Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Analyze and optimize individual pages for conversion performance. Use when the user wants to improve conversion rates, diagnose why a page is underperforming, or increase the effectiveness of marketing pages (homepage, landing pages, pricing, feature pages, or blog posts). This skill focuses on diagnosis, prioritization, and testable recommendations— not blind optimization.
.claude/skills/dokhacgiakhoa-page-cro/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 41% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 68% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 46% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 47% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 85% | 0% |
You are an expert in page-level conversion optimization. Your goal is to diagnose why a page is or is not converting, assess readiness for optimization, and provide prioritized, evidence-based recommendations. You do not guarantee conversion lifts. You do not recommend changes without explaining why they matter.
Before giving CRO advice, calculate the Page Conversion Readiness & Impact Index.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 22,418 | 31,062 | +39% | 1 | 1 | 0% | 3,596 | 5,066 | +41% | 0 | 0 | — |
case-02 | fail→fail | 8,917 | 10,264 | +15% | 1 | 1 | 0% | 1,456 | 2,301 | +58% | 0 | 0 | — |
case-03 | fail→pass | 20,960 | 28,854 | +38% | 1 | 1 | 0% | 3,343 | 5,612 | +68% | 0 | 0 | — |
case-04 | pass→fail | 18,839 | 24,383 | +29% | 1 | 1 | 0% | 3,971 | 5,065 | +28% | 0 | 0 | — |
case-05 | pass→fail | 20,443 | 13,295 | -35% | 1 | 1 | 0% | 3,388 | 2,825 | -17% | 0 | 0 | — |
case-06 | pass→fail | 17,985 | 30,126 | +68% | 1 | 1 | 0% | 2,870 | 5,673 | +98% | 0 | 0 | — |
case-07 | pass→pass | 14,945 | 20,226 | +35% | 1 | 1 | 0% | 2,159 | 3,882 | +80% | 0 | 0 | — |
case-08 | pass→pass | 20,385 | 21,780 | +7% | 1 | 1 | 0% | 3,418 | 4,278 | +25% | 0 | 0 | — |
case-09 | pass→pass | 20,672 | 26,163 | +27% | 1 | 1 | 0% | 3,294 | 5,001 | +52% | 0 | 0 | — |
case-10 | pass→pass | 19,023 | 18,988 | -0% | 1 | 1 | 0% | 3,117 | 3,695 | +19% | 0 | 0 | — |
case-11 | pass→pass | 17,688 | 24,354 | +38% | 1 | 1 | 0% | 2,959 | 4,596 | +55% | 0 | 0 | — |
case-12 | pass→pass | 16,188 | 16,918 | +5% | 1 | 1 | 0% | 2,655 | 3,518 | +33% | 0 | 0 | — |
case-13 | pass→pass | 14,361 | 19,280 | +34% | 1 | 1 | 0% | 2,381 | 4,081 | +71% | 0 | 0 | — |
case-14 | pass→pass | 22,538 | 20,673 | -8% | 1 | 1 | 0% | 3,474 | 4,050 | +17% | 0 | 0 | — |
case-15 | fail→pass | 18,805 | 25,546 | +36% | 1 | 1 | 0% | 2,979 | 4,364 | +46% | 0 | 0 | — |
case-16 | fail→pass | 18,202 | 21,668 | +19% | 1 | 1 | 0% | 2,862 | 4,205 | +47% | 0 | 0 | — |
case-17 | pass→pass | 17,329 | 22,406 | +29% | 1 | 1 | 0% | 2,704 | 4,426 | +64% | 0 | 0 | — |
case-18 | fail→pass | 18,691 | 29,643 | +59% | 1 | 1 | 0% | 2,892 | 5,346 | +85% | 0 | 0 | — |
case-19 | fail→fail | 6,861 | 7,284 | +6% | 1 | 1 | 0% | 932 | 1,600 | +72% | 0 | 0 | — |
case-20 | pass→pass | 17,546 | 17,814 | +2% | 1 | 1 | 0% | 2,712 | 3,638 | +34% | 0 | 0 | — |
case-21 | fail→pass | 20,665 | 27,204 | +32% | 1 | 1 | 0% | 2,795 | 4,387 | +57% | 0 | 0 | — |
case-22 | pass→pass | 23,353 | 22,569 | -3% | 1 | 1 | 0% | 3,634 | 4,347 | +20% | 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 +14 percentage points is the difference between those two pass rates over the 22 comparable cases. 3 cases got worse with the skill loaded, and they are 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.