ThinkWithPrompt by SynotticThinkWithPrompt

Plan a card sort or tree test to fix my navigation

Sets up an information architecture study, including the method, cards or tree, tasks, participants, and how to read the results.

Product & UX DesignFew-ShotStandardRewrite & EditPlanFill-in Template

The prompt

Help me plan an information architecture study to improve my product's or site's navigation. What I am organizing: [e.g. app settings, help center, ecommerce categories, intranet] The problem: [e.g. "people can't find billing", "menu has grown to 14 items"] Current structure: [PASTE THE MENU OR SITEMAP] Content items: [LIST OR PASTE THE PAGES/FEATURES] Users: [WHO, AND HOW MANY I CAN RECRUIT] Tools I have: [e.g. an online card sorting tool, spreadsheets, in-person sessions] Give me: 1. Which method fits and why: open card sort (discover how people group things), closed or hybrid sort (test my categories), or tree test (test whether people can find things in a proposed structure). Often the answer is a sort first, then a tree test. 2. For a card sort: 30-60 cards written in plain user language without repeating category words that bias grouping, plus instructions for participants. 3. For a tree test: the text-only tree, and 8-10 realistic find-it tasks phrased without using the label words, each with the correct answer location. 4. Participant numbers (common guidance: about 15-30 for a card sort, 50 or more for a tree test online) and screening criteria. 5. How to analyze results: similarity matrix and agreement for sorts; success, directness, and first-click for tree tests; what thresholds would make me change a label. 6. Pitfalls to avoid.

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

  • PASTE THE MENU OR SITEMAP
  • LIST OR PASTE THE PAGES/FEATURES
  • WHO, AND HOW MANY I CAN RECRUIT

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