A lead form is a purchase. The visitor pays with their contact details and their expectation of what happens next; you deliver something in return. When a form underperforms, the price is wrong for the goods.
And the price is higher than most marketers think, because a lead form doesn't just cost keystrokes. It costs the visitor a phone call they may not want.
What each field really costs
Email: cheap. It's a filterable address.
Phone number: expensive. It's a commitment to be interrupted, by a salesperson, at a time not of their choosing. This one field ends more lead forms than any other.
Company size, role, budget: expensive in a different currency. They signal that the visitor is about to be evaluated and possibly rejected, and some people decline to be evaluated today.
Anything unexplained: expensive by default, because the visitor supplies the worst plausible reason for the field's existence.
What you're offering in return
Be honest about it. "Contact us" offers nothing — the visitor gives their details and receives a sales conversation they must now manage.
Compare that to: a specific document, a working tool, a scan result, a price. Something delivered immediately, whose value the visitor can assess before deciding to talk.
Most underperforming lead forms are priced like a purchase and stocked like a request for a favour.
Set the expectation at the button
The strongest single improvement to most lead forms is a sentence next to the submit button telling the visitor exactly what happens next: who contacts them, how, and when. "A coach emails you within one working day. No phone calls unless you ask."
That sentence costs nothing and removes the largest unstated fear.
Qualifying fields are a deliberate trade
Every qualifier improves lead quality and reduces lead volume. That's a legitimate trade — but make it deliberately.
If the sales team is starved, remove qualifiers and qualify in the follow-up. If they're drowning, add one qualifier and explain why it's there. What you shouldn't do is inherit five qualifying fields from a template and then wonder about the conversion rate.
Where the leak actually is
Arrive-and-never-start means the page failed before the form: unclear offer, or an ask that outweighs it.
Start-and-stop means a field. It is almost never the first, and disproportionately the phone number.
Submit-and-no-reply means your leak is in the follow-up, not the form — and it's the most expensive kind, because you paid for those leads twice.
Defrixa's free scan reads the structural half: the weight of the ask, the clarity of the offer, whether one action dominates. Which field ends the typing is behavioural, and needs the snippet.
Common questions
Only if a call is genuinely the next step and you say so. Making it optional, with a reason beside it, usually recovers most of the loss.
Often, when the first step is cheap. The gain comes from starting, not from the steps themselves — put the easy question first.
Rarely. It names your process. "Get a quote," "See a sample report," "Book a 20-minute call" name what the visitor receives.