There are a number components involved in effective A/B split testing. Mastering these can give any business a boast and help you get the kind of results you need to increase your subscribers and sales so you can generate more revenue and grow your brand.
A/B testing, also referred to as split testing, involves testing two similar marketing elements. For example, you might wish to test the headline on a sales letter for your product to make it more profitable, or choose between two email subject lines to see which one works best (generates the most opens, most business).
Once you determine which is best, you can then improve your conversion rate for that particular page or email. You can then publish the best one, and leave as is, or split test again and again until you are completely clear as to which one works best.
Part Art, Part Science
A/B split testing is part art, part science. You often have to take an educated guess as to what motivates your audience in order to boost your sales. Then you will confirm, or not, whether your guess was correct.
The analytics tools available for split testing, such as Google Analytics and Experiments, will show you the numbers of visitors and clicks. Your sales reports will show you if your sales have gone up, down, or remained the same.
The Components of a A/B Split Test
The essential components of a split test are as follows:
- Decide on one element to change, such as the headline.
- Create two different versions of the page, and call them A and B.
- Divide your traffic between the two pages.
- Evaluate what the traffic does. Do they click, subscribe, buy?
- Run the test for as long as it takes until you get a clear winner. Use that as your “control” page.
- Go back to Step 1 and run another test using a headline (technically headline C).
Note: If neither A nor B are a clear winner, choose the version you like best, and run a second test against version until you get a winner.
How Long Should the Test Be Conducted?
The answer is that you should run the test as long as it takes to get a clear winner within what is termed a statistically significant number of site visitors. For example, send 100 people to each site. If page A has a 5% conversion rate (that is, 5 out of 100 people make a purchase or subscribe) and page B has a 3% conversion rate (3 out of 100 people take action), then A is clearly the winner.
What Should a Marketer Split Test?
We have already mentioned headline in sales letters and subject lines for emails. But there are many other elements you can test as well in an effort to get more clicks, subscribes and sales. For example, changing the color of the “Buy Now” button from red to green or blue could make a huge difference in conversions. You won’t know until you test it. Anyone who remembers Amazon’s old color scheme can rest assured that they split test their website over and over again until they got the current blue and gold one.
In most cases, marketers will be testing sales landing pages, with the goal of trying to increase sales. In this case, adding certain elements to a page might result in higher conversions. Some items to add would include:
- High-quality graphics
- A money-back guarantee certificate near the sales button
- E-commerce trust symbols to the site
- A video sales letter added to the page, with a Buy Now button underneath, and so on.
If an element raises the response rate, then keep it, use the new version as the “control”, and split test a new element.
It can take time, but A/B testing can significantly improve all aspects of your business, so if you are not already doing it, now would be a perfect time to start.