Dont Test Everything
What I’m about to say is borderline heresy in marketing…Hey marketers, you don’t have to test everything.
Marketers love to test things to see which wording, which copy, which button, which message performs better than the others. It’s a noble pursuit. If there was a marketing version of the hippocratic oath it would be “no variable un-tested”
And there’s a ton of value in testing and experimenting in marketing. It’s truly a routine part of the job and we do it here at TradePending, but I believe that we’ve reached a point of over-testing.
There’s a couple big fundamental pieces needed for testing to be successful.
First – the size of your audience matters. If you’re testing something that only reaches a few hundred people, it’s really not worth the effort. You can kinda make it worth the effort, and here’s point #2, if the length of the test is long enough to give you meaningful data.
Finally, and the most important consideration, when you get the results back, what are you actually going to change, and is that change going to make a significant difference?
Here’s a real-life example of a company I know well, and we’ll spare names to protect the innocent. They decided to run tons of tests on their website to see if they could improve their conversion rates.
While the length of the tests were good, the website didn’t have enough volume for any of the test results to produce any meaningful results. The audience was too small.
And here’s the part that really stung for them. They tested so many different variables they ended up losing their cohesive feel, their story, their messaging, and website conversion dropped.
Test the stuff that can really make a difference. The rest – well maybe try testing NOT testing everything for a while.