Hey there I’m running an experiment using sticky bucketing where primary attribute is user id and fallback is anonId that is from cookie. The experiment starts off from logged out state. I’m seeing around 25% of users who get bucketed into one variant but when they log in, they get rebucketed into control so our control size is larger than expected and probably throwing off our results

or more like 10-15%

When working correctly, sticky buckets should automatically upgrade the identifier attribute from fallback to primary, and you'd still see both keys in the bucket docs. So for a user who started as anonymous and then logged in, you should see both of these:

gbStickyBuckets__anonId||[anonId]
gbStickyBuckets__id||[id]

... and for the experiment(s) in question, they should show the same bucket value in both docs. From the user perspective, they should just use the highest priority sticky bucket doc matching their attributes and not both. Are you able to recreate any of the dropoff / variation hopping when testing locally? You can use our DevTools chrome/firefox extension to make the debugging here a bit easier.

i think the issue was that we had defaulted

id

to

anonId

when user wasn't logged in, but that meant that the

id

value was changing after login rather than falling back to anonId when logged out. i am trying to see if changing to record a new

userId

attribute will work which is allowed to be empty (as id is not allowed empty)