Contact forms: every field costs you submissions

The long form always arrives the same way. Someone on the team gets tired of asking the same question on the phone, so it becomes a field. Then the scheduling question becomes a field. Then a dropdown for how they heard about you, because marketing wanted to know.

Nobody ever removes one. A year later the form has eleven fields, three of them required, and the owner is wondering why the site went quiet.

Why does adding a field cost me submissions?

Because a form is not one decision, it is a series of them, and every field is a fresh chance to quit.

Think about who is filling this out. Not a patient researcher at a desk. Someone standing in a kitchen with a leak, on a phone, one-handed, while a contractor is on the other line. Each field they hit is a small negotiation: do I know this answer, do I want to give it, is this worth it.

Some fields are worse than others. A required field they cannot answer yet is the most expensive kind. Asking for a project budget when they genuinely do not know the budget does not filter anybody, it just ends the session. Same with a dropdown that has no option matching their situation.

Address fields are the other common killer on mobile, because they multiply into four or five separate boxes and none of them autofill correctly.

Every field is a question you could have asked on the phone after they were already talking to you.

The best evidence for the underlying principle comes from ecommerce rather than from local business, and it is worth stating carefully. The Baymard Institute, which has tracked checkout flows since 2012, reports that the average checkout in 2024 runs 5.1 steps and 11.3 form fields, that 17% of users have abandoned a purchase because of checkout complexity, and that step count has barely moved in more than a decade. Its load-bearing conclusion after that decade of testing is that the number of steps matters far less than the number of fields users have to consider. Baymard sells UX research and benchmarking, and this is checkout research, not quote-form research. It supports the direction. It does not price your form.

Which fields should a local business actually keep?

Start with the ones you cannot reply without.

A name, so you know who you are talking to. One contact method, phone or email, and let them choose which. And one open text box asking what they need. That is a working form for most service businesses.

Add a field only when you can finish this sentence honestly: without this answer, I cannot respond at all. Not cannot respond as efficiently. Cannot respond.

Two exceptions earn their place. Location, if you serve a defined area and taking a job outside it is genuinely impossible. And preferred contact time, if your business runs on scheduled calls. Both of those save a real round trip.

Everything else is you making the visitor do your intake work before they have any reason to trust you. The button label carries weight too, and a call to action people actually act on is more specific than "submit."

Put a tappable phone number beside the form as well, because a share of your buyers were never going to type anything. If you do that, make sure the line is answered. CallRail's analysis of 1.1 million de-identified conversations, published January 2025, put missed call rates at 32% in health care, 28% in legal, 14% in home services and 9% in real estate. CallRail sells call tracking and its sample is its own customers, which if anything makes those figures optimistic.

What is the benchmark for the right number of fields?

There is not one, and you should be suspicious of anyone who quotes you a number.

We went looking for it. No study with a disclosed sample and a disclosed method exists for contact or quote forms on local service websites. Not for the ideal field count, not for the abandonment rate per added field, not for the conversion difference between a four field and an eight field form on a plumber's site. The figures that circulate in agency decks are either ecommerce checkout research relabelled, or they have no traceable primary source at all.

That absence is the useful finding. It means two things in practice.

First, any field count benchmark in a proposal is either borrowed from ecommerce, where the buyer has a cart and a reason to persist, or it is invented. Ask which. The answer tells you what kind of vendor you are talking to.

Second, the only valid data about your form is your form's data. Start events and completion events on your own site, over a month, beat every published number, because there is no published number. You can see form abandonment without buying an expensive tool, and that is the measurement that actually applies to you.

You also cannot split test your way out of this, and the arithmetic says why. Conversions needed per variant is roughly 16 ÷ r², where r is the relative lift and the 16 is 2(1.96 + 0.84)² rounded, 1.96 for 95% confidence and 0.84 for 80% power. Read backwards, a hundred conversions per variant is exactly what detects a 40% lift (derived), and cutting one field does not move a form by two fifths. A site producing a dozen leads a month will not get there on a single question. Cut the fields you cannot justify, watch direction over a quarter, and stop pretending it is an experiment.

