Skip to main content
Linking Variations
James Harber avatar
Written by James Harber
Updated over 11 months ago

Sometimes, you want to link variations together – ensuring that users who saw a variation in one test will see a specific variation in another.

There are two ways to do this in Webtrends Optimize:

Option 1: Using Targets instead of tests

Targets allow for rules-based Variation selection, where tests are purely based on % split.

The rule you would need to create should cover:

  • User has seen this variation before OR

  • User has seen a linked variation before OR

  • User hasn’t seen any variation from any of these tests AND User falls into our % split (e.g. 50%).

Once written down this way, which we recommend, it’s easy to visualise the blocks and groups for rules. Creating this in the interface is then fairly straightforward. You will need the relevant cookie values – hop through the experiments in the Force Experiment Widget as needed to set cookies.

You will need to create these rules for each variation/group in advance, and then use them as Variation Segments (not to be confused with the Segment on the Target/Test as a whole) when creating your Target.

Option 2: Proactively building cookies / local storage

What makes us pick one experiment over another? Decisions are persisted in client-side storage, and this is looked up as an intiial check to see if a decision had been made in the past for that user. If so, we persist their experience.

This is the mechanism we can leverage for your test.

The format of our project cookies is:

_wt.project-[your domain id]-[test alias] = 1||[projectid]-[testid]-[experimentid]||[has pageview tracked]||[throttle result]

for example: _wt.project-2123456-ta_AAA = 1||2000046-2000053-2000056||false||IN

We can simply create these cookies in your variations, in which case the user will have pre-determined entry into future tests.

WT.helpers.cookie.set("_wt.project-2123456-ta_AAA", "1||2000046-2000053-2000056||false||IN", 30);

To find these cookie values, simply enter the test using the Force Experiment Widget to create the cookie, and copy it’s values. Be sure to set the “has pageview tracked” value to false though, to scope your metric tracking correctly.

Did this answer your question?