Onboarding asks for 9 fields—users finish 3 and ghost at step 2

You've built a signup form with nine fields. Name, email, mobile, company, role, team size, budget range, how-you-heard-about-us, and a password. The analytics tell you most people complete three and vanish around step two. So what's actually happening in that moment between field three and field four — and why does it feel, from the user's side, less like a form and more like a decision they're not ready to make?

The form is asking for trust the user hasn't decided to give yet

There's a well-known finding in behavioural economics — Kahneman and Tversky's work on loss aversion — that people weigh potential losses roughly twice as heavily as equivalent gains. On a signup form, every field is a small, immediate cost: time, cognitive effort, and for fields like mobile number or budget range, a disclosure that feels like risk. The benefit — whatever your product does — is abstract and deferred. You're asking users to pay now for a payoff they can't see yet.

That asymmetry explains a lot of what looks like irrational drop-off. A user who abandons at field four isn't lazy. They're running a fast, mostly unconscious cost-benefit calculation, and the cost side of the ledger just got heavier than the benefit side. The nine-field form didn't fail because nine is a magic number. It failed because the form front-loads cost before the user has any reason to believe the benefit is real.

Behavioural scientists call the underlying pattern variable-ratio reinforcement when rewards arrive unpredictably — it's why some interactions feel sticky. Onboarding forms rarely have that quality. They're predictable in the worst way: every field is a known, boring cost with no reward until the very end. There's no reason to keep going except discipline, and discipline is not something you can ask a stranger for.

Where the ghosting actually happens (and it's rarely where you think)

Most teams assume people quit at the field that looks hardest — the budget range, the mobile number. In practice, drop-off clusters around a different pattern: the point where the form stops feeling like a conversation and starts feeling like an interrogation.

Consider a mid-sized Australian SaaS company that rebuilt its onboarding in 2023. Their original flow asked for eight fields across two screens. Completion sat at 31%. When they cut it to four fields on a single screen — name, email, company, and a single dropdown for "what are you trying to do?" — completion jumped to 68%. Same product, same traffic source, same offer. The only thing that changed was the shape of the ask.

But here's the part that surprised them: the four-field version didn't lose them qualified leads. In fact, sales reported better conversations, because the "what are you trying to do?" dropdown gave them more useful context than the budget range ever had. The field they thought was essential was the one nobody wanted to answer, and the field they almost cut was the one that actually helped.

The lesson isn't "fewer fields always wins." It's that fields have different psychological weights, and the heaviest ones — anything that feels like a commitment, a judgement, or a disclosure — should be moved as late as possible, or removed entirely until trust exists.

Progressive disclosure beats a single heroic form

There's a reason video games don't teach you every mechanic in the first level. They introduce one system, let you succeed, then add the next. The reward loop — try, succeed, get a small unlock — is what keeps people engaged. Onboarding forms can borrow the same structure without feeling gimmicky.

The practical version looks like this:

Step one: the minimum viable ask. Name and email. That's it. Get the account created. The user has now invested something, which triggers a mild commitment bias — they're slightly more likely to continue because they've already started.

Step two: the contextual ask. One or two questions that visibly change what happens next. "What's your team size?" is fine if the answer actually alters the dashboard, the pricing view, or the onboarding checklist. If it doesn't change anything, don't ask it yet.

Step three: the profile ask. Role, company, how-you-heard. These are enrichment fields. They can live in settings, or be collected gradually over the first week through in-app prompts that feel like helpful nudges rather than gatekeeping.

The key is that each step should feel like the user is getting something, not giving something. A progress bar helps, but only if the steps are genuinely small. A progress bar on a nine-field form just makes the mountain visible.

What the drop-off is really telling you

When 70% of users ghost at step two, the instinct is to blame the copy, the button colour, or the mobile layout. Sometimes those matter. But more often, the drop-off is a signal about perceived value, not perceived effort.

Ask a blunt question: if a user completed this form, what would they get in the next 60 seconds? If the answer is "access to a dashboard they don't understand yet" or "a sales call in three days," the form is asking for trust before it's earned. If the answer is "they can immediately see something useful, personalised to what they just told us," the form has a reason to exist.

There's a concept from decision science called the "peak-end rule" — people judge an experience largely by its most intense moment and how it ends. For onboarding, the end of the form is the moment of highest friction. If the last thing a user does before entering your product is answer a question they resent, that resentment colours everything that follows. If the last thing they do is click a button that immediately shows them something relevant, the whole experience reads as smooth.

Where to take this next

The forward-looking move isn't to shave one more field off the form. It's to stop treating onboarding as a data-collection exercise and start treating it as the first act of the product experience. Every field is a design decision about what the user should feel at that moment — curiosity, momentum, or doubt.

Run a simple experiment this month. Take your current form and mark each field with a single label: essential to create the account, changes what the user sees next, or nice to have for marketing. Anything in the third bucket gets cut or deferred. Anything in the second gets moved as early as it can go without breaking the flow. Anything in the first stays, but gets rewritten so it reads like a question a helpful person would ask, not a form a compliance officer would design.

Then watch what happens at step two. If the ghosting stops, you've learned something more useful than any A/B test on button colour: you've learned that users weren't unwilling to engage. They were unwilling to pay trust they hadn't been given a reason to extend. The form was never the problem. The order of the ask was.