What’s all this buzz about A/B testing, and what does it have to do with my business?
That’s a fair question, actually two fair questions.
Let’s take them one at a time.
A/B testing is a way for businesses to test the effectiveness of new elements on their web pages before publishing them.
Whatever the change — say, a new way for an e-commerce site to move buyers through the checkout process — you channel half of your site visitors to the page without the change (the “control” group), and the other half to the page with the redesigned element.
You measure which one works better — in the case of the e-commerce site, this might be something like how many visitors click on a “Proceed to Checkout” button.
Said differently, A/B testing is all about SEO and finding a way to better optimise your website or your clients’ websites.
What does this have to do with your business?
If you do it right, A/B testing will boost your conversion rates and your profits, assuming that profits are what your site is about.
If you’re part of a marketing agency, A/B testing will do the same for the companies you market, enhancing your brand and increasing your credibility.
Can We Talk?
Sam owns a website hosting business. Sam is smart, so he makes sure his site moves visitors to some well-placed calls to action, such as a phone number, an online form, or a “put me on your mailing list” button.
One morning in the conference room, Samantha (Sam’s marketing guru) is hot to add a chat feature to the site.
But Bob (the ever-cautious CFO) thinks this is too intrusive and will turn people off.
Sam contacts a reliable SEO pro who runs an A/B test for him.
Turns out Samantha was right and Bob got it wrong.
A significant number of visitors directed to the new feature jump at the chance to chat their way to smart answers in real time, and the ones who do are more likely to sign on the dotted line.
The Devil in the Details
So, now that you’ve got the basic idea, it’s time to start A/B testing and jack up those conversion rates, right? Not so fast.
Like any good experiment, an A/B test has to be thoughtfully designed and well-executed.
Do it wrong or misinterpret the results, and you could do more harm than good.
Getting it Right
Here they are, along with some sound advice:
- Sift through the Hype and Keep Expectations Realistic: To read some of the case studies, you’d think A/B tests always work and always push conversion rates into the upper stratosphere. Not so. The fact is that about 7 out of 8 tests have no significant impact on conversion. Success is more about doing multiple tests over time and gradual change.
- Don’t be a Conclusion-Jumper: Your results might look great, but they could be preliminary. It could be that your sample was too small, for example. You have to determine if the results are statistically significant before declaring victory.
- Trick or Treat: A/B testing at its core is about understanding your audience and what motivates them by doing some smart conversion research. There are no quick fixes and no bags of tricks which will turn a so-so website into a sales behemoth.
A/B testing can increase conversion rates for your clients, but it’s not a quick fix.
It has to be done thoughtfully by experienced professionals who understand the science and can separate hype from hard evidence.
If you work with these kinds of pros and keep your expectations realistic, A/B testing will enhance user experience and add to the bottom line.