Why microcopy first

Rebuilding a search algorithm or redesigning ad placement takes engineering time and product buy-in. Rewriting the words a user sees at the moment of confusion takes an afternoon — and often removes a meaningful share of the friction on its own. That made copy the fastest lever available while the larger fixes were still in planning.

Before → after

"Save" "Save to Board"
"Looking for ideas you've saved." "No matching results. Try refining your keywords."
Unlabeled star icon "Highlight this category"

Why each change works

"Save" → "Save to Board." The original was ambiguous about where content goes. Naming the destination removes a full beat of hesitation before the tap.

The empty-search message. The old copy described a different feature entirely (saved ideas, not search), which is confusing at the exact moment a user is already unsure why nothing appeared. The rewrite names the actual state — zero results — and gives an immediate next action.

The unlabeled star icon. An icon with no label asks every new user to guess its function. A five-word label removes the guess entirely, at negligible visual cost.

The throughline

None of these rewrites are clever. That's deliberate — the goal of microcopy in a product like this is to disappear, not to be noticed. Each change traces back to a specific moment identified in the empathy map or the audit, not a general "make it friendlier" instinct.