A/B testing is the most effective way to improve conversion rate
Generating more sales, leads, or subscribers from your website traffic is no easy feat. Once a user hits your website, there is still a long way to go to convince that user to take the action you want them to take. Getting a higher conversion rate from these visitors is a key step in optimizing your digital sales funnel. One of the most effective ways to improve your website conversion rate is A/B testing.
At its core, A/B testing is very simple. A/B testing is a way to compare two versions of something to determine which version performs better. On websites, you can A/B test anything from layout designs to pricing. Basic A/B testing works by sending half of your website traffic to one version and the other half to the alternative version, then measuring which version outperformed the other. It is important to make sure the traffic split is random so external factors are not skewing the results.
A/B testing can significantly move the needle for your website conversion rate. Updating your website based on results of A/B testing leads to an improved experience for your site visitors. Often, your intuition on what your traffic wants does not hold up in reality. Take the A/B test below as an example.
This A/B test of a lead gen form illustrates how wrong our assumptions can be. Most people would assume adding an eTrust image would improve form completions. After all, it makes the potential leads feel like there is an added layer of security!
In reality, this is not the case. The results of this test ended with Version B leading to a 12.6% rise in form completions. In other words, excluding the certified privacy image actually improved the performance of the form! As it turns out, images of this type are usually associated with payments and credit card information. As a result, many users assumed that they were about to pay for something and decided not to complete the form. This is a perfect example of why A/B testing is important! The data often reveals information that our instinct does not.
Common A/B Testing Use Cases
Now that we have established the value of A/B testing, let’s take a look at some common A/B tests that can seriously improve your website conversion funnel!
The psychological effect of the basic style elements of your website is something that is hard to predict but easy to measure with A/B testing. Some of the most basic A/B testing can be done without redesigning your webpage, rewriting copy, or adjusting your website flow.
Take color for example. For long-form text, you generally should have black text with a white or very light background. But for smaller text elements on your page, experimenting with color can be a simple experiment with a high potential upside. For color-related A/B tests, your call to action (CTA) is a great place to start. You want your CTA to stand out on your page and be the clear action that the user is encouraged to take. Take a look at the AnalyticsBox navigation bar, for example. Is there any doubt which link stands out the most and which action we want users to take?
Try out a few color combinations and see which one performs the best for your CTA! You can also experiment with colors for the navigation bar, secondary page elements, and even your logo!
Another aesthetic element to test is text font. This goes for both the font itself and font size. It is generally good to have a consistent font across your website, and it is usually a fairly simple experiment to play around with a few fonts to see which, if any, outperforms the others. If you don’t know which fonts you want to try, this guide can be a good kickoff point!
Landing Pages & Calls to Action
In the marketing industry, a landing page is generally considered a standalone web page created specifically for a marketing or advertising campaign. It is designed as an “entrance” to your website after a visitor clicks on a link from social media, Google, email, etc. It usually has a singular purpose, which is to get the visitor to take a particular action. Landing pages don’t have to be standalone pages, and even your homepage can be used as a landing page for generic campaigns.
A call to action (CTA) is a written directive with the goal of encouraging users to engage in a certain behavior. On your website, common calls to action include making a purchase, scheduling a meeting, and providing contact information. The primary CTA is the main action that you want your visitors to do.
Almost every element of a landing page has the potential to be a fruitful A/B test. For example, experimenting with options and labels on your site’s navigation bar can be a low-hanging fruit. Sometimes, websites often have too many options on their main menu, and users are either overwhelmed or end up taking actions that do not drive the conversion funnel. Keep it simple. Don’t believe us? Look at how basic Google’s primary landing page is.
The page background is a simple A/B test as well. Experiment with images, color patterns, and even footage in an attempt to move the needle. For AnalyticsBox, we ended up determining that this stock footage performed very well for us.
The call to action is usually the most important A/B test you can run on your landing pages. As noted above, the link color and text are relatively straightforward experiments. But while they may be simple, they can make a huge impact. For example, Humana, a health insurance provider, wanted to raise their clickthrough rate for a particular banner on their website. They performed an A/B test to compare their existing CTA to a new version, and experimented with a few elements. One of the elements tested was changing the text of the CTA from “Shop Medicare Plans” to “Get Started Now”. This particular test alone drove an astounding 192% increase in clickthrough rate.
If your testing is more advanced, you can perform more complex tests that focus on overall page structure as well.
- Should you have long-form or short-form copy on your landing page?
- Should you include images and videos or only include text?
- Should the existing copy be replaced by a new text variation?
- Should the CTA be near the top of the page or should the user have to scroll to find it?
These are just a few questions you should ask yourself when thinking about experimenting with structural changes to your page layout!
