ux-behavior-design

Diagnoses user inaction bottlenecks using the Fogg B=MAP behavior model.

Updated Jul 2, 2026
One-click install
npx skills add https://github.com/geekatron/jerry-claude-plugin --skill ux-behavior-design-geekatron
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ux-behavior-design
Source: https://github.com/geekatron/jerry-claude-plugin/tree/main/skills/ux-behavior-design
Command: npx skills add https://github.com/geekatron/jerry-claude-plugin --skill ux-behavior-design-geekatron

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Teams often see users failing to complete desired actions (checkout abandonment, incomplete onboarding, ignored CTAs) without knowing whether the cause is low motivation, excessive friction, or a missing prompt. This Skill applies BJ Fogg's B=MAP behavior model to systematically diagnose which behavioral factor falls below the action threshold, replacing guesswork with a structured elimination algorithm. ## Core Features & Use Cases - B=MAP Factor Assessment: Scores Motivation (three motivator pairs plus intrinsic/extrinsic/social dimensions), Ability (six Fogg simplicity factors with limiting-factor identification), and Prompt (Spark/Facilitator/Signal type, timing, placement) on evidence-backed 1-5 scales. - Bottleneck Elimination Algorithm: Executes a 4-step convergent diagnosis (prompt, ability, motivation, multiple) ordered by intervention difficulty, producing a severity-classified primary bottleneck with a full evidence trace. - Intervention Recommendations: Generates 3-5 prioritized interventions targeting the diagnosed bottleneck, classified by effort-to-impact ratio and marked with LOW synthesis confidence pending validation. - Use Case: A team sees 23% cart abandonment. The Skill maps B=MAP factors, identifies a missing above-fold CTA as a prompt bottleneck, and hands off the diagnosis to HEART metrics for measurement baselining. ## Quick Start Ask the /user-experience skill why users are not completing a specific action, such as "Diagnose the behavioral bottleneck in our checkout flow using B=MAP."

Frequently Asked Questions about ux-behavior-design

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

FAQPage Schema
How do I diagnose why users are not completing an action?

Define the target behavior in Fogg's format ("After [CONTEXT], I will [SPECIFIC BEHAVIOR]"), then assess Motivation, Ability, and Prompt factors. The 4-step elimination algorithm checks prompts first, then ability friction, then motivation, halting at the first factor below threshold.

What is the Fogg B=MAP behavior model?

B=MAP states that behavior occurs when Motivation, Ability, and a Prompt converge above their thresholds simultaneously. It is a convergence model, not multiplication: high motivation cannot compensate for a missing prompt or excessive friction.

When should I use behavior design instead of heuristic evaluation?

Use heuristic evaluation first to identify usability issues in an interface, then behavior design to trace severe issues to behavioral root causes. Behavior design explains why users fail to act; heuristics evaluate interface compliance with usability principles.

Can B=MAP diagnosis work without analytics data?

Yes, it supports a Qualitative Assessment Mode using interface artifacts and user-provided descriptions. The output is explicitly marked as degraded mode, with severity estimated from factor scores and intervention recommendations flagged as requiring empirical validation.

What are the six Fogg simplicity factors?

The six simplicity factors are Time, Money, Physical Effort, Brain Cycles, Social Deviance, and Non-Routine. The lowest-scoring factor is the limiting factor, and only interventions targeting that factor raise overall ability above the action threshold.

What are the limitations of AI-based behavior diagnosis?

Factor ratings depend on the quality of evidence provided, and motivation is hardest to assess remotely since it involves internal psychological states. Bottleneck diagnoses carry MEDIUM confidence and interventions LOW confidence, requiring validation through user testing or A/B experiments.