"Ad click-through is improving, but we're still losing people at the form." The final gate of B2B marketing is form optimization — and it is often the highest-leverage place left to act.
A single field removal can move CVR by several percentage points. That means reducing CPA without touching your ad spend. This guide covers the decision framework for which fields to keep, how to structure steps, how to reduce psychological friction, and what to measure — with before/after examples throughout.
The four levers that drive form CVR
Address them in order of impact:
| Priority | Lever | Impact on CVR |
|---|---|---|
| 1 | Number of fields | Very high (each field removed lifts CVR measurably) |
| 2 | Psychological weight of each field | High (phone number is disproportionately heavy) |
| 3 | Step design | Medium–high (depends on field count) |
| 4 | Trust messaging | Medium (copy and design surrounding the form) |
Changing the button color is not in the top four. Making phone number optional usually is.
1. Cut fields ruthlessly — the "need to follow up" test
For every field, ask: would sales be unable to follow up without this? If the answer is yes, keep it required. If no, make it optional or delete it.
Must be required
- Company name
- Contact name (full name or family name — follow industry norms)
- Email address
These three are enough to identify the lead and send the first follow-up.
Should be optional
- Phone number
- Job title / department
- Company size
- Industry
These help sales prioritize, but requiring them drops CVR by 20–40% in observed patterns — phone number in particular has outsized impact. Mark them optional and add them to the first sales conversation instead.
Should be deleted entirely (with before/after)
| Field | Why delete | What to do instead |
|---|---|---|
| Phonetic spelling of name | Not needed for email or calls | Delete |
| Department | Appears in email signature | Delete (even optional, fill rate is under 5%) |
| "How did you hear about us?" | Low response rate even optional; UTM parameters are more reliable | Delete; use UTM tracking |
| Date of birth | Irrelevant in B2B | Delete |
Before (typical over-fielded form): Company, name, phonetic name, department, title, email, phone, company size, industry, "how did you hear about us" — 10 fields → CVR ~2–3%
After (trimmed): Company, name, email (required) + phone, company size (optional) — 5 fields → CVR ~6–9% (same creative, same targeting)
Cutting fields in half and tripling CVR is not unusual. This is the highest-ROI intervention available on most lead forms.
How to mark optional fields visibly
Write "(optional)" or "optional" next to the label — not just a footnote at the top saying "fields without * are optional." Visual distinction between required and optional reduces anxiety about the form without lowering optional fill rates.
2. Step design — make the first step feel easy
Single-screen vs. multi-step
| Format | Field count | Key advantage | Key disadvantage |
|---|---|---|---|
| Single screen | ≤5 fields | Fewer interactions, faster | All fields visible at once — can feel like a lot |
| Multi-step | 6+ fields | Higher completion rate | Requires more design effort |
Multi-step design principles
Order steps from lightest to heaviest psychological load
- Step 1: Company name, contact name (familiar, low-stakes information)
- Step 2: Email address (the core required ask)
- Step 3: Phone, company size, etc. — with "optional" explicitly labeled
Opening with the easiest questions reduces the chance of abandonment before the respondent has committed. Starting with "phone number" is one of the most reliable ways to kill completion.
Progress bar effect
"Step 2 of 3" or "One more question" activates the psychology of sunk-cost commitment: "I've come this far, I'll finish." This measurably improves completion rates. However, the progress bar needs to communicate "almost done" — "Step 1 of 7" displayed at the start can discourage rather than encourage.
Session persistence
If someone fills in 3 fields and navigates away, losing their progress on return sharply increases re-abandonment. Saving to sessionStorage or localStorage while the form is in progress reduces this drop-off.
3. Error display — tell people what to fix, not just that they failed
Errors cause more abandonment than you might expect — often not because the error itself is a problem, but because users don't know which field to fix or how. Error messages that communicate clearly recover most of those abandoners.
