Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Identify the Ideal Customer Profile (ICP) from research data with demographics, behaviors, JTBD, and needs. Use when defining your ICP, analyzing PMF survey data, or understanding who your best customers are.
.claude/skills/phuryn-ideal-customer-profile/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-13 | ✗→✓ | ▲ Improved | 270% | 0% |
| case-05 | ✓→✓ | ＝ Same ✓ | 43% | 0% |
| case-06 | ✓→✓ | ＝ Same ✓ | 69% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | 5% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | 69% | 0% |
Identify your Ideal Customer Profile (ICP) from research and survey data. This skill synthesizes customer research to define the customer most likely to find value, retain, and expand with your product.
Who are they from a firmographic and personal perspective?
How do they work and make decisions?
What are they trying to accomplish?
What problems does your product solve?
Collect research about actual and potential customers:
Identify customer cohorts and their value:
Extract firmographic patterns:
Map decision-making and adoption patterns:
Articulate what they're trying to accomplish:
Synthesize specific problem areas:
Use $ARGUMENTS to pass:
A comprehensive ICP definition including:
Based on Jobs to Be Done theory by Clayton Christensen and customer profiling methodology. Combines behavioral data with motivational insights to define actionable customer profiles.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-05 | pass→pass | 19,425 | 18,021 | -7% | 1 | 1 | 0% | 3,008 | 4,292 | +43% | 0 | 0 | — |
case-06 | pass→pass | 18,402 | 22,592 | +23% | 1 | 1 | 0% | 3,050 | 5,155 | +69% | 0 | 0 | — |
case-01 | pass→pass | 38,040 | 29,205 | -23% | 1 | 1 | 0% | 6,280 | 6,566 | +5% | 0 | 0 | — |
case-02 | pass→pass | 17,583 | 28,150 | +60% | 1 | 1 | 0% | 2,897 | 4,902 | +69% | 0 | 0 | — |
case-03 | pass→pass | 16,547 | 32,523 | +97% | 1 | 1 | 0% | 2,666 | 5,348 | +101% | 0 | 0 | — |
case-04 | pass→pass | 23,946 | 19,574 | -18% | 1 | 1 | 0% | 3,308 | 4,923 | +49% | 0 | 0 | — |
case-07 | pass→pass | 19,012 | 23,941 | +26% | 1 | 1 | 0% | 3,210 | 5,034 | +57% | 0 | 0 | — |
case-08 | pass→pass | 17,012 | 28,532 | +68% | 1 | 1 | 0% | 2,711 | 5,780 | +113% | 0 | 0 | — |
case-09 | pass→pass | 18,500 | 23,395 | +26% | 1 | 1 | 0% | 2,858 | 4,828 | +69% | 0 | 0 | — |
case-10 | pass→pass | 13,482 | 17,013 | +26% | 1 | 1 | 0% | 2,128 | 4,012 | +89% | 0 | 0 | — |
case-11 | pass→pass | 13,143 | 22,601 | +72% | 1 | 1 | 0% | 2,022 | 4,834 | +139% | 0 | 0 | — |
case-12 | pass→pass | 12,146 | 25,108 | +107% | 1 | 1 | 0% | 2,176 | 5,407 | +148% | 0 | 0 | — |
case-13 | fail→pass | 8,731 | 23,673 | +171% | 1 | 1 | 0% | 1,337 | 4,949 | +270% | 0 | 0 | — |
case-14 | pass→pass | 19,890 | 28,879 | +45% | 1 | 1 | 0% | 3,283 | 6,124 | +87% | 0 | 0 | — |
case-15 | pass→pass | 21,160 | 23,881 | +13% | 1 | 1 | 0% | 2,887 | 5,186 | +80% | 0 | 0 | — |
case-16 | pass→pass | 19,983 | 33,987 | +70% | 1 | 1 | 0% | 3,003 | 5,403 | +80% | 0 | 0 | — |
case-17 | pass→pass | 13,141 | 24,682 | +88% | 1 | 1 | 0% | 1,655 | 4,421 | +167% | 0 | 0 | — |
case-18 | pass→pass | 17,841 | 22,213 | +25% | 1 | 1 | 0% | 2,966 | 4,833 | +63% | 0 | 0 | — |
case-19 | pass→pass | 13,407 | 19,028 | +42% | 1 | 1 | 0% | 2,047 | 4,427 | +116% | 0 | 0 | — |
case-20 | pass→pass | 26,936 | 33,468 | +24% | 1 | 1 | 0% | 3,371 | 5,257 | +56% | 0 | 0 | — |
case-21 | pass→pass | 13,512 | 11,159 | -17% | 1 | 1 | 0% | 2,380 | 3,141 | +32% | 0 | 0 | — |
case-22 | pass→pass | 16,299 | 21,044 | +29% | 1 | 1 | 0% | 3,567 | 5,794 | +62% | 0 | 0 | — |
case-23 | pass→pass | 19,582 | 21,723 | +11% | 1 | 1 | 0% | 3,475 | 4,906 | +41% | 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. 23 cases were attempted. The headline lift of +4 percentage points is the difference between those two pass rates over the 23 comparable cases.
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.