Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Segment users from feedback data based on behavior, JTBD, and needs. Identifies at least 3 distinct user segments. Use when segmenting a user base, analyzing diverse user feedback, or building a segmentation model.
.claude/skills/phuryn-user-segmentation/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 60% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 15% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 39% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 278% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 391% | 0% |
Analyze diverse user feedback to identify at least 3 distinct behavioral and needs-based user segments. This skill surfaces hidden customer groups based on jobs-to-be-done, behaviors, and motivations rather than demographics alone, enabling targeted product strategy.
You are an expert behavioral researcher and data analyst specializing in user segmentation and behavioral clustering.
Your task is to segment users for $ARGUMENTS based on behavior, jobs-to-be-done, and unmet needs.
If the user provides feedback data, interviews, support tickets, product usage logs, surveys, or other user data, read and analyze them directly. Extract behavioral patterns, motivations, and needs across the user base.
For each identified segment (minimum 3):
Segment Name & Overview
Behavioral Characteristics
Jobs-to-be-Done & Motivations
Key Needs & Pain Points
Current Product Fit
Differentiated Value Proposition
Segment Prioritization
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-05 | pass→pass | 14,360 | 12,399 | -14% | 1 | 1 | 0% | 2,262 | 2,793 | +23% | 0 | 0 | — |
case-01 | fail→pass | 17,404 | 33,223 | +91% | 1 | 1 | 0% | 3,163 | 5,048 | +60% | 0 | 0 | — |
case-02 | fail→pass | 26,048 | 26,457 | +2% | 1 | 1 | 0% | 4,490 | 5,150 | +15% | 0 | 0 | — |
case-03 | fail→pass | 20,171 | 28,147 | +40% | 1 | 1 | 0% | 3,471 | 4,839 | +39% | 0 | 0 | — |
case-04 | pass→pass | 20,367 | 17,607 | -14% | 1 | 1 | 0% | 3,388 | 4,315 | +27% | 0 | 0 | — |
case-06 | pass→fail | 14,320 | 20,056 | +40% | 1 | 1 | 0% | 2,635 | 4,414 | +68% | 0 | 0 | — |
case-07 | fail→pass | 5,864 | 18,301 | +212% | 1 | 1 | 0% | 1,034 | 3,909 | +278% | 0 | 0 | — |
case-08 | fail→pass | 6,114 | 25,620 | +319% | 1 | 1 | 0% | 981 | 4,815 | +391% | 0 | 0 | — |
case-09 | fail→pass | 8,532 | 31,359 | +268% | 1 | 1 | 0% | 1,152 | 4,719 | +310% | 0 | 0 | — |
case-10 | fail→pass | 18,119 | 19,981 | +10% | 1 | 1 | 0% | 3,120 | 4,138 | +33% | 0 | 0 | — |
case-11 | fail→pass | 5,773 | 25,059 | +334% | 1 | 1 | 0% | 750 | 5,138 | +585% | 0 | 0 | — |
case-12 | fail→pass | 15,593 | 26,819 | +72% | 1 | 1 | 0% | 2,620 | 5,225 | +99% | 0 | 0 | — |
case-13 | fail→pass | 6,575 | 20,863 | +217% | 1 | 1 | 0% | 1,060 | 4,354 | +311% | 0 | 0 | — |
case-14 | fail→pass | 5,872 | 23,797 | +305% | 1 | 1 | 0% | 848 | 4,727 | +457% | 0 | 0 | — |
case-15 | fail→pass | 16,117 | 31,057 | +93% | 1 | 1 | 0% | 2,700 | 4,832 | +79% | 0 | 0 | — |
case-16 | fail→pass | 8,194 | 18,227 | +122% | 1 | 1 | 0% | 1,042 | 4,021 | +286% | 0 | 0 | — |
case-17 | pass→pass | 18,798 | 24,442 | +30% | 1 | 1 | 0% | 2,369 | 5,123 | +116% | 0 | 0 | — |
case-18 | fail→pass | 6,734 | 26,837 | +299% | 1 | 1 | 0% | 1,074 | 4,894 | +356% | 0 | 0 | — |
case-19 | pass→pass | 17,083 | 28,703 | +68% | 1 | 1 | 0% | 2,225 | 4,920 | +121% | 0 | 0 | — |
case-20 | fail→pass | 4,883 | 24,276 | +397% | 1 | 1 | 0% | 779 | 5,089 | +553% | 0 | 0 | — |
case-21 | fail→pass | 4,669 | 34,090 | +630% | 1 | 1 | 0% | 774 | 6,966 | +800% | 0 | 0 | — |
case-22 | pass→pass | 15,122 | 26,222 | +73% | 1 | 1 | 0% | 2,658 | 5,108 | +92% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +68 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.