Digital marketers emphasise the importance of A/B testing to improve conversation rates. However, what many experts fail to mention is the A/B testing essential to optimising your website and its performance as an eCommerce. Not all A/B testing is necessary, and testing everything on your site can waste time and money.
The best way to know what you need to test? Set goals. You cannot prove a negative. You need to understand what works on your website by setting measurable goals. Quantifiable results will indicate the areas of your website that require improvement.
So what are some of the most important things that need to be A/B tested on your website? The following are our top 19 picks.
Typography
Typography is proven to affect the performance of your marketing, but testing each font is not practical. Thankfully, there is a tremendous amount of research available on the effect of typefaces on different audiences.
1. Serif vs. Sans Serif
Serif typefaces are accented with various widths for each line in a character and have flourishes (for example, Times New Roman). However, sans serif typefaces are plain with a consistent width (like Arial).
Serif has been commonly used in print, as well as used in multiple languages that use Latin script. The individual letters are distinctive and easier for us to process as we read a book or newspaper, for example.
The serif typeface, Georgia, is the most commonly used typeface on the internet. It is easy to read and familiar. Computer monitors generally have a resolution of 100 pixels, compared with 1,000 for print. The clear font makes it easier to see characters as they lack detail and flourishes.
Sans is the best typeface for your website. Start with this simple base, and you won’t need to A/B test the font. You can have confidence choosing the sans font that suits your brand.
2. Colors
Black (dark) text on a white (light) background is the traditional color scheme and one that most people find the most comfortable to view.
For CTAs and other impactful text elements, you can choose the colors that suit your design, but remember, if it stands out, it attracts clicks.
3. Font Size
Tahoma is believed to be the most legible font at 10 points, Verdana and Courier at 12, and Arial at 14 points.
You do need to A/B test the differences in user engagement and click-throughs according to the size of the font. In most cases, because people are engaging with eCommerce sites on their devices, larger fonts are more popular, but this is not always the case and testing will help you to determine the best choice for your site.
4. Typefaces
Typeface testing should be very focused. Choose a few that are good for your brand and image and test those. It is easy to get too obsessive with typeface, so allow your design team to make choices and run a few tests to make a final decision.
Call to Action
The call to action (CTA) is an essential element on any landing page. You need to combine and test all the elements of your CTA; position, color and text. The conversion rates can be changed dramatically simply by changing any one of these elements.
5. Position
Designers will tell you that the CTA must go above the fold on the home page. However, not according to website traffic. Locating your CTA below the fold could reportedly increase your conversion rate by 304 percent. Test your CTA location above the fold, below the fold, in the middle/left/right of the page in relation to text elements.
6. Color
Color features heavily in testing. There are theories that a red CTA button will outcompete a green one with a 21 percent increase in conversions. Other tests show that a green “add to cart” button attracts 35.81 percent more sales than a blue one. It is important to test colors and incorporate colour theory and psychology into your testing.
Contrasting colors are distinctive and stand out against other page elements, and overall, this theory seems to work best in A/B testing. You should test this element on a regular basis and make changes as your testing indicates what is most effective for your conversion rates.
7. Text
Your CTA text is one of the first things that people see on a landing page. You need to test it using various lengths, pronouns, power words, and verbs.
At different times, different CTA texts will resonate with consumers. From ‘Try Me’, to ‘Learn More’ to ‘Sign Up Now’, there are any number of phrases that will motivate people to action.
A/B Test: Pricing Schemes
This section examines how you set free trials and money-back guarantees in accordance with the price point for your product/software.
8. Freemium vs. Free Trial vs. Money Back Guarantees
Product demos are important. Online vendors usually offer one of three free trial models:
- a basic freemium product with limited features that can be used without payment
- a time-sensitive free trial that includes full functionality
- a time-sensitive money-back guarantee for dissatisfied customers
You will need to test different trial types for a few months to establish which model works best for your conversion rate optimization.
9. Free Trial Length
Knowing how long to offer your free trial is another aspect of A/B testing that needs to be considered with your sales team.
For some SaaS providers, they have found that shorter trials, like 7 or 14-day trials motivate sales far more than 30-day trials, in which time some consumers realize they do not use the service as they expected. Test what offers are most attractive to your potential customers.
10. Pricing Each Plan
Trying different prices for plans in conjunction with your sales team is an important aspect of any product launch. You need to understand your competitors, the average price point and how you might price your products or services within the range expected by your target audience. Some brands can inflate prices because of their reputation, others aim for below market average to attract a bigger market share. You need to test your price points and how you can use psychology (such as 99 cents being more attractive than a 1 dollar price point) to drive sales.
You can also offer a decoy pricing model. By offering a much higher price for the entry-level compared with a mid-tier option, users are likely to spend more without investigating the offer fully.
A/B Test: Landing Page Copywriting
One of the main parts of great content is great copy. It takes time to get the written content right and it should be tested for its persuasive powers just like other A/B testing of your site.
11. Short-Form or Long-Form Copy
Studies have shown that the power of long-form copy is not lost. Long-form copy produced 7.6 percent more leads (and better-quality leads) than short-form copy for a SaaS site. However, a gym chain saw 11 percent more conversions with shorter copy.
You need to test your copy for your audience. Some business types do not need copy, like clothing retailers, while others require explanation so prospective customers know what they are buying, like SaaS providers.
12. Video vs. Text Sales Pages
Video copy is a gamble. Again, you need to test it for your market. However, as the internet evolves, video content is becoming increasingly popular. You can test video content, but be sure that your load speeds are ready for the challenge or even the test will prove unreliable as people leave the site because of issues with tech rather than a preference for content.
Video content does work extremely well for merchants who want to show rather than sell their products. Demo videos, fashion shows or interviews can be a great way to provide informative content.
13. Actual Text
You need to test different versions of your text-based copy. You should publish, then have your content team work on updating and improving the content as your business grows. You will never have the perfect word combination, but you must keep working on it to find the content that will resonate with your target audience. It should be in your brand voice and motivate traffic to complete sales.
General A/B Tests for Your Website
There are many things that need to be A/B tested on your website that fall under sales funnels, website design/structure, security and more.
14. Number of Columns
Multiple-column landing pages are less appealing than a single splash page.
Conversion rates improved by 680.6 percent when one company changed its two-column pricing page to a single-column page. Landing pages should be clear, easy to ready and have comparable information easy to follow without traffic having to take further action.
15. Background Images and Patterns
Your landing page background has a subliminal effect on your traffic. Test your backgrounds using solid colors, images or video. Do not assume that just because it looks on a competitor’s website that it is actually what traffic wants to see.
16. Navigation Links
Navigational menus change form every few years. They come in and out of fashion as hamburgers, drop down, tabs and other forms of menu options. You need to test potion, style, color and text for the optimum choices.
17. Link Color
Test the link color to drive traffic from your blog to your website. While testing your external CTAs is not often a consideration, it is considered essential to the improvement of your conversions. If traffic does not notice or feel motivated by your CTA on a blog, they do not become traffic and customers. They are already qualified leads, so testing the CTA’s is important.
18. Contact Form Fields
To get contact/quote requests from your website the format of your contact form is critical to your conversion rate.
Test the number of fields (keep it to a minimum) and the types of fields (checkbox vs. drop-down) to elicit more form submissions.
People want things to be fast and easy and do not want to spend time reading your submission form, they want you to respond as an interested business ready to provide great customer service.
19. Fast Checkout Process
Numerous case studies have proven that single-step checkouts almost always convert significantly better than multi-page checkouts. Test a single-page check out and autofill options.
