Diamond Prompting in UX Work

There are two opposing styles of prompts for using AI in UX:

• Exploratory prompting that starts with few preconceptions, except for providing the AI with context about your current project. These prompts are meant to benefit from AI’s inherent ideation capabilities and broaden your thinking about the current problem. Examples include:
  
  • suggest 3 different user research plans with small, medium, and large budgets, employing a variety of research methods
  • list 20 different test tasks for a usability study of this design
• Detail-refining prompting that builds on user-curated top hits from the exploratory stage. Examples include:
  
  • write a detailed screener for users with persona X
  • write out a complete description of task ideas 3 and 5 and include a motivating scenario for each. (This could be considered an advanced version of accordion editing, except that the detail phase might be in a later session.)

The best results come from alternating these two styles of prompting.

UI design is not art. UX methods are not academics.

Here’s a schematic of the recommended process for alternating the two styles of prompt in UX projects:

Diamond prompting of AI in UX work: we first use explorative prompts to broaden our understanding of the problem at hand, after which we use detail-refining prompts to arrive at the specific solution. (Ideogram)

When using AI to define how we should proceed in a UX project, we should use diamond prompting at multiple levels:

• Strategic scope: These decisions are at the level of the entire design project, focusing on creating an overall UX project plan and budget. This scope can span several months or a few years.
• Tactical scope:  After defining the project plan, we move to considering the process level, determining steps to progress from one stage to the next. This scope can stretch over several weeks or a few months.
• Operational scope: Finally, we consider the activity level, defining specific actions for an individual activity. For example, what are the specific actions needed to run a usability study in a few days? The scope of this level will usually only be several days or a few weeks. If an activity takes more than two weeks, it’s really a process and not an activity and should be kicked up to the tactical level for refinement.

Four interesting examples of that additional nuance coming from qualitative research:

• Senior staff were better than junior staff at creating prompts for the AI to help them in drafting these documents. (While you might have hoped that expensive senior staff would outperform cheaper junior staff, it’s not a given that senior people would actually be better than juniors at using a completely new technology. Their extra experience might have been useless.) The reason seniors were better is that the best results came from feeding the AI more detailed and descriptive context in the prompts.
• Junior staff very quickly improved their prompting skills. This echoes past findings that AI is a seniority accelerant.
• AI use changed minds: UX people who used to be skeptical about AI “discovered unforeseen benefits and capabilities, turning apprehension into endorsement” after they got hands-on experience with using AI for their jobs. (This is similar to the finding that the percentage of business professionals who were optimistic about the future of AI almost doubled with personal AI experience, whereas concerns about the potential negative implications of AI were almost cut in half for respondents with hands-on experience with AI use.)
• The best results came from using diamond prompting, as discussed above.