. A commitment surface is honest when the user actually wants the thing and you’re helping their present self keep faith with their earlier one. It curdles the moment you’re using consistency to keep someone pointed at your goal instead of theirs. Build the version where, if the user’s intention changes, changing the stated commitment is as easy as making it. The pull should be a handrail, not a cage.

Reminders ask the user to trust you. A commitment they can see asks them to trust themselves — and that’s a debt people are far more willing to pay.