Watching people abandon a form taught me more about UX than any design system

Starting a goal tree from a blank page is intimidating. So we built a way to browse goal trees other people already made and adopt one as your starting point instead. We call it a trail: pick someone's marathon plan or reading habit setup, and it becomes your own tree from day one, not a blank canvas you have to fill in alone.

This update is about the screen for submitting a trail.

The first version was one long form: pick a goal, then a wall of fields underneath it. I tried it on myself and lost track of how much was left. Category, title, description, then more fields I'd forgotten I needed to fill. No sense of progress, just scrolling.

So I split it into three steps: pick, fill in, review. A thin progress line connects them. You can click back to any completed step instead of hitting a browser back button and losing your inputs.

Small change on paper. In practice it turned a form people bailed on halfway through into one that feels like it has an end.

Two smaller decisions came out of the same pass:

- Trail links now use a short numeric address instead of a long UUID. Nobody wants to paste that into a text message.

- The trail and settings screens open as a modal over your tree instead of taking the whole page. You never lose the thing you were looking at when you clicked in.

None of this is complicated engineering. What I keep relearning: a feature living behind one screen doesn't tell you where people get stuck. You have to watch someone hit the wall themselves, or hit it yourself trying to use your own product honestly.

Retic is still in pre-registration at retic.io while I build out the flow for previewing and adopting someone else's trail into your own tree.

If you've shipped a multi-step form before: how many steps is too many before people start dropping off? Did you find that out from data, or from just watching someone use it?