I've noticed what seems to be a trend: more and more clients are accepting a new - and dangerous - definition of testing marketing campaigns. We have to stop it, or we could lose a powerful tool.
The scenario looks like this: you think something needs to be tested. You assume that the test is to learn something which both you and the client don't know is certain yet but may have a hypothesis about.
Then the client tells you that he or she expects the test to have a positive ROI. You ask, 'but wait a minute, this is just a test, we don't know if it's
going to work or not.' The client replies; 'Sorry, no guarantee, no test.'
This is bad practice that should be addressed head on. Pressed with an ongoing requirement to justify every penny, coupled with the general culture
of fear that seems to be a modus operandi in so many companies, clients eems to be ever more sensitive to risk.
All testing is a bit of a risky exercise but its aim is to produce insight and not immediate sales. Insight, that will be priceless in the long term and will help clients avoid wasting their advertising budgets.
While I was working on Tesco.com's email programme years ago we tested 14 different subject lines alone! No wonder that they understand so well what works and what doesn't for their customers.
I wish more clients out there realised the value of testing lies in the insight it provides and not immediate ROI.