I'm looking into doing some A/B testing and am not looking to assess a single landing page (as is I think is the focus of the Multivariate module, but want to assess a new theme's usability. I want exactly what the split test module describes here:

"This modules enables the split testing of two Drupal themes. Each alternating anonymous visitor to the site is assigned a different theme, and if the user reaches the action page (such as a thankyou page) the conversion is recorded. The split test ends when the total number of visitors to test, is reached."

So, is Multivariate looking to extend to do this? Does Multivariate already do this, but just isn't described that way?

Interested in more usability/accessibility work in the Drupal community.

CommentFileSizeAuthor
#1 multivariate.png29.79 KBfederico

Comments

federico’s picture

StatusFileSize
new29.79 KB

Hi, I don't want to open a new issue so I'm reopening this one. I've created an A/B test, one original page (a webform), one variation, both with the same thankyou page. I've visited the original page with different browsers and clearing the cache, and only once I saw the variation. Anyway, I've submitted the form using both versions, and I got following data http://drupal.org/files/issues/multivariate.png (view attached image).

Is this working? Am I missing something? I just don't understand the results. I thought I was going to see something like:
Version A: Visitors=100 Sucess=10 (10%)
Version B: Visitors=95 Sucess=50 (53%)

Thanks for your help!

soyarma’s picture

Status: Active » Postponed

Hey federico--I don't know for sure why you had that issue, I'd need to see a bit more of your config, but I imagine that your issue is long passed.

Mgifford--I'm working to make Multivariate context aware, so you will actually be able to do this in the future.

mgifford’s picture

Nice @soyarma - I look forward to testing it! Hopefully I can help bring the functioning from dev to a stable release.