# Form Cro

> Form Conversion Rate Optimization (Form CRO) workflow skill. Use this skill when the user needs Optimize any form that is NOT signup or account registration — including lead capture, contact, demo request, application, survey, quote, and checkout forms and the operator should preserve the upstream workflow, copied support files, and provenance before merging or handing off.

- Skill: `diegosouzapw/form-cro-2` (Agent Skill, multi-file: 7 files)
- Install (CLI): `npx skillmds@latest add diegosouzapw/form-cro-2`
- Raw SKILL.md: https://api.skillmd.com/api/skills/diegosouzapw/form-cro-2/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Marketing & Growth
- License: Unknown
- Author: diegosouzapw (https://skillmd.com/u/diegosouzapw)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/diegosouzapw/form-cro-2

---


# Form Conversion Rate Optimization (Form CRO)

## Overview

Use this skill to optimize forms that are **not** signup or account-registration flows, including:
- lead capture forms
- contact forms
- demo request forms
- quote/request-for-pricing forms
- booking or appointment forms
- application forms
- survey forms
- donation forms
- checkout and purchase-supporting forms

Preserve upstream workflow, copied support files, and provenance during implementation or handoff.

This skill is for improving **completion rate, completion quality, and recovery from friction** without reducing necessary data quality, accessibility, consent clarity, or instrumentation.

Core stance:
- optimize for the form's business job first
- remove or defer fields only when they do not justify their friction
- treat validation, mobile usability, and error recovery as conversion issues
- instrument the form as a measurable funnel, not a black box
- preserve user input whenever possible during failures and retries

## When to Use

Use this skill when the user needs to:
- audit or improve a low-converting non-registration form
- reduce abandonment on a contact, quote, demo, checkout, or application flow
- simplify fields, labels, layout, validation, or submit behavior
- decide whether a form should stay single-step or become multi-step
- improve mobile completion, autofill, accessibility, or error recovery
- add form analytics so optimization can be measured instead of guessed

Do **not** use this skill as the primary workflow for:
- signup or account-registration optimization
- identity, legal, or compliance interpretation beyond UX implementation
- payment gateway integration redesign unrelated to form UX
- deceptive patterns such as forced opt-ins, disguised consent, or hidden costs

## Inputs

Collect as many of these as possible before changing the form:
- form URL, component file, or design artifact
- form type and business goal
- current fields and which are required
- traffic source or user intent, if known
- device split, especially mobile share
- baseline metrics: view rate, start rate, completion rate, error rate, abandonment point
- validation rules and backend constraints
- legal/compliance requirements for required fields, consent, and disclosures
- analytics already in place

If data is missing, proceed with a structured heuristic audit and clearly label assumptions.

## Output Format

Return results in a concrete operator-ready format.

Your default output should contain these sections in order:
1. **Form Summary** — form type, business goal, audience, and main friction hypothesis
2. **Observed Issues** — prioritized list of conversion blockers with severity
3. **Recommended Changes** — field, layout, validation, mobile, trust, and measurement changes
4. **Implementation Notes** — exact copy, markup, behavior, or analytics details to change
5. **Measurement Plan** — events, funnel stages, and success metrics
6. **Risks / Tradeoffs** — what data could be lost, what legal or operational checks are needed
7. **Next Step** — the smallest safe change set to ship first

When reviewing an existing form, prefer a table like this:

| Area | Current behavior | Problem | Recommendation | Expected impact |
|---|---|---|---|---|
| Fields | Phone + company mandatory | High friction for low-intent leads | Make optional or defer | Higher starts and completions |
| Validation | Errors only after submit | Late recovery | Add inline validation for format-sensitive fields | Lower error abandonment |
| Mobile | Email keyboard not triggered | Slower entry | Use correct input type and autocomplete | Faster completion |

When rewriting or implementing, end with a **ship-ready change list** containing only actionable items.

## Workflow

1. **Define the form's job**
   - Identify the primary business outcome.
   - Identify the user's immediate goal.
   - Distinguish essential data from "nice to have" enrichment.
   - Check whether downstream teams truly use each field.

2. **Map the current friction path**
   - List visible fields, optional fields, hidden fields, consent blocks, and submit states.
   - Note where users must think, search memory, switch context, or fear commitment.
   - Identify likely abandonment points: first screen, validation, captcha, long text, payment, or unclear CTA.
   - Review desktop and mobile separately.

3. **Reduce or defer unnecessary data collection**
   - Remove fields that are not used or not needed at this stage.
   - Convert some required fields to optional if they are helpful but non-essential.
   - Defer advanced qualification until after the main conversion.
   - Consolidate duplicated questions.
   - Prefer smart defaults, autofill, and inferred values where appropriate.

4. **Improve structure, labels, and flow**
   - Prefer a simple single-column layout.
   - Keep labels persistent and close to fields; do not rely on placeholder-only labeling.
   - Group related fields with clear headings.
   - Use field-level help only where users are likely to hesitate.
   - Choose single-step unless the form is long enough that chunking clearly reduces cognitive load.

5. **Tune validation and error recovery**
   - Decide deliberately between browser-native and custom validation behavior.
   - Validate format-sensitive fields early enough to prevent wasted effort.
   - Show errors near the affected field using specific, corrective language.
   - Preserve entered values after errors and retries.
   - Move focus to a clear summary or first failing field when submission fails.

