Leveraging A/B Tests to Optimize Your Website

If you manage a website chances are its not performing at full potential. Don’t take it personally; most sites aren’t. Its not that most sites are managed poorly, it’s just that there are always improvements to be made. The best way to improve site elements and ensure that you and your users are getting the most out of your site is performing a simple A/B test. Tools like Optimizely and Visual Website Optimizer make A/B testing very easy and straightforward. 

You can test the effectiveness of literally anything on your site, but some site elements are more important in terms of driving engagement and conversions than others. Below I’ll highlight a few of those elements.

1. Calls To Action (CTA’s). A CTA is a site feature that prompts the user to perform a desired behavior. This can be reading, purchasing, signing up, downloading, or sharing. Testing CTA’s can really move the needle, and a lot of times these can be very small, seemingly minor changes.

  • Change the text on the CTA. For example, ‘Buy now’, ‘Purchase’, ‘Proceed to Checkout’, and ‘Add to Cart’ are all variations you might want to test for an ecommerce site.
  • Test whether text, icons, or text + icons are more effective at converting users.
  • Test different color and shape permutations. This may seem insignificant, but there is a big difference between the effectiveness of a gray CTA and a red CTA.

2. Funnels. The funnel is made up of the steps in a conversion process. If a user has to go through 3 pages of filling out information to complete a conversion, then each page is a step in the conversion funnel. Because users are very sensitive to giving a website personal information, small changes in the funnel process can yield huge results.

  • Try removing anything distracting from the task at hand (i.e. promotions, offers, non-essential information, etc.). More often than not, simpler is better
  • Test the number of pages. Does stuffing more information into one or two pages perform better than spreading that information across several pages?
  • Remove any navigation to other pages from the funnel. This is often referred to as a ‘captive checkout’ and is utilized by Amazon among other sites.

3. Forms. On the effort scale for internet users, filling out any kind of form is a 12 on a scale of 1-10. You might as well be asking the user to perform hard physical labor. Try to make the process of filling out a form as easy as possible.

  • Test the length. Try to stick to required information only. This might mean taking stock of what you’re really looking to get from a sign up form. Do you really need the person’s physical address or is that just something you’d like to have? What % increase in registrations would make not collecting some additional information worth it?
  • Address people’s fear of what you might do with their information. If you can guarantee that their personal info won’t be used for spam, say so. It could greatly increase people’s willingness fill your form out.
  • Try asking for different information in your form fields where possible. For example ‘business email’ vs. ‘email’ or ‘work phone’ vs. ‘cell phone’.

4. Social Media. Social presence is becoming more important for everyone. The amount of impressions to which a single social share translates can really go far in increasing brand awareness. Test the following things to increase those shares:

  • Change the size and placement of the social icons. Bigger can be, but doesn’t always equal, better.
  • Create and test your own social media icons to fit the look, feel, and branding of your site. Users value consistency so you might find that the stock images for the social networks don’t jive well with your website experience.

Developing a process for continual testing and improvement will ensure that your finger is on the pulse of your user base, delivering to them the best content for achieving desired site outcomes. You may find that key site elements are already performing quite well compared to other variations, but you wont know for sure unless you test.

There are 0 Comments

Leave a comment