Many digital advertising agencies spend a great deal of time targeting the right keywords and ads for their pay-per-click (PPC) campaigns. But once your campaign matures, the real value comes from getting more people to convert on your landing pages.
High converting landing pages should be the focus of your digital advertising efforts because they have a direct effect on conversions. But if you’re not displaying landing pages that truly resonate with your audience, your site isn’t living up to its potential. Split testing helps you make educated decisions about your landing pages.
First Things First: The Components of a PPC Campaign
There are three basic parts of any PPC campaign, and they are:
- The keyword or placement you’ll bid on. These are the terms or sites you think potential buyers are using to find your products or services.
- The ad you’ll serve when a user searches your keyword.
- Once a person clicks on your ad, they’ll be taken to a landing page. You know how a landing page is supposed to work: It encourages the visitor to take action, so they convert to a lead.
What Is A/B Split Testing, and Why Do You Need It?
As a concept, A/B split testing is simple: You test two versions of a landing page simultaneously to see which gets the most people to convert at the best price.
Why A/B Test?
As a general rule, the easier it is for a prospect to understand what you want them to do, the more likely they are to do it. When someone “lands” on your landing page, they need to understand the value of it right away. They won’t tolerate confusion. The goal of split testing is getting them to that point. You make careful changes to the user experience and collect data on audience behavior after each change, so you can pinpoint which had an effect.
Find out which landing pages resonate with your visitors. Test your layout and navigation, so you can optimize your pages for users. When you optimize your site based on split testing, you prevent valuable leads from going somewhere else because they’re uncomfortable or confused by the design, copy, or function of your site.
Types of Testing
As we alluded to earlier, there are two types of testing you can perform. A/B split testing and multi-variate testing. Here’s the basic breakdown:
A/B Split Testing
What It Is
A/B testing simply tests one variant against another.
When to Use It
You’ll want to use A/B testing when there is one element on your page that you need to test. A/B testing can also be better when your site has a small amount of relative traffic.
Example of A/B Testing
For example, on one landing page you have a green call-to action button, and on another you have a red one. They’re in exactly the same place on the page, so the only difference is color. If your landing-page visitors click on the green button more than the red, you know the color of the button made a difference, and you optimize your site with green call-to-action buttons.
What It Is
This type of testing works with combinations of variables.
When to Use It
This testing is best used when comparing two pages with multiple design elements that differ from one another. I usually use multi-variate testing when I’ve tested the larger opportunities via A/B testing. Just be prepared for the tests to take longer, since you’ll need more data to achieve a statistically significant result.
Example of Multi-Variate Testing
For example, you use a certain image and headline in combination with one another on a landing page, and you use a different combination on another page. This type of testing makes it harder to pinpoint exactly what does and doesn’t resonate with your visitors, but it can be a great way to determine the impact of an overall design on your audience.
What Should You Test on Your Landing Page?
So which components of your landing page could make a difference when it comes to conversions? Here are some of the most influential parts of the page you might choose to test:
CTAs (Calls-to-Action) – Are you offering a free download, a paid or free trial, or a certain percentage off your product? Try testing different CTAs to see which ones convert better. It could be the difference between a button saying, “Download Now” vs. “Get Your Free Guide.”
Imagery – A best practice is to gear your imagery toward your target audience. If your audience falls into an older demographic, try testing pictures of older people and see which ages produce better click-through or simply more engagement. Perhaps images of people who are 40 to 50 years old perform better than images of people who are 60 and up because even though your audience is 60 and older, they think of themselves as young-at-heart.
Headlines – The most important job of a good headline? Make your product’s or service’s unique selling points (USPs) clear. You can do this in a catchy, trendy way or in a more straightforward way. Try testing those styles against each other!
Understanding When You Have a Winner – Statistical Significance
Before you make any decisions about which landing page is converting better, you’ll need to establish your measure of statistical significance.
Statistical significance is the amount of data you need to collect to ensure your test is determining a clear winner between your two landing pages and your results aren’t just due to random chance. As a general rule, you should have a good idea of statistical significance after about 100 conversions or so. Most testing platforms will help you gauge the statistical significance of your tests.
Once you’ve determined a clear winner, you have two options:
- Either do further testing of other page components, or
- Run with the winning page.
The Right Tools for the Job – Testing Platforms
There are several tools you can implement to run your tests, and it’s important to evaluate the features and benefits of each and how those features align with your testing goals. Check out some of the top tools for testing your landing pages.
This is a free tool with code you can use to test different variations of your landing pages. You can even use the Visual editor to create variations of your landing page without changing the code.
Customize your landing page experience for different demographic segments of your customers, and choose which visitors you’d like to include in your A/B test.
Another great thing about Google Optimize is it integrates with other Google tools, such as Analytics, which allows for a more seamless tracking experience and give you more visibility into the overall success of a campaign
Optimizely is another great tool you can use to test both your landing pages and your entire site. Test on any device and any format, including mobile, and use its robust Stats Engine to make your decisions.
Optimizely also uses a visual editor for easy variations, allows complex (and simple) segmentation, and provides you with custom snippets. Custom snippets are pieces of code that ultimately allow your site to run at optimum speeds while you run A/B testing. They also allow you to run multiple tests at once and across multiple properties.
One downside to Optimizely in comparison with Google Optimize is the pricing; there is no free version of this app. The good news? You can use a free trial to compare it to other platforms.
VWO (Visual Website Optimizer)
The major benefit of VWO is that it allows you to create landing pages without using any code. You can then use its reporting functions to make appropriate changes. Everything on VWO is mobile-optimized, and it even has a heat-map function to show you what areas of your landing page attract visitors’ attention.
VWO provides a large list of best practices from experts that are customized by industry, and the company offers 24/7 support.
The ease and functionality of VWO does come at a price, and, again, there’s no free version of the tool.
Optmyzr provides quite a few data features that will give you advanced insight into your campaigns. While less sleek-looking than some of the other tools, this app gets the job done efficiently. The visual design feature offered by Optimizely and VWO is not something you’ll get with Optmyzr.
This tool integrates with Google Analytics but offers less support at the lower tiers than other solutions. The lite version of Optmyzr starts at $208/month.
A/B Testing in Action (Examples)
We pulled some great examples of landing page A/B testing from behave.org, so you can see what it looks like in action.
Testing Two Different CTA Buttons
Culligan suspected their “GET A QUOTE” button would be the more popular CTA because water treatment is expensive, and the word “quote” sounds more personal than “pricing.”
They were right, and the winning landing page lead to a 104% increase in form submissions.
Testing Icon Placement
Comfort Keepers used a digital marketing agency to test whether putting the call icon at the bottom of the landing page would increase phone calls, and it paid off!
The winning landing page is the one with the phone icon both at the top and bottom of the page, increasing phone calls by 74.3%.
Testing Landing Page Image
The Boat Owners Association wanted to test whether a serious-looking captain or a “happy” captain increased sign-ups to their annual membership and on-water towing services.
It turns out the serious-looking captain portrayed authority and increased conversions by 32.8%.
Now It’s Your Turn
Now that you’ve seen the impact testing can have on conversions, it’s time to start testing your landing pages. Even with your knowledge and business expertise, we know getting started can be daunting, so we recommend hiring someone with conversion rate optimization (CRO) experience to help you get your testing off the ground.
Frontier Marketing can help you find the right tools, use them to their full capabilities, and start optimizing your landing pages today. Contact us to learn more about how our experts can help increase your landing page conversions and get the most out of your PPC campaigns.