6. **Optimize mobile completion**
   - Use correct input types, autocomplete tokens, and inputmode values.
   - Trigger the right keyboard for email, phone, numeric, and URL fields.
   - Minimize long dropdowns and awkward date pickers when better patterns exist.
   - Keep tap targets comfortable and avoid layout shifts while typing.

7. **Reduce trust friction at the point of commitment**
   - Make the submit action explicit about what happens next.
   - Place reassurance near sensitive fields or the submit area.
   - Clarify privacy, pricing, response expectations, or card safety where relevant.
   - Remove ambiguous or overly aggressive CTA copy.

8. **Instrument the form as a funnel**
   - Track form viewed, started, field error, step progressed, submitted, success, and failure.
   - Capture the field or step causing errors without logging sensitive raw values.
   - Segment by device, traffic source, and major form variant.
   - Ensure success events reflect actual completion, not just button clicks.

9. **Ship the smallest meaningful test**
   - Prioritize changes with high impact and low implementation risk.
   - Avoid changing every variable at once if measurement matters.
   - Record hypotheses, expected effect, and rollback criteria.

## Examples

### Example 1: Lead capture form audit

**Input**
```text
Form type: B2B demo request
Fields: first name, last name, work email, phone, company, job title, company size, country, product interest, message
All fields required except message
Mobile traffic: 68%
Observed issue: high abandonment on first screen
```

**Output**
```text
Form Summary
- Goal: collect qualified demo requests
- Main friction hypothesis: too much required qualification before user commitment

Observed Issues
1. Too many required fields for top-of-funnel intent
2. Phone and company size create commitment anxiety
3. Mobile completion is slowed by long field list and likely keyboard switching

Recommended Changes
- Keep required: first name, last name, work email, company
- Make optional: phone, job title, company size, message
- Defer product interest to post-submit routing if possible
- Use single-column layout with persistent labels
- Add autocomplete and input types for name, email, tel, organization
- Change CTA from "Submit" to "Request demo"

Measurement Plan
- Track view, first interaction, validation error, submit attempt, success
- Track field_error events for email and phone format only
- Compare completion rate and qualified-meeting rate before/after
```

### Example 2: Checkout support form improvement

**Before**
```text
- Coupon code shown expanded by default
- Billing and shipping fields duplicated
- Errors appear only after submit at top of page
- Cardholder name and phone both required without explanation
```

**After**
```text
- Collapse coupon field behind an optional action
- Offer "billing same as shipping" by default when valid
- Show inline field errors and focus the first invalid field on submit
- Explain why phone is needed or make it optional if not operationally required
```

Expected result: fewer interruptions, faster completion, lower error-driven abandonment.

For a fuller worked example, open [`examples/worked-example.md`](examples/worked-example.md).

## Best Practices

Do:
- verify which fields are operationally necessary before removing or requiring them
- optimize around user intent and funnel stage
- keep labels visible at all times
- use explicit required/optional treatment consistently
- preserve user-entered values through validation and server-side failures
- instrument errors, starts, and completions as separate events
- review mobile behavior directly, not only desktop screenshots
- ensure error text explains how to fix the problem
- align CTA copy with the actual next step

Do not:
- remove fields that legal, payment, fulfillment, or routing systems actually require
- ask for sensitive information earlier than needed
- rely on placeholder text as the only label
- use vague error copy such as "invalid input" with no remedy
- fire success analytics on button click instead of confirmed completion
- collect sensitive raw values in analytics payloads
- force multi-step flow just because the form feels long without testing whether chunking helps
- use fear, deception, or hidden consent to lift apparent conversion

## Troubleshooting

**Symptoms:** High abandonment before first submit, especially on mobile.

**Solution:** Reduce visible required fields, shorten first-screen content, verify correct mobile keyboard/input semantics, and test whether one or two intimidating fields can be deferred.

**Symptoms:** Many submit attempts but low successful completion.

**Solution:** Audit validation timing, error clarity, and whether entered values are preserved after failure. Confirm that server-side errors are surfaced clearly and near relevant fields.

**Symptoms:** Users start the form but drop at a specific field.

**Solution:** Check whether the field is truly required, whether the label is ambiguous, whether formatting expectations are unclear, or whether the field introduces commitment anxiety.

**Symptoms:** Analytics says conversions are healthy, but backend completions are lower.

**Solution:** Verify that the success event is triggered only after confirmed completion, not on button click or optimistic client state.

**Symptoms:** Mobile users underperform desktop users by a wide margin.

**Solution:** Review keyboard types, autofill tokens, field spacing, dropdown usability, sticky footers, and any layout shifts caused by inline errors or virtual keyboard appearance.

**Symptoms:** Accessibility fixes seem to conflict with visual simplification.

**Solution:** Keep persistent labels, programmatic associations, descriptive errors, and logical focus order. Simplification should remove friction, not remove orientation.

## Additional Resources

- [`references/domain-notes.md`](references/domain-notes.md) — open this for decision tables on field reduction, step count, validation choice, semantic/mobile attributes, and analytics taxonomy.
- [`examples/worked-example.md`](examples/worked-example.md) — open this for a concrete before/after audit and implementation-style recommendation set.

## Related Skills

No related local skills were provided in the source context.