CONVERSION

Microcopy rewrite with rationale

Use when buttons, labels, and helper text are vague and you need alternatives you can defend.

microcopy-rewrite-with-rationale.md

Download .md

You are rewriting interface microcopy. Every rewrite must have a stated reason.

ELEMENTS TO REWRITE (current text, element type, where it appears): {{CURRENT_MICROCOPY}}
WHAT THE USER IS TRYING TO DO AT THIS POINT: {{USER_GOAL}}
WHAT HAPPENS IMMEDIATELY AFTER THEY ACT: {{NEXT_STEP}}
BRAND TONE CONSTRAINTS: {{TONE_RULES}}

For each element, give: Current | Problem with it | Option A (clear and plain) | Option B (specific about outcome) | Which you recommend and why | Character count of the recommendation.

Rules for the rewrites:
- Button text must say what happens next, not what the user is doing generically. Avoid "Submit", "Continue" where a specific verb exists.
- Helper text goes above the field if it is needed to fill it in, below if it is only reassurance. Say which for each.
- Never promise something the next step does not deliver.
- No exclamation marks. No "just", "simply", "easy", or "quick". No em dashes.
- If the current copy is already fine, say so and move on rather than producing a change for its own sake.

At the end, flag any element where the wording problem is really a flow problem that copy cannot fix.

Fill in before running

Replace each placeholder with your own detail. The more specific you are, the less the model invents.

{{CURRENT_MICROCOPY}}
{{USER_GOAL}}
{{NEXT_STEP}}
{{TONE_RULES}}

Getting a better result

Give the character limits your design imposes, or you will get options that do not fit.
Include the confirmation screen copy too, so the promise and the delivery can be checked against each other.
The final flag list is worth reading first - some copy problems are structural.

Questions about this prompt

When is a microcopy pass the right move rather than a page rewrite?

When individual buttons, labels and helper text are vague and you need alternatives you can defend in review. It declines to change copy that is already fine, so it will not manufacture work. If the sections arrive in the wrong order, the section order prompt is the better starting point and changes what the copy must do.

What do I need in front of me before running it?

The current text with element type and position, what the user is trying to do at that point, what happens immediately after they act, and your tone rules. Add the character limits your design imposes or you will get options that do not fit. Include the confirmation copy so promise and delivery can be checked together.

What comes back, and which part is worth acting on?

Per element: the problem, a plain option A, an outcome specific option B, a recommendation with reasoning and a character count. Read the closing list first though. It flags elements where the wording problem is really a flow problem, and no button label fixes a step that should not be there in the first place.

What is the mistake that costs me here?

Shipping option B everywhere because it sounds more confident. Button text has to describe what actually happens next, and an outcome led label that overpromises converts a copy win into a support ticket. Respect the helper text placement calls too: above the field when it is needed to fill it in, below when it is only reassurance.