What is A/B Testing and How Can It Help Your Site?

I once heard A/B testing described as the internet equivalent of sex in high school: everyone says they are doing it, but few really are.  A/B testing is simply the systematic process of testing various elements of your website to understand how subtle changes affect user behavior.  By employing testing strategies, you can increase your conversions, improve open rates on email marketing campaigns, and drive visitor focus to a specific area of your website (i.e. drawing their eye to your email sign-up form).

So How Do I Know If I Need to be A/B Testing?

That’s a great question, and the answer is really easy: you do.

Let me explain. If you are anything less than 100% satisfied with the performance of all your efforts online, then testing needs to be part of your toolkit. Do you wish more people visited your site?  Bought your products?  Increased your profit per customer?  Clicked on your ads?

Of course you do.

Marketers and web professionals are, by nature, creative people.  Inspiration strikes and projects take shape, websites are put on the internet, products are created.  That’s the beauty of the entrepreneurial spirit – the speed at which life can be breathed into a product that has the potential to solve real problems, attract customers, and positively influence both the world and your bank account is staggering.

But when reality sets in and things aren’t moving as well as you’d like, or they are moving but you want to increaa-b test on writeworkse the capture of data, revenue, and attention, the application of more systematic tools such as A/B testing can dramatically accelerate the rate at which you reach those goals.

A Primer on Testing

So A/B testing is simply a way to say that you’re testing a single variable.  Let me explain.  When a visitor lands on your website, she sees your opt-in box.  It is blue.  In an A/B test scenario, when a second visitor lands on your site, everything is exactly the same except for the color of your opt-in – now it’s red.  Using simple software tool such as Google Optimizer, you repeat this experiment over the course of a few days and several hundred visitors to your site.

At the end of the test, you look at the results and you find out that only 3% of visitors signed up when your opt-in was blue, but 5% signed up if it was red.  Right now you may be thinking, 3% – 5%.  So what? Well, let’s assume your site receives 1000 visitors per day.  That adds up to an additional 7,300 people that will join your mailing list this year, thanks to one simple change.

Over time, you may continue to test additional variables – what if you tested out two different headlines, or incentives for signing up?  Or what if you moved the position of the box to a different place on your website?  Either of these factors could have an increased impact on your conversion rates, allowing you to improve on them over time.

More sophisticated testing – called multivariate testing – allows you to test multiple variables at the same time such as different combinations of color and headlines for example.  This can put you closer to your optimal conversions even more quickly, but are more complex to manage. If you need multivariate testing, I would recommend hiring an expert to conduct this for you.  Plus the results that you’ll see when employing testing for the first time, those incremental improvements in the ability to persuade website visitors to follow your calls to action, are often enough.  When dipping a toe into the pool of testing, start by looking at single variables to improve.

How Do I Know What to Test?

A common question from website owners is how to know which variable to test.  The following list will get you started in terms of looking at different features that you might want to prioritize to quickly improve bottom-line effecting results.

Opt-in to your email list If you operate a website and are collecting emails, that’s a great thing.  An email list is your most valuable asset, in the instance of your site getting de-indexed, issues with your hosting provider, or any other problems, an email list allows you to still immediately reach your customers.  So look at variables that could be affecting your opt-ins, such as placement of the box, color of the box, what information you’re asking for, the language you’re using, and more.

Email headlines If you do a lot of email marketing, consider conducting an A/B test on your headline with 10% of your list.  Write two versions of the headline and send those out.  Then compare conversion rates, and use the headline with the higher open rate on the remaining 90% of your list.

Call to actions What’s the big call to action on your page?  For example, if you operate a site that sells vitamins, your call to action is to get people to buy your product.  So take a look at all the ways you invite them buy.  Maybe it’s through the placement of a “shop” link in the menu bar.  Could you position that differently?  Turn it a color that makes it “pop” more?  If you are selling the same product through a squeeze page, consider looking at factors as simple as the difference between “Buy Now!” and “Click Here to Order!”  Isolate the individual factors that communicate your call to actions and refine them for maximum conversions.

I’m convinced – so how do I get started?

In a future post, we’ll look at the technological aspects of A/B testing.  But if you’re ready to dive in and want an easy to use and low cost (free) solution, Google offers a tool called Google Optimizer and recently launched Google Content Experiments (see video below) that you can start the process within minutes.

If you’re interested in learning more, or seeing specific case studies of A/B testing in action, check-out the following recommended reading: