CRO & UX
The form and checkout leaks costing you leads
Most sites lose more buyers at the form than at the ad. Here is a leak audit you can run this week, with the fixes that recover the drop-off.
TL;DR
- The biggest conversion leaks are rarely the headline or the price. They are the form and the checkout: too many fields, confusing validation, slow mobile, and missing trust at the moment of decision.
- Run the audit in this order: count the fields, test the errors, check it on a real phone, look for trust signals, and pull the funnel analytics to find the exact step where people leave.
- Most fixes are cheap: cut fields, fix error messages, reserve space so the layout does not jump, and put trust signals where the doubt is.
- You cannot fix a leak you cannot see. Instrument the funnel first, then change one thing at a time and watch the step-level drop-off.
- See how we approach conversion optimization, and why website performance is part of the same problem.
Where leads actually leak
The drop-off is usually at the form, not the first impression.
Teams spend weeks on the hero and the headline, then send the buyer to a form that asks for twelve fields and rejects the phone number for a format reason it never explained. The visitor who liked your offer enough to start the form is the most valuable person on the site, and that is exactly where most sites lose them. A small improvement at the form recovers buyers who already wanted to convert.
The pattern repeats in checkout. People abandon not because they changed their mind about the product but because of friction: a forced account, a surprise shipping cost, a step that fails on mobile, or a layout that jumps while they type. Each of these is a leak you can find and close without touching your traffic or your offer.
The good news: because these buyers are high-intent, the fixes pay back fast. You are not trying to attract new visitors, you are stopping the ones you already paid for from leaving at the last step.
You cannot fix a leak you cannot see. Instrument the funnel first, then change one thing at a time and watch the step-level drop-off.
The leak audit
Seven checks to run this week, in order.
01
Count the fields, then cut them
Every field is a chance to abandon. List each one and ask whether you truly need it to start the conversation. Phone number, company size, and "how did you hear about us" can usually wait until after the lead is captured. A form that asks for name, email, and one real qualifier converts far better than one that asks for everything upfront.
02
Test what happens when it goes wrong
Submit the form with a missing field, a wrong email format, and a long pause. Good error UX shows the message next to the field, in plain language, the moment the problem is clear, and never clears what the user already typed. Bad error UX rejects the whole form, scrolls away from the problem, or wipes the inputs. The error path is where careful buyers quietly give up.
03
Use it on a real phone
Open the form on an actual mobile device, not a desktop emulator. Check that the right keyboard appears (numeric for phone, email for email), that tap targets are big enough, that the layout does not shift while the keyboard opens, and that you can complete the whole thing one-handed. Mobile is where most traffic is and where most forms are worst.
04
Put trust where the doubt is
At the moment someone hands over an email or a card, doubt spikes. That is where a trust signal earns its place: a short privacy line under the email field, a security note at payment, a recognisable logo or a real result near the submit button. Trust signals buried in the footer do nothing; placed at the point of decision, they recover hesitant buyers.
05
Remove forced steps in checkout
A forced account before purchase, a surprise cost at the final step, or a multi-page checkout that hides progress all cost sales. Offer guest checkout, show the total cost early, and make progress visible. Every step you remove between intent and completion is leads recovered.
06
Check the speed of the form itself
A form that lags on each keystroke, or a submit button that takes seconds to respond, reads as broken. Heavy scripts and layout shifts at the form are a conversion problem as much as a performance one. See our Core Web Vitals guide for why interaction delay (INP) matters most exactly here, at the point of decision.
07
Read the funnel, not the average
A single conversion rate hides where the leak is. Instrument each step (page view, form start, field errors, submit, success) and look at the drop between steps. The step with the biggest fall is your priority. Without step-level data you are guessing, and guesses cost as much as the leak.
At a glance
- 1
- Change one thing at a time. Bundle five fixes and you will never know which one moved the number.
- 3
- A starting target for most lead forms: name, email, and one real qualifier. Everything else can wait.
- 5+
- Instrument at least these funnel steps: view, start, error, submit, success. The biggest drop is your priority.
Fix it in this order
A practical sequence once you have found the leaks.
✓
Instrument the funnel so you can see step-level drop-off before changing anything.
✓
Cut every field that is not needed to start the conversation.
✓
Fix error messages: inline, plain language, and never wipe what the user typed.
✓
Reserve space for errors and dynamic content so the layout does not jump.
✓
Add a guest checkout path and show total cost early.
✓
Place one trust signal at each point of doubt (email, payment, submit).
✓
Test the whole flow on a real phone, one-handed.
✓
Change one thing at a time and compare step-level drop-off before and after.
Common questions
What teams ask when leads stall at the form.
How many fields should a lead form have?
As few as it takes to start a useful conversation. For most B2B lead forms that is name, email, and one qualifying question. Every extra field lowers completion, so the rest (phone, company size, budget) should be captured after the lead exists, in a follow-up or a later step. The exception is when a field genuinely improves lead quality enough to justify the drop in volume, which you should measure rather than assume.
What is the single most common conversion leak?
Poor error handling. A form that rejects an input without explaining why, scrolls away from the problem, or wipes what the user typed loses careful, high-intent buyers at the worst possible moment. Inline validation in plain language, shown next to the field and without clearing the form, is one of the highest-return fixes you can make.
How do I find where people drop off?
Instrument the funnel as discrete steps: page view, form start, field-level errors, submit attempt, and success. Then look at the drop between each step rather than the overall conversion rate. The step with the largest fall is your priority. Session recordings and a quick test on a real device fill in the why once the analytics show you the where.
Does page speed really affect form conversion?
Yes, and it matters most exactly at the form. Interaction delay (the lag between a tap or keystroke and the response) is felt sharpest when someone is actively typing or trying to submit. A form that stutters reads as broken and gets abandoned. This is why conversion and performance are the same problem at the point of decision. See our Core Web Vitals guide.
Should I run an A/B test for every change?
Only when you have the traffic to reach significance in a reasonable time. Below that, A/B testing leads to false conclusions from noise. For lower-traffic sites, make one well-reasoned change at a time, watch step-level drop-off over a few weeks, and rely on clear before-and-after differences. Reserve formal A/B testing for high-traffic pages where the volume supports it.
Is guest checkout worth it if I want accounts?
Yes. Forcing account creation before purchase is one of the most reliable ways to lose sales. Offer guest checkout, complete the sale, then invite the buyer to create an account afterwards using the details they already entered. You get the conversion and most of the accounts, without the forced step that drives people away.
How we approach it
Measure the funnel, then close one leak at a time.
We start by instrumenting the funnel, because a single conversion rate hides the leak. Once we can see step-level drop-off, we fix in order of impact: usually fields first, then error UX, then mobile and trust, then speed. We change one thing at a time and compare the step-level numbers, so we know which fix moved the result and which did not.
Performance is part of the same job, because interaction delay hurts most at the form. See how we approach conversion optimization, and our Core Web Vitals guide for the speed side of the point of decision.
Concrete solution
Bring the operational risk.You get a clear diagnosis and a concrete next step.
We are the right fit if you want a team that pushes back when it matters. See outcomes and metrics →
Reviewing first?
Company evidenceon the site.
Engagements with commercial outcomes on Work. Team bios and operating model on About. Nothing to download. Review it before you commit to a call. Open to review. Commit when ready.