What not to do
- "There is 1 error" shown only after the submit button is pressed, with no field-level indication
- The error location requires scrolling to find
- Red text only — invisible to users with red-green color deficiency
- "This field is required" with no specifics
What to do
- Real-time validation on focus-out: show the error the moment the user leaves the field
- Field-level specific message: "Email address format is invalid — example: name@company.com"
- Triple signal: icon + color + text — do not rely on color alone
- Auto-scroll to the first error field when submit is pressed with multiple errors
Implementing real-time validation alone has reduced form-error abandonment by 30–50% in observed cases.
4. Trust messaging — remove doubt before the submit button
High-converting form pages address doubt before it becomes hesitation. Most of the effective trust messaging sits outside the form fields themselves.
Above the form
- Visual preview of what they will receive: whitepaper cover, video thumbnail, screenshot of the tool
- Volume and time to consume: "20 pages · 30 minutes" / "3-minute overview video"
- Delivery timing: "Instant automated delivery" / "We'll be in touch within one business day"
- Privacy stance — be specific: "No cold calls" / "Newsletter is opt-in, not automatic" / "We use your information only for follow-up on this request"
"No cold calls" added to a phone field has been shown to increase voluntary phone number fill rate. Counter-intuitive, but it works: it removes the fear that prompted hesitation in the first place.
Below the form
- Total downloads or total users ("already used by 2,500 companies")
- Customer logos (with permission)
- SSL badge and privacy policy link
What not to put near the form
- Long terms of service (users won't read them and some will leave)
- Links to other pages of your site (every exit link is a potential abandonment)
- "Check out these related articles" (people who navigate away before submitting rarely come back)
5. Mobile optimization — eliminate forms that cannot be submitted
60–70% of lead-form traffic arrives on mobile. A form designed for desktop and carried over unchanged will have input friction that drives abandonment.
Mobile readiness checklist (the minimum)
- Tap target height ≥48px on all input fields and buttons
- Email field uses
type="email"→ email keyboard opens automatically - Phone field uses
inputmode="tel"→ number pad opens automatically -
autocompleteattributes set → browser autofill works - "Submit" button stays accessible during vertical scroll
- Error messages appear below the field, not above it (above is often hidden by the keyboard)
Test on a real device, not just the browser's device emulator. Keyboard-scroll interactions in particular behave differently on physical hardware.
6. Post-submission experience — the last chance to deepen the relationship
The thank-you page is not just confirmation. It is your last opportunity to move the lead forward before they navigate away.
What to include on the thank-you page
- Immediate download link on the page — don't rely only on email. Email sometimes lands in spam; downloading directly from the page removes that uncertainty.
- 3 related content pieces — blog posts, webinar links, or case studies related to what they just downloaded
- One next-action CTA — "Book a 30-minute consultation" / "Request a demo" / "See how other [industry] teams use this"
- Time commitment on the CTA — "15-minute online demo" / "30 minutes, no preparation needed"
Redesigning the thank-you page alone has doubled or tripled same-day demo requests in observed cases, with no change to the upstream form or ad.
CV tracking setup via GTM
When the thank-you page represents a CV event, push to the dataLayer:
window.dataLayer = window.dataLayer || [];
window.dataLayer.push({
event: 'lead_form_complete',
form_type: 'document_download',
company_size: '50-100' // if collected as optional field
});
Tying this to Google Ads, GA4, Yahoo! Ads, and LINE Ads CV tags lets you see which ad creative, audience segment, and keyword is driving form completions — not just clicks.
KPIs: what to measure and what to watch out for
The metrics that matter
| Metric | How to measure | Watch out for |
|---|---|---|
| Form CVR | GA4 goal completions / form page views | Break down by device and traffic source |
| Per-field abandonment | Hotjar or Microsoft Clarity form analysis | Reveals exactly where people leave |
| Error rate | Custom event push on validation failure | What % of submit attempts fail |
| Mobile vs. desktop CVR | GA4 device-type segment | Large gap = UI problem |
| Lead quality (opp rate) | CRM + manual or automated tagging | CVR alone cannot reveal this |
The CVR-only trap
Optimizing purely for CVR can push you toward "remove everything," which raises submission count while lowering lead quality. If deals sourced from the form dry up, CVR improvement was counterproductive.
