A/B testing has become an industry standard, but are you doing it right? The EDM experts at MailChimp have shared their secrets.

How long to run?

A/B testing aims to optimise your electronic mail-outs. By sending tweaked versions of an email, you can see which gets the best response.

To get reliable indicator that one variation is superior to another, MailChimp ran some broad tests. They found that different metrics require different wait times.

They report that the best variation can be accurately predicted 80% of the time by running the test:

• For opens, wait two hours.
• For clicks, wait one hour.
• For revenue, wait twelve hours.

How long is your list?

MailChimp recommends a sample of 5000 subscribers for each tested variation.

For market researchers that don’t have that reach, they recommend testing with your entire list.

No, the results won’t be directly implementable for the campaign you’ve used for testing, but they will form a good basis for future campaign content decisions.

Worth the wait?

Yep. MailChimp research shows that A/B testing works.

A/B-tested campaigns get better open and click rates than untested campaigns, ultimately generating an average of around 20% extra revenue.

Of course, every list can have its own quirks: Does your EDM experience vary significantly from MailChimp’s recommendations?

Copy Transmission is a Melbourne-based agency :: Better Brands. Loud & Clear.