# A/B Test attribution

**URL:** <https://community.hubspot.com/t/a-b-test-attribution/117979>\
**Category:** CMS Development\
**Tags:** professional\
**Created:** [October 9, 2024, 11:17am UTC](https://community.hubspot.com/t/a-b-test-attribution/117979 "2024-10-09T11:17:51Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![wmflan39](https://d10pqhm3lwygmk.cloudfront.net/user_avatar/community.hubspot.com/wmflan39/32/45434_2.png) [@wmflan39](https://community.hubspot.com/u/wmflan39)\
**Post date:** [October 9, 2024, 11:17am UTC](https://community.hubspot.com/t/a-b-test-attribution/117979/1 "2024-10-09T11:17:51Z")

</div>

I’ve started an A/B test on my home page

- A-version displays links to features and conversison channel links
- B-version displays only conversion channel links

These conversion paths have selection pages prior to the converting landing pages (ex: select food type \> select meat \> select beef), so visitors to the home page do not necessarily click directly to the landing page, so I cannot pass the version info directly into the forms from my CTAs.

How can I associate the new contact with the test version they experienced a few pages prior?

---

<div class="post-metadata">

**Author:** ![kennedyp](https://avatars.discourse-cdn.com/v4/letter/k/49beb7/32.png) [@kennedyp](https://community.hubspot.com/u/kennedyp)\
**Post date:** [October 9, 2024, 6:07pm UTC](https://community.hubspot.com/t/a-b-test-attribution/117979/2 "2024-10-09T18:07:47Z")

</div>

Hey @wmflan39! Thanks for this question.

[This Idea](https://community.hubspot.com/t5/HubSpot-Ideas/Custom-events-on-AB-Tests/idi-p/953921)notes a similar issue that could be resolved if custom event tracking were more customizable. That said, I also found[this thread](https://community.hubspot.com/t5/Blog-Website-Page-Publishing/TEST-AB-based-on-became-a-customer-date/m-p/771987)where @louischausse suggested using A/B tests at a CTA level instead of the page level. Depending on how those selection pages are linked, maybe you could explore this option.

I’d also like to invite @aviafriat26 and @Josh to provide any tips or tricks they may have for tracking A/B versions when the conversion does not occur on that page directly. Thanks y’all!

Best,  
Kennedy

---

<div class="post-metadata">

**Author:** ![FarioConsulting](https://avatars.discourse-cdn.com/v4/letter/f/a587f6/32.png) [@FarioConsulting](https://community.hubspot.com/u/FarioConsulting)\
**Post date:** [October 9, 2024, 6:43pm UTC](https://community.hubspot.com/t/a-b-test-attribution/117979/3 "2024-10-09T18:43:07Z")

</div>

Hi William,

You could append a URL parameter to track which version of your home page they encountered during their initial visit. For example, you could use ?v=1 for Version A and ?v=2 for Version B. This way, even though visitors go through multiple steps before converting, you’ll still be able to trace back to the correct test variant