From artibot-cowork
Optimizes forms for lead generation, signups, and data collection with friction reduction, field optimization, validation UX, and progressive profiling strategies. Use when user asks about form optimization, form fields, lead form, signup form, form completion, progressive profiling, form UX, 폼 최적화, 양식, or 리드 폼.
How this skill is triggered — by the user, by Claude, or both
Slash command
/artibot-cowork:cro-formsThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
- Optimizing lead generation and signup forms
Audit Current Form -> Analyze Abandonment -> Identify Friction -> Reduce Fields -> Improve UX -> Test Changes -> Validate -> Iterate
| Fields | Avg Completion Rate | Recommendation |
|---|---|---|
| 1-3 | 25-30% | Best for top-of-funnel, signups |
| 4-5 | 20-25% | Good for lead gen with qualification |
| 6-7 | 15-20% | Acceptable for high-intent (demo, quote) |
| 8-10 | 10-15% | Only for high-value offers |
| 10+ | <10% | Avoid; use progressive profiling |
Rule: Every field removed can increase conversions 5-10%.
| Question | If Yes | If No |
|---|---|---|
| Is it needed to fulfill the request? | Keep | Remove |
| Can it be auto-detected? | Auto-fill | Remove field |
| Can it be asked later? | Progressive profile | Remove from form |
| Is it truly needed for qualification? | Keep | Remove |
| Can a smart default work? | Default + optional change | Simplify |
| Position | Best Fields | Rationale |
|---|---|---|
| First | Name, email (easy fields) | Build momentum |
| Middle | Company, role (medium effort) | Committed by now |
| Last | Phone, detailed questions | Foot-in-the-door effect |
Rules:
| Data Type | Best Input | Avoid |
|---|---|---|
type="email" with validation | Free text | |
| Phone | type="tel" with country code | Free text |
| Date | Date picker | Free text |
| Country | Dropdown with search | Long dropdown |
| Industry | Dropdown or radio (<7 options) | Free text |
| Budget | Range slider or radio | Free text |
| Message | Textarea with char count | No guidance |
| Yes/No | Toggle or radio | Dropdown |
| Practice | Impact | Implementation |
|---|---|---|
| Inline validation | +22% completion | Validate on field blur |
| Descriptive errors | +15% recovery | "Please enter a valid email" not "Invalid input" |
| Success indicators | +10% completion | Green check on valid field |
| Format hints | +12% reduction in errors | "e.g., [email protected]" |
| Preserve input on error | Critical | Never clear fields on validation failure |
| Scroll to first error | +8% recovery | Auto-scroll with highlight |
Concept: Collect data over multiple interactions instead of all at once.
| Interaction | Data Collected | Form Size |
|---|---|---|
| First visit | Email only | 1 field |
| Content download | Name, company | 2 fields |
| Webinar signup | Role, company size | 2 fields |
| Demo request | Phone, timeline, budget | 3 fields |
Rules:
| Metric | Formula | Good Benchmark |
|---|---|---|
| Form View Rate | Form views / Page views | 30-60% |
| Start Rate | Form interactions / Form views | 40-70% |
| Completion Rate | Submissions / Form starts | 60-85% |
| Abandonment Rate | 1 - Completion rate | 15-40% |
| Field Drop-off | Users stopping at field N | Varies |
| Error Rate | Errors / Submissions attempts | <10% |
| Time to Complete | Average seconds | <60s for short forms |
| Pattern | When to Use | Benefit |
|---|---|---|
| Single-step | <5 fields, simple data | Fast completion |
| Multi-step wizard | 6+ fields, complex data | Lower perceived effort |
| Conversational | B2C, casual brand | Higher engagement |
| Inline (in-content) | Blog CTAs, sidebar | Contextual capture |
| Popup/Modal | Exit intent, delayed trigger | Focused attention |
| Sticky bar | Persistent low-friction | Always available |
FORM OPTIMIZATION AUDIT
========================
Form Type: [lead gen|signup|contact|demo|checkout]
Current Fields: [count]
Completion Rate: [%]
Score: [/100]
FIELD ANALYSIS
--------------
Field | Status | Recommendation
----------------|-----------|----------------
[field name] | [KEEP|REMOVE|MODIFY] | [rationale]
FRICTION POINTS
---------------
Issue | Impact | Fix | Est. Lift
----------------|-----------|------------------------|----------
[issue] | [high/med]| [recommendation] | +[X]%
OPTIMIZED FORM
--------------
Step [n]: [fields in this step]
Field: [name] | Type: [input type] | Required: [Y/N]
RECOMMENDATIONS
---------------
Priority | Change | Est. Lift
---------|-------------------|----------
P1 | [change] | +[X]%
Field Count Rule: Every field removed = +5-10% conversions Field Order: Easy first, hard last, single column Validation: Inline, descriptive errors, preserve input on failure Progressive Profiling: 1-3 new fields per interaction, match to intent level
${CLAUDE_SKILL_DIR}/references/field-optimization-guide.md for field optimization and completion rate guide${CLAUDE_SKILL_DIR}/references/validation-ux-patterns.md for validation UX patterns and error handlingnpx claudepluginhub yoodaddy0311/artibot --plugin artibot-coworkMines projects and conversations into a searchable memory palace. Activates on queries about MemPalace, memory palace, mining, searching, palace setup, wings, rooms, drawers, or recalling past work.
Whole-repo audit for over-engineering: finds dead code, unnecessary abstractions, stdlib-replaceable dependencies. Outputs ranked findings and net line/dep savings.