Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Communicate design's contribution to business and user outcomes in terms that resonate with stakeholders.
.claude/skills/owl-listener-design-impact-reporting/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 48% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 83% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 50% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 26% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 47% | 0% |
You are an expert in measuring and communicating the value of design work to leadership, cross-functional partners, and the broader organization.
You build the evidence and narrative that connects design decisions to measurable outcomes — so design is treated as a strategic investment, not a cost center or aesthetic layer.
Design impact is often diffuse, lagged, and shared with other functions. A better onboarding flow increases conversion — but so does a marketing campaign and a pricing change that launched the same quarter. Design impact reporting requires:
Connect design work to three levels:
What users do as a result of the design:
What the product achieves:
What the business achieves:
A recurring (quarterly) snapshot of key metrics across active design work:
For significant shipped work:
When controlled experiments are available:
For leadership and headcount conversations:
Quantitative metrics alone are incomplete. Pair them with:
Every impact report needs:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 14,073 | 16,702 | +19% | 1 | 1 | 0% | 2,634 | 3,911 | +48% | 0 | 0 | — |
case-02 | pass→pass | 13,511 | 12,405 | -8% | 1 | 1 | 0% | 2,719 | 3,673 | +35% | 0 | 0 | — |
case-03 | pass→pass | 16,529 | 17,510 | +6% | 1 | 1 | 0% | 2,620 | 3,996 | +53% | 0 | 0 | — |
case-04 | pass→pass | 18,112 | 15,393 | -15% | 1 | 1 | 0% | 3,699 | 3,906 | +6% | 0 | 0 | — |
case-05 | fail→pass | 10,526 | 14,127 | +34% | 1 | 1 | 0% | 1,864 | 3,411 | +83% | 0 | 0 | — |
case-06 | fail→pass | 14,037 | 13,882 | -1% | 1 | 1 | 0% | 2,192 | 3,282 | +50% | 0 | 0 | — |
case-07 | pass→pass | 13,573 | 14,519 | +7% | 1 | 1 | 0% | 2,233 | 3,184 | +43% | 0 | 0 | — |
case-08 | pass→pass | 7,950 | 10,524 | +32% | 1 | 1 | 0% | 1,406 | 2,724 | +94% | 0 | 0 | — |
case-09 | pass→pass | 11,918 | 12,413 | +4% | 1 | 1 | 0% | 1,792 | 2,876 | +60% | 0 | 0 | — |
case-10 | fail→pass | 14,916 | 11,730 | -21% | 1 | 1 | 0% | 2,387 | 3,003 | +26% | 0 | 0 | — |
case-11 | fail→fail | 12,586 | 17,745 | +41% | 1 | 1 | 0% | 2,159 | 3,624 | +68% | 0 | 0 | — |
case-12 | fail→pass | 14,468 | 14,375 | -1% | 1 | 1 | 0% | 2,311 | 3,403 | +47% | 0 | 0 | — |
case-13 | pass→pass | 13,706 | 12,593 | -8% | 1 | 1 | 0% | 2,214 | 3,066 | +38% | 0 | 0 | — |
case-14 | pass→pass | 7,181 | 3,501 | -51% | 1 | 1 | 0% | 1,364 | 1,548 | +13% | 0 | 0 | — |
case-15 | fail→fail | 14,119 | 16,829 | +19% | 1 | 1 | 0% | 2,223 | 3,523 | +58% | 0 | 0 | — |
case-16 | fail→fail | 14,608 | 12,688 | -13% | 1 | 1 | 0% | 2,313 | 2,930 | +27% | 0 | 0 | — |
case-17 | fail→fail | 13,855 | 14,125 | +2% | 1 | 1 | 0% | 2,418 | 3,227 | +33% | 0 | 0 | — |
case-18 | pass→pass | 16,569 | 17,028 | +3% | 1 | 1 | 0% | 2,601 | 3,860 | +48% | 0 | 0 | — |
case-19 | pass→pass | 10,985 | 11,607 | +6% | 1 | 1 | 0% | 1,932 | 2,978 | +54% | 0 | 0 | — |
case-20 | fail→pass | 15,189 | 16,071 | +6% | 1 | 1 | 0% | 2,563 | 3,444 | +34% | 0 | 0 | — |
case-21 | pass→pass | 11,605 | 14,389 | +24% | 1 | 1 | 0% | 1,904 | 3,201 | +68% | 0 | 0 | — |
case-22 | pass→pass | 13,771 | 14,068 | +2% | 1 | 1 | 0% | 2,216 | 3,147 | +42% | 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 +27 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.