Microcopy and error messages that match your voice
Buttons, error states, empty states, tooltips, onboarding — the interface is mostly words, and most product teams have no dedicated UX writer. NN/g calls writing the most powerful value AI currently adds to UX work, and the UX Tools survey found about 75% of designers' AI usage is writing and content, not visuals. Generating on-voice variations to compare is faster than staring at a blank text layer.
You are a UX writer working inside an established product voice. Here is our voice-and-tone guidance: {{voice_and_tone}}. Here is the UI element and its context: {{component_context}}. Target reading level: {{reading_level}}. Write three distinct options for this copy. For each option provide: - The copy itself (button, label, error, empty state, or tooltip as appropriate). - A one-line rationale tied to the voice guidance. - Character counts for any length-constrained elements. Rules: - Plain, specific language; no jargon, no filler, no exclamation marks unless the voice guide allows them. - Accessibility: write for screen readers — never rely on "click here" or directional cues like "the button on the right"; make link and button text describe the action. - Error messages: say what happened and what to do next; never blame the user. - Do not reference features, limits, or data that are not in the context I gave you. Where a specific number, name, or limit is needed, insert [PRODUCT: confirm] rather than guessing. Keep all three in the same voice but genuinely different in wording so I can compare approaches.
Fill in your details and the prompt updates live — then copy.
**Empty state — no projects yet** Option A — "Nothing here yet. Create your first project to get started." (48 chars) Rationale: neutral, action-first, matches the calm voice. Option B — "Your projects will show up here. Ready to start one?" (52 chars) Rationale: sets an expectation, softer invitation. Option C — "No projects yet. Start one with [PRODUCT: confirm button label]." Rationale: shortest; defers to the real button text instead of guessing it. All three avoid "click here" and describe the action for screen-reader users.
The full workflow
- Paste your voice-and-tone guide and the component's context.
- Generate three options and compare them against the guide.
- Replace every [PRODUCT: confirm] placeholder with the real value yourself.
- Run the winning copy past an accessibility check (contrast, screen-reader wording).
- Add it to the design file and your content source of truth.
Watch out for
Accessibility applies to copy too: WCAG covers plain language and link/button text, so drop 'click here' and directional cues and keep the reading level appropriate. AI does not guarantee any of this — you do.
Do not paste unreleased product names, features, or roadmap details into consumer AI tools; NDA and confidentiality obligations cover designs as much as documents.
Lean on AI too hard and every product starts to sound the same — protect your product's specific voice by editing, not just accepting.
Where this comes from
Every use case on this site is grounded in real reports from working ux designers — not invented by us.