What does one abandoned form actually cost?

Run it with your own numbers. Five lines and an hour of your attention.

Start with the click price. LocaliQ and WordStream's 2026 search advertising benchmarks put the median cost per click for Home and Home Improvement at $8.33 and the median cost per lead at $90.92. Use your own account figures if you have them; use these as a stand-in if you do not.

Set the spend. 200 clicks a month at $8.33 is $1,666.

Count the form. Suppose your analytics shows 60 form starts and 18 completions. Your cost per lead is $1,666 divided by 18, which is $92.56, close to the category median. Forty-two people started and left.

Price the abandonment. Those 42 starts cost you 42 times $8.33, about $350 a month in clicks that reached the form and walked away. Over a year, $4,200 in traffic you paid for and lost at the last step.

Now price a repair. If cutting two fields recovers five of those 42, you have 23 leads on the same spend and your cost per lead falls to $72.43. That is a $20 drop per lead, achieved with no bid change and no budget increase, and it applies to your organic and referral traffic at the same time.

Two honest notes. LocaliQ sells advertising services, its 2026 sample size is not published, and its "averages" are technically medians. And the five recovered submissions is your assumption, not a finding; nobody has published a lift figure for local lead forms because, as above, the study does not exist. The point of the arithmetic is not the answer. It is that you can now price a field, which turns a design argument into a business decision.

What about qualifying leads before they contact me?

The instinct is right and the placement is wrong.

Qualification belongs on the page, in your own words, before the form. Say who you work with, what a typical project runs, and what you do not do. People who do not fit read that and leave without filling anything in, which is exactly what you wanted. That costs you nothing and it does not punish the good leads.

Qualification inside the form does the opposite. It penalizes everyone equally, including the customer who was ready to hire you, because they all have to answer the same eight questions to get to you.

The strongest version of this is publishing a price range on the page, because it filters harder than any form field can and it does the filtering before anyone spends effort.

If you are running paid traffic yourself, this matters more, not less. Managing ads yourself takes real hours every week, and the hours you save screening bad leads are not worth the good leads a long form silently turned away.

Does a shorter form mean worse leads?

Sometimes, and the trade is usually still worth taking.

A short form will bring you some inquiries that go nowhere. The question is not whether that happens, it is what one wasted five-minute call costs you against what one lost customer costs you. For most local service businesses those two numbers are not close.

There is also a quieter effect. The customer with the biggest, most urgent problem is often the one with the least patience for a form. Emergency work, same-day bookings, high-value jobs where the buyer is annoyed and moving fast. Long forms filter those out first.

If unqualified inquiries genuinely are drowning you, fix it in the reply rather than the form. A short screening question in your first response costs you nothing and costs the visitor nothing at the moment they were deciding whether to bother.

Spam is the other objection, and extra fields are the wrong answer to it. A honeypot field hidden from humans, a time-on-page threshold, or a server side check will stop most automated submissions without asking a real buyer for anything. Adding a required question to deter bots taxes every human to inconvenience a script.

What to do this week

Open your form and count the fields. Write next to each one what you actually do with that answer. Any field where the honest answer is "nothing" or "it would be nice to know" comes off this week.

Then fill the form out yourself on your phone, thumb only, standing up. Not on your laptop. Every awkward tap you hit is one a customer hit too, and they did not have your motivation to finish.

Run the five line cost calculation above with your real click cost. Field arguments end quickly once someone puts a dollar figure on a dropdown.

Check that the confirmation actually works. Submit a real test and confirm the message arrives in the inbox someone reads, not a spam folder. Forms break silently, and the ninety-day window is when it usually happens, which is part of what should be checked after a website launches.

Last, put the phone number next to the form. Some people will never fill anything in and will happily call.

Questions about your contact form?

Email me at eric@seod.com with a link to the page your form sits on. I will go through it field by field and send you back a list of which ones to cut, which to make optional, and which one to add if something obvious is missing.

It takes me a few minutes and there is no call attached to it. If your form is already lean, I will tell you that and point at what I would look at instead.

There is also more on conversion in the rest of the library.