The Friction Audit: Every Field You Should Delete

Forms grow. That’s the first thing to understand about them. Nobody sits down and designs a signup form with nine fields. Forms get to nine fields the way attics get full: one reasonable addition at a time. Marketing wants a company name so they can segment. Sales wants a phone number so they can call. Someone adds a “how did you hear about us” dropdown because it seemed useful in a meeting. Each addition was defensible. The sum is a form that visitors abandon.

The reason this happens is that the cost of a field is invisible to the person adding it. The benefit is concrete: a column in a spreadsheet, a segment in a dashboard. The cost is a stranger, somewhere, deciding not to bother, and that person leaves no trace. You never meet the people your form drove away. So organizations keep making the same trade, concrete benefit against invisible cost, and the invisible cost wins every time without anyone noticing.

When I review websites, and when GazeSite’s conversion reviewer looks at a page, the model is a skeptical visitor with other tabs open. For that visitor, every field is a small tax, but not all taxes are equal. The real cost of a field isn’t the typing. It’s the deciding. “Email” costs almost nothing; people type their email on autopilot. “Phone number” costs a lot, not because phone numbers are long but because the visitor has to stop and think about what you’ll do with it. Someone is going to call me. Do I want that? A field that triggers that kind of deliberation doesn’t just risk being left blank. It gives the visitor a reason to reconsider the whole transaction. They came to try your product and now they’re negotiating with themselves about whether they want a sales call. You put that negotiation in their head.

So here is the audit I recommend, and it’s simple enough to do in ten minutes. For each field on your form, ask one question: what happens with this value in the first week? Not what could happen. Not what someone might someday query. What actually happens. If the honest answer is “it sits in the database,” delete the field. If the answer is “sales uses it to prioritize leads,” ask the follow-up: would sales rather have this field, or would they rather have the extra leads the field is costing? Framed that way, almost everyone chooses the leads. The field only survived because nobody ever framed it that way.

Certain fields fail this test so reliably that I’ll name them. The confirmation email field, where you type your address twice, is a relic; people paste, and the field catches nothing. The company size dropdown is enrichment data you can buy or infer later. “How did you hear about us” is a survey wearing a form field’s clothes, and it belongs after signup, if anywhere. The phone number, unless you are literally going to call every signup within the day and that call is the product, should go. And the saddest one: required fields that exist only because the form library made everything required by default and nobody changed it.

The principle underneath all this is that a form is not a questionnaire. It’s a gate. The purpose of a gate is to let people through. Every question you ask at the gate should be one you cannot proceed without, and for most products that list is startlingly short: an email, maybe a password, maybe a URL if the product operates on one. Everything else can be asked later, and here’s the part people miss: later, the answer is better. A user who has seen your product and gotten value from it will happily tell you their company size, their role, their phone number. The same person, asked at the gate, lies or leaves. You’re not choosing between having the data and not having it. You’re choosing between bad data now and good data later, plus all the users the shorter form let through.

I feel this personally because I got it wrong myself. Early versions of my own checkout asked for more than they needed, on the theory that I’d want the information eventually. What I actually wanted, it turned out, was customers. The information I thought I needed I could get by looking at the URL they submitted, which they had to give me anyway because scanning it is the product. The best form field is one where the answer is the thing the user came to do.

There’s a general lesson here that goes beyond forms. Friction is invisible to the people who create it, because they never experience their own product as a stranger. The stranger’s experience has to be deliberately simulated, by watching someone use the site, or by an audit, or just by the discipline of asking, for every single thing you require of a visitor: what do we do with this in the first week? Most of the time the answer is nothing. Delete it. Nobody has ever abandoned a form for being too short.