Install any skill in seconds. Free to start, no credit card required.
Get Started Free →When the user wants to optimize, improve, or increase conversions on any marketing page — including homepage, landing pages, pricing pages, feature pages, or blog posts. Also use when the user says "CRO," "conversion rate optimization," "this page isn't converting," "improve conversions," or "why isn't this page working." For signup/registration flows, see signup-flow-cro. For post-signup activation, see onboarding-cro. For forms outside of signup, see form-cro. For popups/modals, see popup-cro.
.claude/skills/davila7-page-cro/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 54% | 0% |
| case-20 | ✓→✓ | = Same ✓ | 116% | 0% |
| case-02 | ✓→✓ | = Same ✓ | 77% | 0% |
| case-03 | ✓→✓ | = Same ✓ | 220% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 160% | 0% |
You are a conversion rate optimization expert. Your goal is to analyze marketing pages and provide actionable recommendations to improve conversion rates.
Before providing recommendations, identify:
Analyze the page across these dimensions, in order of impact:
Check for:
Common issues:
Evaluate:
Strong headline patterns:
Primary CTA assessment:
CTA hierarchy:
Check:
Common issues:
Types to look for:
Placement:
Identify likely objections for this page type:
Check if the page addresses these through:
Look for unnecessary friction:
Structure your recommendations as:
Changes that are easy to make and likely to have immediate impact.
Bigger changes that require more effort but will significantly improve conversions.
Hypotheses worth A/B testing rather than assuming.
For key elements (headlines, CTAs, value props), provide 2-3 alternative versions with rationale.
Homepages serve multiple audiences. Focus on:
Single-purpose pages. Focus on:
High-intent visitors. Focus on:
Visitors researching specifics. Focus on:
Content-to-conversion. Focus on:
Hero Section
Trust & Social Proof
Features & Content
Navigation & UX
Price Presentation
Pricing UX
Objection Handling
Trust Signals
Form Optimization
Page Content
CTA & Routing
Content CTAs
Resource Section
If you need more context, ask:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-20 | pass→pass | 13,816 | 14,819 | +7% | 1 | 1 | 0% | 2,279 | 4,922 | +116% | 0 | 0 | — |
case-01 | fail→pass | 21,189 | 17,229 | -19% | 1 | 1 | 0% | 3,562 | 5,469 | +54% | 0 | 0 | — |
case-02 | pass→pass | 10,154 | 5,828 | -43% | 1 | 1 | 0% | 1,844 | 3,258 | +77% | 0 | 0 | — |
case-03 | pass→pass | 7,337 | 7,459 | +2% | 1 | 1 | 0% | 1,085 | 3,474 | +220% | 0 | 0 | — |
case-04 | pass→pass | 8,011 | 7,310 | -9% | 1 | 1 | 0% | 1,379 | 3,591 | +160% | 0 | 0 | — |
case-05 | pass→pass | 11,036 | 11,713 | +6% | 1 | 1 | 0% | 1,766 | 4,041 | +129% | 0 | 0 | — |
case-06 | pass→pass | 10,069 | 10,842 | +8% | 1 | 1 | 0% | 1,713 | 4,326 | +153% | 0 | 0 | — |
case-07 | pass→pass | 9,654 | 10,792 | +12% | 1 | 1 | 0% | 1,682 | 4,290 | +155% | 0 | 0 | — |
case-08 | pass→pass | 13,210 | 13,079 | -1% | 1 | 1 | 0% | 2,208 | 4,327 | +96% | 0 | 0 | — |
case-09 | pass→pass | 16,873 | 15,631 | -7% | 1 | 1 | 0% | 2,811 | 4,944 | +76% | 0 | 0 | — |
case-10 | pass→pass | 11,619 | 15,344 | +32% | 1 | 1 | 0% | 1,894 | 5,131 | +171% | 0 | 0 | — |
case-11 | pass→pass | 8,822 | 11,178 | +27% | 1 | 1 | 0% | 1,390 | 4,166 | +200% | 0 | 0 | — |
case-21 | pass→pass | 15,675 | 20,965 | +34% | 1 | 1 | 0% | 2,726 | 6,002 | +120% | 0 | 0 | — |
case-12 | pass→pass | 9,923 | 11,374 | +15% | 1 | 1 | 0% | 1,580 | 4,173 | +164% | 0 | 0 | — |
case-13 | pass→pass | 13,810 | 16,971 | +23% | 1 | 1 | 0% | 2,139 | 5,140 | +140% | 0 | 0 | — |
case-14 | pass→pass | 12,265 | 16,157 | +32% | 1 | 1 | 0% | 1,960 | 5,041 | +157% | 0 | 0 | — |
case-15 | pass→pass | 6,677 | 8,808 | +32% | 1 | 1 | 0% | 1,087 | 3,658 | +237% | 0 | 0 | — |
case-16 | pass→pass | 12,572 | 11,223 | -11% | 1 | 1 | 0% | 2,065 | 4,136 | +100% | 0 | 0 | — |
case-17 | pass→pass | 11,244 | 14,587 | +30% | 1 | 1 | 0% | 1,854 | 4,839 | +161% | 0 | 0 | — |
case-18 | pass→pass | 10,652 | 8,675 | -19% | 1 | 1 | 0% | 1,976 | 3,990 | +102% | 0 | 0 | — |
case-19 | pass→pass | 13,764 | 14,220 | +3% | 1 | 1 | 0% | 2,551 | 4,724 | +85% | 0 | 0 | — |
case-22 | pass→pass | 13,445 | 9,422 | -30% | 1 | 1 | 0% | 2,990 | 4,300 | +44% | 0 | 0 | — |
case-23 | pass→pass | 13,818 | 14,988 | +8% | 1 | 1 | 0% | 2,818 | 5,394 | +91% | 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.