ThinkWithPrompt by SynotticThinkWithPrompt

Design permission requests that users actually accept

Plans when and how to ask for notifications, location, camera, or contacts access, with pre-prompts, copy, and fallbacks if users decline.

Product & UX DesignFew-ShotStandardAI & PromptingFill-in Template

The prompt

Help me design how my app asks users for permissions. App and platform: [WHAT IT DOES; iOS / Android / web] Permissions needed: [e.g. push notifications, location, camera, contacts, tracking] Why each is needed and what the user gets: [BENEFIT TO THE USER, NOT TO US] When we ask now: [e.g. all at launch] Current opt-in rates: [NUMBERS OR "unknown"] For each permission give: 1. The best moment to ask: in context, right when the user tries the feature that needs it, not at first launch unless the app cannot work without it. 2. Whether to use a custom pre-prompt screen before the system dialog, and its copy: headline, one-line benefit, primary and "Not now" buttons. Note that once a user denies the system dialog, the app often cannot show it again, so the pre-prompt protects that one chance. 3. The purpose string text for the system dialog (specific and honest). 4. What happens if they decline: a graceful fallback, how to re-ask later, and how to guide them to settings if needed. 5. A metric to track. Rules: never imply the app will not work when it will; do not use guilt or misleading button labels; follow current platform guidelines and privacy rules in my regions, and confirm against the latest store policies. Return a table plus the copy for each prompt.

Highlighted [brackets] are blanks to replace with your own details.

Fill in the blanks and use this promptMore Product & UX Design prompts

What to fill in

  • WHAT IT DOES; iOS / Android / web
  • e.g. push notifications, location, camera, contacts, tracking
  • BENEFIT TO THE USER, NOT TO US
  • e.g. all at launch
  • NUMBERS OR "unknown"

How this prompt is built

This prompt follows the Few-Shot structure (Few-shot prompting). A few-shot prompt includes one or more worked examples of the input and the output you want, so the model can copy the pattern instead of guessing it.

Related prompts

Find more prompts like this