Why You Should A/B Test Your Website
Testing your website is important. Testing helps you make better, more informed decisions on page design, content and conversion strategy. Testing provides you the insight you need to confirm how a website is performing. The results garnered from testing help you identify what may not be functioning as well as you may have thought.
A/B testing is one of the most effective forms of testing for companies who are looking to improve user experience, conversion rates, time on page and any other user metric you can name.
So What is A/B Testing?
Also known as split testing, A/B testing is a method of comparing two versions of a web page (A and B) to see which one performs better. Users are funneled into one version or the other (at random) when they visit the site. By tracking the way users interact with the page, you can determine which version (A or B) is most effective.
Why Should I A/B Test?
The goal of any website is to drive the visitor to perform an action — whether it’s to purchase a product or service, sign up for a newsletter, donate to a cause, click on a link, etc. A/B testing can be extremely useful in improving the customer experience and converting more of your existing traffic by discovering methodologies that keep them engaged with your website longer, thereby increasing the likelihood of them interacting with your content. The benefit to converting existing traffic is much more cost-effective than paying for acquiring new traffic. And optimizing your current conversion rates will increase your bottom line.
What Can I Test?
There’s really no limit to what you can test with the A/B method, but the items you do test should be closely associated with your goals. For example, if the goal of your landing page is to get visitors to sign up for your newsletter, your focus should be on the content and images on the page that are most closely related to the newsletter, such as the color of the button the visitor clicks on to sign up.
Here are a few things you should consider changing in your A/B tests:
- Body copy
- Call to action buttons (color and copy)
A/B testing doesn’t have to be limited to web pages. You can also A/B test off-site marketing initiatives, such as emails and advertisements by switching up headlines, landing pages, special offers and body copy.
Best Practices When Testing
1. Set goals early
To begin, make sure you set clear goals for the tests. Goals – such as increasing your clickthrough rates or increasing the number of newsletter signups – will help guide your A/B test by outlining exactly what you want your users to do. For many A/B testers, following the scientific method (yes, the one from your first elementary school science class) is particularly helpful. Make an initial observation of your visitors’ behaviors, develop a hypothesis and make predictions – after you do this, you’ll then conduct the experiment and evaluate your results based on the previously identified goals.
Also, it helps to know how many participants you will need to get the right amount of confidence in your testing.
2. Cookie your users
For on-site A/B testing, make sure that your users are cookied when they visit your site. This ensures that their browser will recognize that they are visiting the A test page and won’t direct them to the B test page if they refresh the page or click ‘back’.
3. Assign traffic weight accordingly
When A/B testing, you can set the percentage of users being funneled to page A vs. page B. You should determine that percentage based on the type of campaign you’re running. If you’re launching a brand new campaign from scratch, it makes sense to give equal weight to each landing page, as you do not know how either of the pages will perform without previous data. If you want to improve an existing page with A/B testing, you should give each new page a smaller percentage of traffic than the existing page, as you don’t want to risk the new pages performing poorly.
4. Make sure your tests are SEO-friendly
Google has four basic guidelines to help ensure that your next A/B test won’t affect your search engine rankings. Check them out here.
Case Study: ShareProgress’ Mobile Ad
ShareProgress, a company that helps progressive organizations grow their campaign base, was interested to find out how to best optimize their mobile ads on Facebook for their clients. They wanted to see if a text-based ad could outperform an ad with a graphic, so they set up an A/B test with a clear objective, hypothesis and testing parameters. They tested the two versions below – which do you think more users clicked on?
If you guessed Version A, you’re correct! The “Share the Petition” image was more directive and performed 101% better than the other graphic images.
Interested in A/B testing for your website? Contact Mike at firstname.lastname@example.org or call 410-732-6542 and ask for Mike.