Mum's gone to do an A/B test
A reality show illustrates a simple and entertaining retail sales experiment.
The other night I found myself watching a documentary on BBC2 following the fortunes of the Iceland supermarket chain and their colourful owner Malcolm Walker. One interesting section showed the thought process behind a retail a/b test.
Walker appeared in the company’s in-house test store musing as to what he saw as the confusing labelling of wine. “Why all these labels?”, observed the discerning oenophile. “Why not just have three colorful blocks - ‘good red wine’, ‘good white wine’, ‘good rosé wine’? We’re going to try it in five stores and see how it does.”
Some time later, he’s in a real-life store explaining how the experiment went. “We tried having just these three types of wine,” he explained, pointing at some generically branded bottles, “and they didn’t sell”.
“But then we tried putting the ordinary wine back alongside, and wine sales in this store are now up 15%.” Which I thought was interesting. The colourfully basic branding on the wine (and I’ve seen similar types in more upmarket stores like Waitrose) didn’t sell until it was contrasted with more conventional and less stand-out labels, then it either did or attracted attention to the other brands which raised overall sales. Cheers!