As of 13 August 2026, AI can create a UX research plan for your business.

What to actually do

1. Use a tool built for this

   The route this page recommends

2. Do it yourself

   Second choice

   A chat interface, chat-fluent skill, and roughly 30 minutes until you can act on the result.

   How to actually do it

   1. Open a document and write the business, product, target users, decision to be made, known uncertainty, existing evidence and practical constraints.
   2. Paste those details into the prompt, replacing every bracketed slot, and ask the chatbot to create the plan.
   3. Answer any essential follow-up questions using information from your product records, customer support material, analytics or conversations with your team.
   4. Compare the proposed objectives and research questions with the decision you actually need to make, removing questions that would not change an action.
   5. Check the method, participant criteria, consent wording, privacy steps and accessibility steps against your recruitment process and current data-handling arrangements.
   6. Ask a colleague who understands the product or users to challenge the assumptions, then revise the plan and agree who will recruit, run sessions, analyse evidence and approve the final recommendations.

   Prompt

   Create a practical UX research plan for my business using the information below.
   
   Business and product: [describe the business, product or service]
   Decision this research must support: [state the decision]
   Research problem or uncertainty: [state what you do not know]
   Target users: [describe who they are, including relevant UK context]
   What users currently do: [describe the current behaviour or alternative]
   Existing evidence: [paste analytics findings, support themes, interview notes or other evidence]
   Research constraints: [time, budget, access to participants, team capacity and tools]
   Privacy or accessibility requirements: [state any requirements]
   
   Produce:
   1. A short research brief and the decision it will inform.
   2. Three to five prioritised research objectives.
   3. Neutral research questions linked to each objective. Do not write leading questions.
   4. The recommended method or combination of methods, with a reason for each choice and a note on what it cannot answer.
   5. A participant profile, inclusion criteria, exclusion criteria and a realistic recruitment approach.
   6. A sample-size rationale without inventing statistical confidence or certainty.
   7. A discussion guide or test script with timings, opening consent wording, tasks and follow-up questions.
   8. A plan for recording, organising and analysing the evidence.
   9. A synthesis framework showing how observations become findings and findings become recommendations.
   10. Risks, biases, ethical and privacy considerations, accessibility considerations and mitigation steps.
   11. A timeline, responsibilities, deliverables and definition of done.
   
   Separate facts taken from my input from your assumptions. Mark every assumption clearly. If information is missing, ask up to five essential questions first. Do not invent users, evidence, recruitment numbers, research findings or business constraints. Keep the plan proportionate to the decision and suitable for a UK business. End with a short checklist of decisions I must approve before research starts.

   Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.

3. Hand it to a person

   The distant third

   A person who owns the outcome does this end to end, worth it when the failure is dear.