The right end metric is cost per closed deal (CPO = ad spend ÷ number of new customers), or at minimum cost per sales-qualified lead. If CVR goes up but CPO worsens, something went wrong downstream.
A/B testing: how to avoid common mistakes
One change per test
Testing "make phone optional" + "add a progress bar" + "split into 2 steps" simultaneously tells you nothing about causation. One change at a time.
Run for at least two weeks
Lead generation volume often varies by day of week and by where you are in the month. One week of data often reflects noise rather than signal.
Minimum sample size
≥200 sessions per variant with ≥95% statistical significance before declaring a winner. Moving fast on small samples risks calling an improvement that was actually random variation — or vice versa.
Summary
The highest-leverage steps for lead form CVR, in order:
- Cut or make optional: phone number, industry, company size, department — these alone are often responsible for 30–50% CVR gaps between forms
- Order fields from lightest to heaviest — company and name before email before phone
- Implement real-time validation — field-level, specific, icon-plus-color-plus-text
- Add trust messaging above the form — preview of the asset, delivery timing, explicit "no cold calls"
- Design the thank-you page — immediate download link, 3 content suggestions, one CTA
- Set up CV tracking — GA4, Google Ads, and beyond; measure through to lead quality
A perfect form redesigned in one shot matters less than one improvement per week with data review. The cadence of iteration is what compounds.
If you're building a new lead-capture form, Repoan's document-download and free-trial templates are built around the minimum-field configurations described here.
- Website embedding (see embed survey on website) — place the form inside your LP so visitors don't navigate away to a separate form page
- Brand theme (see brand-themed form design) — logo, colors, and fonts set once and applied across the form automatically
- Cloudflare Turnstile (see bot and spam protection) — blocks bot submissions that pollute your lead list by default
- Custom thank-you page (see thank-you page optimization) — post-submission download links, consultation CTAs, and related content
- CV tracking (see form conversion tracking) — GTM dataLayer integration for GA4, Google Ads, Yahoo!, and LINE Ads
AI chat generates an initial question set from a brief description of your goal; you refine by chat. This reduces time-to-first-draft for a new lead form significantly.
FAQ
Q1. Won't removing fields lower lead quality?
It depends on what you remove. Company, contact name, and email are sufficient to identify the lead and execute follow-up. Phone number, company size, and industry can be collected in the first sales conversation — they don't need to be gated at the form. That said, if your CRM uses these fields for automated scoring, keeping them optional (rather than deleting) is worth considering.
Q2. Does multi-step really improve completion rate?
For forms with 6+ fields, splitting into 2–3 steps tends to outperform a single long screen. The key variable is what goes on Step 1 — if you crowd 4+ questions onto it, the benefit reverses. Aim for 2–3 questions on Step 1, prioritizing the easiest to answer. The goal is to get the respondent into a "yes, I'm doing this" state before asking for anything sensitive.
Q3. What tools can I use to A/B test my lead form?
GA4 supports basic experiment tracking via custom events. For more control: A/B Tasty, VWO, and Kameleoon are common choices. A simpler approach for small teams is to alternate between two form URLs and compare GA4 Explore segments. Repoan's GTM dataLayer integration lets you track CV events from each form variant in GA4 without additional tooling.
Q4. Our sales team insists on required phone number. How do I make the case?
Show the absolute-number comparison. Example: "Required phone gives us 3% CVR with 100% phone collection = 30 phone numbers per month. Optional phone gives us 7% CVR with 50% voluntary fill = 35 phone numbers per month." The counterintuitive result — more phone numbers by making it optional — is common enough to be a useful conversation starter. The data from your own form will be more persuasive than the general case.
Q5. We have no thank-you page right now. Where do I start?
Start with a page that does just one thing: put the download link directly on the page (don't make them wait for email). Then add in sequence: ① three related content links → ② one consultation or demo CTA. This stepwise approach lets you measure the impact of each addition rather than changing everything at once.