A/B testing your emails
Two subject lines per step, measured on opens and clicks, with a winner.
Subject lines decide whether anything else in the email matters. They are also the cheapest thing to test, because the rest of the message stays the same.
How it works
- A flow step can carry a second subject line and a split percentage.
- Sends, opens and clicks are counted separately for each variant.
- A winner is recorded, so the result is a decision rather than a chart somebody has to interpret.
Read clicks, not opens
Open rates have been unreliable since mail clients started pre-fetching images on the recipient’s behalf — a proportion of your “opens” are a privacy proxy, not a person. Clicks are a real action.
Test on opens if that is all you have, but decide on clicks and, where the volume allows it, on revenue.
What it does not do
It will not tell you a result is significant. On a list of a few hundred, a five per cent difference is noise, and calling a winner from it will send you confidently in a random direction. Small lists should test big differences — a different angle, not a different adjective.
Stop reconciling stock by hand
One catalogue, six channels, one stock figure that stays right.