build-personas

Builds evidence-tagged personas and Jobs-to-be-Done statements from synthesized research data.

1|Updated Jul 13, 2026
One-click install
npx skills add https://github.com/dineshrevunuru/SuperSkills --skill build-personas-dineshrevunuru
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: build-personas
Source: https://github.com/dineshrevunuru/SuperSkills/tree/main/build-personas
Command: npx skills add https://github.com/dineshrevunuru/SuperSkills --skill build-personas-dineshrevunuru

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Teams routinely build personas that overstate their evidence, invent demographic filler, or get laminated and ignored. This Skill turns synthesized research into honest user models that state exactly what they can and cannot claim, so design decisions rest on real evidence instead of assumption laundering. ## Core Features & Use Cases - Persona type selection: Decision tree for choosing proto, qualitative, or statistical personas based on available data, with an honesty table stating what each type can and cannot claim. - Behavior-based clustering: Clusters synthesized attributes into 2-4 personas by shared goals and behaviors, merging or cutting clusters that drive no distinct design decisions. - JTBD writing: Writes Jobs-to-be-Done statements with situation, outcome, and functional plus emotional success criteria, keeping solutions and feature words out. - Use Case: After synthesizing 9 user interviews into themes, use this Skill to cluster the themes into two evidence-tagged persona cards and three JTBD statements, each stamped with its type and limitations. ## Quick Start Build personas and JTBD statements from my synthesized interview themes, and state what each persona type cannot claim.

Frequently Asked Questions about build-personas

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build personas from user interview data?

First synthesize raw interviews into themes and attributes, then cluster attributes that co-occur in the same participants by shared behavior and goal. Merge clusters that drive identical design decisions and stop at 2-4 personas, tagging every pain point with participant IDs.

Should I use personas or Jobs-to-be-Done for my project?

Use personas when multiple genuinely different user groups have conflicting needs or when prioritization fights arise. Use JTBD when one user type has many contexts and desired outcomes, or when the team fixates on demographics. They combine cleanly when both segments and outcomes matter.

What is the difference between proto, qualitative, and statistical personas?

Proto personas come from team assumptions in a 2-4 hour workshop and must be stamped as unvalidated. Qualitative personas come from 5-30 interviews and are the recommended default. Statistical personas require a survey of 100-500+ respondents with cluster analysis and can claim segment sizes.

Can qualitative personas claim what percentage of users they represent?

No. Qualitative personas built from 5-30 interviews cannot claim segment sizes or statistical representativeness. To state that X% of users match a persona, you need the statistical tier with a survey of 100-500+ respondents and cluster analysis.

When should I not use this persona-building approach?

Do not use it on raw transcripts or notes; run research synthesis first to produce coded themes. It also does not cover journey maps or empathy maps, survey instrument design for the statistical tier, or planning validation research for proto personas.