# A/B test: remove fields from the checkout form

> If we hide optional checkout fields (company, second address line) behind a link, then orders will rise, because a shorter form gets finished sooner, especially on mobile.

- Source: https://bixlow.com/en/ideas/fewer-checkout-fields
- Language: en
- Updated: 2026-11-02
- Publisher: Bixlow

## Why it can work

Count your checkout fields on a phone. Each one means tap, wait for the keyboard, type, close the keyboard. Almost nobody fills in “Company” or “Address line 2”, but everyone has to scroll past them and wonder whether they are required.

A form that looks short gets started. A long one gets left for later, and later never comes.


## How to set it up

Check your platform first. Many of them restrict what can be changed inside the checkout, and Shopify is the strictest: there you may only be able to run this test in the cart or on what shoppers see before paying. On WooCommerce or PrestaShop the checkout is just another page of your site and can be tested like any other.

In Bixlow you describe the change to the agent, it builds it on your site and launches it as an A/B test once you approve it. A/B tests are on the paid plans, starting with Standard.

1. List the fields and mark the ones that are truly optional: company, second address line and, sometimes, a tax ID for people who don't need an invoice.
2. In the variant, those fields hide behind a link such as “Add company or more address details”.
3. Leave required fields alone: the test is about hiding, not about no longer asking for data you need to ship.
4. If most of your traffic is mobile, you can run it on mobile only, where every field weighs more.


## What to measure

|  | Metric |
| --- | --- |
| Primary metric | Orders per visitor |
| Metric that must not get worse | Delivery problems caused by incomplete addresses |
| Where it is measured | On the order confirmation page |

_Your testing tool does not see delivery problems: ask whoever handles shipping to track them during the test._


> **When I would not test it.** If you sell a lot to businesses that need an invoice with their details, hiding those fields makes them hunt. And if your deliveries often fail because of missing flat or door numbers, the second address line may not be as optional as it looks.


## FAQ

### Which fields can I safely hide?

The ones the fewest people fill in. If you can, check what share of your orders include a company or second address line. If it is low, it is a good candidate.

### Wouldn't it be better to remove them altogether?

You could, but then people who need them have nowhere to put them. Hiding them behind a link is the middle ground, and that is what this tests.

### What about address autocomplete instead?

That is a different idea and also worth testing. Test them separately: if you ship both at once you won't know which one moved orders.

## Further reading

- [Previous idea: guest checkout first](https://bixlow.com/en/ideas/guest-checkout-first)
- [How to write a CRO hypothesis](https://bixlow.com/en/glossary/cro-hypothesis)
- [A/B testing without a developer](https://bixlow.com/en/platform/ab-testing)