Conversion Funnel & Pricing
Your conversion funnel is the journey a consumer takes from initial interaction through your desired action. For the sake of this article, we will focus on the conversion funnel elements within your own site. Every stage of the funnel can be optimized towards achieving the targeted action.
For businesses whose targeted conversion is lead generation, a contact form on their website can be an important element to A/B test. The most common contact form test is the number of fields being collected. Usually, less is more when it comes to the number of questions in a contact form. At Neil Patel, they A/B tested their contact form and ended up seeing a 26% increase in conversions after reducing the number of fields from 4 to 3.
For e-commerce sites that are selling products, product descriptions & images are simplistic experiments to perform. What changes can be made to the text describing a given product to encourage a higher purchase rate? Do certain images (or order of images) lead to more conversions than others? If someone is already looking at a product details page on your website, they are demonstrating purchase intent behavior. Reducing lost sales at this stage of the funnel is often very achievable.
Another consideration should be the simplicity of the on-site funnel. Think about how many existing stages are required on your site from an entrance to a conversion? Generally, the fewer the clicks, the better the conversion rate! For online sales, one way to decrease complexity is by reducing the number of pages on your site’s cart path. You can experiment between a single-step vs. multi-step checkout flow to see if capturing all necessary information within a single page increases conversion rate.
The classic experiment that long precedes digital commerce is pricing. Price sensitivity varies wildly across companies and industries. The only way to know for sure what the price sensitivity is for your own business is by testing for yourself! This article is a great place to start if you want to begin A/B testing different price points. With pricing, you can also test the impact of different promotional tools such as coupons, discount rates, and free trial periods.
Advanced A/B Testing
Straightforward A/B testing is a powerful tool on its own, but we can go further. Experimenting with a more sophisticated approach may lead to more insightful and profitable findings.
One way to get the most out of A/B testing is by experimenting with more variations of the same test. In other words, instead of A/B testing, you can try out A/B/C or A/B/C/D testing! If you are a dentist looking to get more business, maybe you want to try out a few different CTAs on your website to schedule an appointment. Instead of just testing the CTA test of “Schedule an Appointment” vs. “Book Now”, maybe you additionally want to include “Book Here” and “Sign Up”.
Many companies that take A/B testing seriously don’t only look at individual tests in isolation. Just because a variation in an experiment performs the best in a standalone test does not mean that it performs the best when taking into account the results of other experiments. Testing combinations of variations often brings about results that are different from the results if you consider each test individually. The reason for this is that the results of one test might prime a user to expect an experience in the next. If you are performing tests all across your site, you should consider how different unique experiments impact your multi-component testing.
Take into consideration the example below. Most people would guess that Version B should perform better. The headline copy and sub-head are more clear and crisp. The results tell a different story; Version A increased leads by 115%. The reason is this experiment was for a landing page of a specific paid marketing campaign with particular messaging. This messaging was better aligned with the text in Version A. Therefore, while Version B might seem better as a standalone landing page, Version A was better when taking into account other components of the funnel.
The farthest level of depth one can go with A/B testing is mass customization. You should consider this option when you determine that the results of an experiment vary in a meaningful way depending on the user. In other words, sometimes you are best off providing different experiences for dissimilar users. If you are a two-sided marketplace like Shutterstock or Amazon, maybe you want customers and providers to have different landing pages for a marketing campaign. After all, your offerings to those two segments are entirely distinct!
It can be more nuanced than that as well: sometimes users behave differently based on if they are a new vs. returning visitor or how they found your website. Someone who navigated to your site from social media might behave differently (and want different things) than someone who found your site from Google. You won’t know for sure until you experiment and dig into the data!
Think about what you want your first experiments to be. The best places to start are elements of your site that are both simple to test and critical to your conversion funnel. CTAs generally meet both of those qualifications, and make excellent first experiments! Once you have a general plan, you need to decide how you want to conduct these tests.
There are many tools and platforms from which you can choose if you want to begin A/B testing your website. One of the most popular platforms in this space is Optimizely. Optimizely is a very powerful tool with features such as multi-page testing and advanced segmentation. Optimizely can get pricey, and you should expect to spend at least five figures for even lower-level plans.
A cheaper option is VWO. They have a good portion of the same functionality of Optimizely, but at a much lower price point. Crazy Egg and Zoho PageSense are worth checking out as well! If you want to truly automate your A/B testing and take a backseat, Smart Websites from AnalyticsBox allow you to turn on automated experiments that optimize your site without any manual intervention!
Once you have your vision and your tools, you are well on your way to optimizing your website through A/B testing! Need some more testing ideas? Click here to get a few more ideas related to website page flow. If you are interested in experimenting with mobile versions of your site, give this short article a read to learn more about how to cater to your mobile traffic.