Forms that capture more: what to remove before adding
Every extra field costs you leads. How many fields to use, what to ask and what not, how to measure where people drop off, and why WhatsApp changes the maths in Colombia.
Digital Marketing, Medical Marketing & AI Consultant · btodigital
Every field you add to a form costs you leads. That is the one rule that holds across every case I have measured, and it is the one almost nobody applies, because the temptation to ask for one more detail always wins.
The right question is not what you would like to know. It is the minimum you need in order to reply.
How many fields should it have?
The minimum needed to contact the person. Almost always two: a name and one way to reach them.
| Fields | When it makes sense |
|---|---|
| 2 (name and contact) | Document download, subscription, first contact |
| 3 to 4 | When you need to route to the right salesperson |
| 5 or more | Only if the form replaces a qualification call |
Everything else gets asked later, in the conversation, which is where people actually answer. A long form does not qualify: it filters by patience, not by buying intent.
Which fields are worth asking for?
It depends what you sell, but two outperform the rest:
- One that routes. City, specialty, company type. Whatever lets you decide who handles it.
- One that qualifies in a line. “When do you need this by?” separates browsers from buyers better than budget, which people do not want to type anyway.
What is rarely worth it: exact job title, company size in brackets, “how did you hear about us” (measure that with the source tag, not by asking) and any dropdown with fifteen options.
What about the button?
It should say what happens next, not “Submit”. “Download the book”, “Book the 30 minutes”, “See the prices”. The button is the last thing someone reads before deciding.
And below it, a line that lowers the risk: how long you take to reply, or that they will not get promotional email. In Colombia that matters.
What about mobile?
That is where most is lost. Three things I always check:
- That the right keyboard appears by itself. Numeric for phone, email for email. It is a one-line attribute and almost nobody sets it.
- That it can be filled one-handed, without zooming.
- That the button is not hidden by the keyboard when it opens.
Open your own form on a phone using mobile data and fill it in. The problem surfaces on its own.
Form or WhatsApp?
In Colombia this decision changes the numbers more than any design tweak.
| Form | ||
|---|---|---|
| Friction | Medium | Low |
| Automatically recorded | Yes | Only with a system behind it |
| Response expectation | Hours | Minutes |
| Risk | Nobody reads it | Nobody replies fast |
What works is having both, with one warning: if nobody answers WhatsApp within minutes, it converts worse than the form, because the expectation is different.
What do you measure?
Not how many forms were submitted. Three things:
- How many people start and do not finish. Visible through session recordings or an event per field.
- Which field they drop off at. Almost always the phone number, or a long dropdown.
- How many of those who submit end up buying. A form bringing leads who never buy is not working: it is generating work.
Point three is what settles the eternal argument between marketing and sales.
And A/B tests?
Carefully. Detecting a moderate improvement needs thousands of visits per variant. With a few hundred a month, any difference you see is chance.
Without that volume, make large changes one at a time and compare month over month, knowing it is a signal and not proof. Removing three fields is a large change. Changing the button colour is not.
Frequently asked questions
How many fields should a form have? The minimum needed to reply, almost always a name and one way to reach them. Every extra field costs leads, and the rest can be asked in conversation.
Does asking for a phone number reduce submissions? Yes, noticeably. It is worth it only if your sales process is phone-based. If you can resolve by email or WhatsApp, make it optional.
What should the button say? What will happen when they click: “Download the book”, “Book the 30 minutes”. Never “Submit”.
Is a form or WhatsApp better? Both, if you can. WhatsApp converts better in Colombia as long as someone replies quickly; if the message sits unanswered for hours, the form performs better.
How do I know which field people abandon? With session recordings, or by firing an event per field in your analytics. It is almost always the phone number or a dropdown with too many options.
How many visits do I need to test two versions? Thousands per variant to detect moderate differences. With fewer, make one large change at a time and observe the full month.
Is it useful to put the form in several places on the page? Yes, on long pages. What does not work is each one asking for different things: it confuses people and complicates measurement.
