A/B Testing is Ruining Your Creativity (And You’re Letting It)
Let’s talk about split testing. The quiet killer of bold ideas.
A/B testing was once the saviour of indecisive marketers. It still has its place — but somewhere along the way we started using it as a crutch. Or worse, a veto machine. I’ve seen entire campaigns diluted into beige oblivion because button colour B outperformed button colour A by 3.2%. You know what else outperforms consistently? Vanilla.
Brands are hesitating to back a strong point of view because the data hasn’t 'proven the concept' yet. And so we stick with what tips the needle slightly today, without considering what moves the brand tomorrow. I once saw a headline rewritten ten times because the original – smart, punchy copy – didn’t get enough clicks in the first hour. The final version? A Frankenstein of keywords and fake urgency. It worked, technically. But if that’s the bar, we may as well write by committee and call it a day.
True brand building isn’t about shaving decimals off bounce rates. It’s about saying something only you can say, in a way that sticks. Don’t scrap the A/B test entirely, but stop using it to dodge conviction. Sometimes the weird idea is the right one — and it needs more than 48 hours to prove its worth. New Zealand’s best campaigns took years to bloom. Let’s give the next one a shot at sunlight.
A/B testing was once the saviour of indecisive marketers. It still has its place — but somewhere along the way we started using it as a crutch. Or worse, a veto machine. I’ve seen entire campaigns diluted into beige oblivion because button colour B outperformed button colour A by 3.2%. You know what else outperforms consistently? Vanilla.
Brands are hesitating to back a strong point of view because the data hasn’t 'proven the concept' yet. And so we stick with what tips the needle slightly today, without considering what moves the brand tomorrow. I once saw a headline rewritten ten times because the original – smart, punchy copy – didn’t get enough clicks in the first hour. The final version? A Frankenstein of keywords and fake urgency. It worked, technically. But if that’s the bar, we may as well write by committee and call it a day.
True brand building isn’t about shaving decimals off bounce rates. It’s about saying something only you can say, in a way that sticks. Don’t scrap the A/B test entirely, but stop using it to dodge conviction. Sometimes the weird idea is the right one — and it needs more than 48 hours to prove its worth. New Zealand’s best campaigns took years to bloom. Let’s give the next one a shot at sunlight.