A/B split testing is an essential part of any marketer’s tool box and is particularly great for email because it’s quick and easy - however the temptation is to act on results too early.

Typically, A/B split testing is done by isolating two small groups of around 5% of your list and sending them two different creative’s, subject lines or offers. Both campaigns are left until they have generated enough response  in the form of opens, clicks or sales to be statistically representative (this generally takes less than six hours). The best performing creative or subject line is then sent to the rest of the list.  Surely nothing could be simpler?

At Alchemy Worx we’ve found that the best performing cell in the first few hours rarely ends up in the lead at the end of the campaign. So there is a real danger that the version you select will not perform as well as the version you reject! In email split testing, patience really is a virtue.

It’s simple: the longer you leave your analysis, the better your results will be. The people on your list that respond fastest are often not representative of the whole list.

Our advice is to resist the urge to act to soon, no matter how much better the initial response for one version appears to be. Hold off for as long as you can, several days as a minimum – a week if possible and you’ll find email split testing really comes into its own as a device for marketers.

If you would like learn more here is a free guide to A/B split test for email marketers published by Alchemy Worx