Prompt UX Patterns for Making Decisions Faster

For those who spend a lot of time working with AI to get to an answer, here are some Prompt UX tips. I thought to share as I spent an inordinate amount of time going back-and-forth with AI to arrive at solutions.

Sometimes it's just a matter of tweaking a prompt to make it easy for you to make decisions for AI. Here are my three favorites.

1. Suggested Answers Pattern

When you have AI ask clarifying questions, make it suggest the answers too.

Don’t:

Ask me clarifying questions to understand the requirements better.

Do:

Ask me clarifying questions and suggest answers based on the context.

Example:
- Q: Who is the main user of this product?
- Suggested Answer: [INFERRED_USER_TYPE]
- [Confirm or correct this assumption]

This is human-in-the-loop at work. Not constant back-and-forth.

2. Pre-Filled Assumptions Pattern

Make AI state assumptions and let you course-correct when things are incorrect.

Do:

Proceed with assumptions based on the context. For example:

"I’m proceeding with these assumptions based on the context provided:
- Budget: [INFERRED_FROM_CONTEXT]
- Objective: [INFERRED_FROM_CONTEXT]  

Please correct wrong assumptions, otherwise I will continue."

3. Progressive Disclosure Pattern

Start with 80%, then allow opting into advanced requirements.

Do:

Please progressively disclose my options. Start with simple and 
progressively introduce advanced options. For example:

"I’ll create a basic diet plan.

Reply 'advanced' if you also need: vegetarian, vegan, specific nutrition requirements."

Contrived example but hope that makes sense.

Hope this helps!

Leave a Reply