Loading skill
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/dokhacgiakhoa-signup-flow-cro/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-13 | ✗→✓ | ▲ Improved | 22% | 0% |
| case-22 | ✗→✓ | ▲ Improved | 39% | 0% |
| case-25 | ✗→✓ | ▲ Improved | 11% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | 44% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | 22% | 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.
Before providing recommendations, understand:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 26,466 | 22,932 | -13% | 1 | 1 | 0% | 3,752 | 3,982 | +6% | 0 | 0 | — |
case-02 | fail→fail | 21,799 | 19,241 | -12% | 1 | 1 | 0% | 2,881 | 4,243 | +47% | 0 | 0 | — |
case-03 | pass→pass | 18,645 | 19,248 | +3% | 1 | 1 | 0% | 2,309 | 3,331 | +44% | 0 | 0 | — |
case-04 | pass→pass | 26,035 | 26,038 | +0% | 1 | 1 | 0% | 3,723 | 4,528 | +22% | 0 | 0 | — |
case-05 | pass→pass | 23,895 | 24,920 | +4% | 1 | 1 | 0% | 3,069 | 4,300 | +40% | 0 | 0 | — |
case-06 | pass→pass | 12,263 | 12,599 | +3% | 1 | 1 | 0% | 2,090 | 3,158 | +51% | 0 | 0 | — |
case-07 | pass→pass | 12,527 | 16,306 | +30% | 1 | 1 | 0% | 1,981 | 2,752 | +39% | 0 | 0 | — |
case-08 | pass→pass | 17,306 | 15,844 | -8% | 1 | 1 | 0% | 1,994 | 2,655 | +33% | 0 | 0 | — |
case-09 | pass→pass | 12,977 | 15,439 | +19% | 1 | 1 | 0% | 2,202 | 2,681 | +22% | 0 | 0 | — |
case-10 | pass→pass | 16,740 | 10,937 | -35% | 1 | 1 | 0% | 2,019 | 2,766 | +37% | 0 | 0 | — |
case-11 | pass→pass | 12,041 | 12,848 | +7% | 1 | 1 | 0% | 1,959 | 3,046 | +55% | 0 | 0 | — |
case-12 | pass→pass | 10,268 | 17,087 | +66% | 1 | 1 | 0% | 1,700 | 2,914 | +71% | 0 | 0 | — |
case-13 | fail→pass | 13,237 | 10,567 | -20% | 1 | 1 | 0% | 2,192 | 2,668 | +22% | 0 | 0 | — |
case-14 | pass→pass | 18,353 | 19,309 | +5% | 1 | 1 | 0% | 2,075 | 3,077 | +48% | 0 | 0 | — |
case-15 | pass→pass | 17,050 | 15,194 | -11% | 1 | 1 | 0% | 1,897 | 2,677 | +41% | 0 | 0 | — |
case-16 | pass→pass | 12,793 | 15,029 | +17% | 1 | 1 | 0% | 2,050 | 2,572 | +25% | 0 | 0 | — |
case-17 | pass→pass | 13,297 | 11,796 | -11% | 1 | 1 | 0% | 2,071 | 2,782 | +34% | 0 | 0 | — |
case-18 | pass→pass | 10,505 | 13,859 | +32% | 1 | 1 | 0% | 1,623 | 2,337 | +44% | 0 | 0 | — |
case-19 | pass→pass | 18,971 | 20,925 | +10% | 1 | 1 | 0% | 2,267 | 3,434 | +51% | 0 | 0 | — |
case-20 | pass→pass | 15,583 | 12,178 | -22% | 1 | 1 | 0% | 1,623 | 2,925 | +80% | 0 | 0 | — |
case-21 | pass→pass | 18,299 | 18,187 | -1% | 1 | 1 | 0% | 2,166 | 2,992 | +38% | 0 | 0 | — |
case-22 | fail→pass | 19,163 | 13,122 | -32% | 1 | 1 | 0% | 2,224 | 3,094 | +39% | 0 | 0 | — |
case-23 | pass→pass | 16,989 | 19,640 | +16% | 1 | 1 | 0% | 1,928 | 3,387 | +76% | 0 | 0 | — |
case-24 | pass→pass | 15,812 | 13,789 | -13% | 1 | 1 | 0% | 1,734 | 2,403 | +39% | 0 | 0 | — |
case-25 | fail→pass | 20,430 | 15,598 | -24% | 1 | 1 | 0% | 2,453 | 2,727 | +11% | 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. 25 cases were attempted. The headline lift of +12 percentage points is the difference between those two pass rates over the 25 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.