Install any skill in seconds. Free to start, no credit card required.
Get Started Free →When the user wants to optimize signup, registration, account creation, or trial activation flows. Also use when the user mentions "signup conversions," "registration friction," "signup form optimization," "free trial signup," "reduce signup dropoff," or "account creation flow." For post-signup onboarding, see onboarding-cro. For lead capture forms (not account creation), see form-cro.
.claude/skills/alirezarezvani-signup-flow-cro/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 92% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 145% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 98% | 0% |
| case-11 | ✓→✗ | ▼ Worse | 156% | 0% |
| case-06 | ✓→✓ | = Same ✓ | 70% | 0% |
You are an expert in optimizing signup and registration flows. Your goal is to reduce friction, increase completion rates, and set users up for successful activation.
Check for product marketing context first: If .claude/product-marketing-context.md exists, read it before asking questions. Use that context and only ask for information not already covered or specific to this task.
Before providing recommendations, understand:
→ See references/signup-cro-playbook.md for details
| Tool | Invocation | Output | |---|---|---| | Funnel drop analyzer | python3 scripts/funnel_drop_analyzer.py --steps funnel.json (or --stdin; --json for pipelines; no arg = embedded demo) | Per-step drop-off %, the worst step named, and severity ranking |
Feed it the step-by-step user counts (landing → form start → form complete → verify → done). The named worst step is where the audit starts; quantify each finding's Impact with its drop-off number.
For each issue found:
Organized by:
Layout & Structure
Field Optimization
Authentication Options
Visual Design
Headlines & CTAs
Microcopy
Trust Elements
Free Trial Variations
Friction Points
.claude/product-marketing-context.md for B2B vs. B2C context, compliance requirements, and qualification data needs before designing the field set. WHEN NOT: skip if user has provided explicit product and compliance context in the conversation.All signup flow CRO output follows this quality standard:
Automatically surface signup-flow-cro when:
| Artifact | Format | Description | |----------|--------|-------------| | Signup Flow Audit | Issue/Impact/Fix/Priority table | Per-step and per-field analysis with severity ratings | | Recommended Field Set | Justified list | Required vs. deferrable fields with rationale, organized by signup step | | Flow Redesign Spec | Step-by-step outline | Recommended multi-step or single-step flow with copy for each screen | | SSO & Auth Options Recommendation | Decision table | Which auth methods to offer, placement, and priority for the target audience | | A/B Test Hypotheses | Table | Hypothesis × variant description × success metric × priority for top 3-5 tests |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-06 | pass→pass | 11,212 | 7,732 | -31% | 1 | 1 | 0% | 2,040 | 3,471 | +70% | 0 | 0 | — |
case-07 | fail→pass | 12,366 | 11,621 | -6% | 1 | 1 | 0% | 2,094 | 4,014 | +92% | 0 | 0 | — |
case-01 | fail→fail | 14,518 | 19,497 | +34% | 1 | 1 | 0% | 2,906 | 6,027 | +107% | 0 | 0 | — |
case-08 | pass→pass | 10,524 | 9,543 | -9% | 1 | 1 | 0% | 1,775 | 3,672 | +107% | 0 | 0 | — |
case-09 | pass→pass | 11,837 | 10,128 | -14% | 1 | 1 | 0% | 1,995 | 3,721 | +87% | 0 | 0 | — |
case-10 | pass→pass | 17,125 | 13,721 | -20% | 1 | 1 | 0% | 2,819 | 4,480 | +59% | 0 | 0 | — |
case-11 | pass→fail | 7,939 | 7,773 | -2% | 1 | 1 | 0% | 1,322 | 3,384 | +156% | 0 | 0 | — |
case-12 | fail→pass | 8,157 | 10,630 | +30% | 1 | 1 | 0% | 1,591 | 3,898 | +145% | 0 | 0 | — |
case-02 | fail→pass | 6,129 | 2,135 | -65% | 1 | 1 | 0% | 1,281 | 2,541 | +98% | 0 | 0 | — |
case-03 | pass→pass | 11,146 | 10,117 | -9% | 1 | 1 | 0% | 1,829 | 3,814 | +109% | 0 | 0 | — |
case-04 | pass→pass | 11,634 | 11,208 | -4% | 1 | 1 | 0% | 2,022 | 3,943 | +95% | 0 | 0 | — |
case-05 | pass→pass | 12,441 | 11,073 | -11% | 1 | 1 | 0% | 2,142 | 4,015 | +87% | 0 | 0 | — |
case-21 | pass→pass | 17,624 | 12,704 | -28% | 1 | 1 | 0% | 2,509 | 4,198 | +67% | 0 | 0 | — |
case-22 | pass→pass | 15,225 | 14,483 | -5% | 1 | 1 | 0% | 2,515 | 4,418 | +76% | 0 | 0 | — |
case-13 | pass→pass | 14,426 | 15,832 | +10% | 1 | 1 | 0% | 2,376 | 4,810 | +102% | 0 | 0 | — |
case-14 | pass→pass | 11,961 | 13,543 | +13% | 1 | 1 | 0% | 2,004 | 4,372 | +118% | 0 | 0 | — |
case-15 | pass→pass | 14,542 | 13,097 | -10% | 1 | 1 | 0% | 2,518 | 4,349 | +73% | 0 | 0 | — |
case-16 | fail→fail | 12,955 | 7,231 | -44% | 1 | 1 | 0% | 2,088 | 3,257 | +56% | 0 | 0 | — |
case-17 | pass→pass | 20,978 | 16,314 | -22% | 1 | 1 | 0% | 3,967 | 4,723 | +19% | 0 | 0 | — |
case-18 | pass→pass | 11,709 | 10,592 | -10% | 1 | 1 | 0% | 2,050 | 3,733 | +82% | 0 | 0 | — |
case-19 | fail→fail | 13,084 | 10,987 | -16% | 1 | 1 | 0% | 2,147 | 4,002 | +86% | 0 | 0 | — |
case-20 | pass→pass | 15,085 | 17,882 | +19% | 1 | 1 | 0% | 2,438 | 5,033 | +106% | 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 +9 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.