A gentleman who had on his website a page with a button to register visitors, decided to do a so-called A/B split test, that is:
Written on A button: Register for free
Written on B button: Start now
Which writing generated the most registrations in your opinion and with what percentage?
Write your answer here: _______________
Then he decided to change colour:
A button colour: GREEN
B button colour: RED
Which colour has generated the most registrations in your opinion and with what percentage?
Write your answer here: _______________
Okay, let’s see the actual results:
The words “Start now” won with a +7%
The RED button won with a +34%
So, in a few hours that landing page has improved the conversion by more than 40%!
Not bad.
But the A/B split test is NOT something that only works online.
In fact, that’s one of the fundamentals of direct response marketing.
And it’s a practice as old as the hills, used already in the early ‘900 (!) by the American advertiser Claude Hopkins. At that time the web was not there, and he used coupons with different identification codes to see which offer or newspaper resulted in more sales.
Usually those who work in B2B don’t think much about A/B split tests, because the numbers involved are relatively small. And since the statistical value needs substancial numbers, which take time to reach in B2B, we often end up saying, “Well, it doesn’t matter…”.
No, actually it DOES matter.
Sometimes even a small change to just one aspect of your offer can make a big difference in terms of conversions.
I close then with a suggestion for any reflections related to your business.
Make an inventory of all the things that you could test in your “message”, intended as any element with which you deal with customers and which customers must respond to by doing a certain action.
They can be, for example:
– ads to click on Google;
– the colour of the product to be chosen on the ecommerce website;
– the title of a registration page;
– one font compared to another;
– written in capital letters rather than in italics;
– even two offers slightly different in content;
– etc.
And then start testing these elements one by one.
(WARNING
Don’t think that since THAT words and the red button in the example worked better, they will work better for you too. The method is right, and it must be adopted. But the results MUST NOT be copied because for you, for your product, in your market, with your customers, they could be different).
Head. Measure. Refine.
Re-head. Re-measurement. Re-fine.
And so on…
Little by little, if you’re just a little persistent, your conversions will improve.
And because each conversion costs money, you’ll get better results at a lower cost.
P.S.:
if you have not yet participated in our 2016 Survey on Marketing in Italy, please give your contribution here at this link, thank you!
