5 Things to A/B Test on Your Website

Published Aug 09, 2018. Updated Jun 14, 2019.

5 Things to A/B Test on Your Website

Your website is the most important piece of marketing and sales material your business will ever have. It’s often the first impression of your company that visitors see, and it can single-handedly determine whether or not they choose to work with or buy from you.

For your website to be as effective as it can, you need to know what resonates with your audience and what doesn’t. The best way to do this is through A/B testing.

What is A/B testing?

A/B testing, also known as split testing, is when web designers and marketers compare two slightly different versions of the same piece of marketing material to determine which an audience responds to more readily.

Much like a school science experiment, it’s best to only test one change at once so you can easily see why one version works better than the other — if you test too many elements at once, you won’t know which one was responsible for the improved performance.

Here are some things we always recommend that our clients test before committing to a final version of their website.

1. Colors

Color plays a big role in how your visitors respond to your website. It can form 90% of a customer’s opinion of a brand in just 90 seconds and can be up to 85% of the reason people decide to buy from a company. That means it’s definitely worth testing so that you get it right.

Start with your website’s color schemes. Does your brand play better with a dark look or a bright one? One study found that light color schemes can boost your site’s performance by up to 1.3%, while dark color schemes can bring as much as a 2% boost in performance. Test this theory to see if it holds true for your audience.

You can also experiment with your brand colors. Remember, each color carries its own set of connotations, so think carefully about how you want your brand to feel instead of simply choosing colors at random. Is your company more serious and professional, or fun and hip? Is your staff mostly older or younger employees? What about your customer base?

color psychology

Next, test how your brand colors work in different areas of your website. For longform copy like blogs and case studies, it’s usually best to use a dark font on a light background, as this is what your visitors’ eyes are most used to reading. But for other areas, like sections of your service and home pages, you may find that a colored background with lighter text helps your message pop.

If you’d like to read more about how color affects your visitors’ experience, check out our previous blog, “How to Use Psychology in Your Web Design.”

read more on the blog: how to use psychology in your web design

2. Imagery

Visuals are an important part of any website because they have a huge impact on how your audience perceives your brand. Choose your imagery carefully and test at least two versions (though more is better) before settling on the one you’ll use.

First, test whether your site performs better with solid or image backgrounds. A solid background is simpler and usually makes text easier to read, while an image lends more personality to your brand and can be a good way to break up sections of your site.

Where you do use images, experiment with the subject matter. If your business is very people-oriented, you may find that images with people in them perform better, while if your business is more industrial you may find that people aren’t necessary in your images. If you’re a veterinarian, test whether your patients’ owners respond better to images of cats or images of dogs.

You may need to cycle through a few different versions before finding the combination that works best for your site, so keep testing until you find the magic formula.

3. Fonts

Fonts are something that most people never think to test, but they’re an important part of your brand presence. The font you choose tells your audience almost as much about your company as color does, so you need to choose carefully and test thoroughly.

First, decide whether you’ll use a serif or sans serif font. Serif fonts are typically seen as more mature and serious while sans-serif fonts are more modern and approachable. Sans serif fonts are also usually easier to read online, but you may find that your audience prefers a mix of the two.

Your font size also matters. You want your body text to be big enough that your viewers can read it comfortably even on mobile device, but not so large that they’re constantly having to scroll more. Your heading text should be large enough to stand out from the body text, but not so large that your readers feel like you’re shouting at them. This can take some adjustment to get right.

read more on the blog: 15+ Stats you should know when redesigning your website

4. Calls To Action

Your call to action is one of the most important parts of your website, and often requires the most testing to get right.

Start with its position on the page. Most web pages include a CTA below the fold — i.e., far enough down on the page that your visitor needs to scroll to see it. In fact, ContentVerve found that doing this could increase conversions by as much as 304%.

It’s also a good idea to test different variations on your CTA button itself. Marketers often recommend using a green button rather than a red one since visitors are conditioned to read red at “stop” (or “no”) and green as “go” (or “yes”). However, Hubspot found that switching to a red button from a green one brought in 21% higher conversions.

CTA buttons

Finally, how you word your CTA matters, too. On most website pages, a CTA that takes the soft sell approach usually outperforms a hard sell.

The most well-known example is this case study of an Obama campaign landing page, where swapping out the pushy “Sign Up” button for the more relaxed “Learn More” brought in 40.6% more conversions — which for the Obama campaign translated to an extra $60 million in donations.

Remember, though, that every audience is different. What works for one company may not work as well for yours, so be sure to test these elements before committing fully to one design.

5. Pricing

Including pricing online is standard for e-commerce and hospitality businesses. But if yours is more service-based, it can be hard to know whether you should include prices or not.

That’s where A/B testing comes in. Measure how well your website performs both with and without pricing to see which your customers and clients prefer. Including pricing often eliminates “tire kickers” who are simply looking for the lowest price rather than the best value, so you may find that including pricing brings fewer inquiries or leads, but the ones that do come through are a better fit for your company.

pricing on websites

If you find that your clientele prefers seeing your pricing upfront, you can still test how to structure that pricing. You may choose to price each service individually or bundle certain ones into a package. Maybe your audience purchases more when you show a higher price slashed in favor of the current one. Maybe a $1,000 service sells faster if it’s actually $997.99.

And if you offer a free trial, experiment with length and whether or not to require a credit card upfront.

We help you speak your customers’ language

Still not sure how to make your website resonate with your target audience? We can help. Our team of design and copy experts know just how to get to the root of your customers’ problems and show them that your company is the obvious solution.

Want to learn how we can help your company? Contact us to learn more.

Related posts

Start your digital transformation today

Ready for a slick new website, increased traffic and more converting customers? Tell us a little about your business and we’ll reach out to get your project underway.

Start your web design project today