: One variant receives way more visitors than the other (Google Content Experiments) I've been running a number of A/B tests using Google Content Experiments, and have had the same problem across
I've been running a number of A/B tests using Google Content Experiments, and have had the same problem across all the tests I've run in the last couple weeks. One variant (the original page) receives about 2.5 times as many visitors as the other.
The pages I'm testing receive about 40,000 visits a week, and I've run single tests for over a week, so it's astronomically unlikely that this imbalance is just a statistical anomaly.
Any ideas on why this imbalance might be occurring?
EDIT: Turns out this is a "feature" (per obscure Google documentation). My question, then, is how can I turn this feature off?
More posts by @Jessie594
1 Comments
Sorted by latest first Latest Oldest Best
the imbalance is probably well explained by the "multi-armed bandit" approach that Google's statistical engines take in the back-end. I have noticed this with other users too. However, have you tried setting the option "Distribute traffic evenly across all variations" to ON? Google does the auto splitting and auto allocation when you don't explicitly say to distribute the traffic.
Terms of Use Create Support ticket Your support tickets Stock Market News! © vmapp.org2024 All Rights reserved.