Loser
Low confidence
Walmart: Site-wide — Big Navigation A/B Test With 8 Confounded UI Changes
Walmart ran a larger redesign experiment of their navigation. If I observed correctly after triple checking, it was a leap variation with at least 8 changes grouped together. Unfortunately, as of this month it seems that the experiment has stopped with no sight of the variation - hinting at a rejection for whatever reason. In retrospect, I do have some ideas along with my personal and mixed bets on these UI changes. All in all, I suspect a possible confounding situation with some negative changes cancelling out the positive.
Hypothesis
If we test a similar change on our any pages as Walmart rejected, we should be cautious — their rejection suggests it underperformed.
Why This Works
Key Learnings
Viewed 42+ times
Related Entries
Test Idea
Winner
Product Category Links on Homepage → +17% Sales
navigationhomepage
Expected lift: 5%-20%
High confidence
Test Idea
Winner
Multiple Steps
navigationlisting
Expected lift: 5%-5%
High confidence
Research Finding
Principle
4+ Variation Tests Win 2.4x More Often With 27% Higher Lift
testing_strategyany
Expected lift: 10%-30%
High confidence
Research Finding
Principle
Test Quality Peaks at 1–10 Annual Tests Per Engineer; Drops 87% After 30
testing_strategyany
High confidence
Test Idea
Winner
Pattern #14: Exposed Menu Options
navigationproduct
Expected lift: 0%-5%
High confidence