How to A/B Test your Landing Page to Maximize Conversions
A/B testing is the basis for improvement in online marketing, especially for landing page optimization. Ads, landing pages, websites and marketing emails should all be tested periodically to ensure your business is getting the best ROI you can.
This article will jump headfirst into the world of landing page A/B testing, and leave you still swimming on the far side. I’ll break it down into the main variables you need to be testing, show you concrete examples of how that test would look and give you real-world case studies to show how testing these variables can affect your landing pages.
A/B Testing Recap
For those of you just dipping your toes into the ocean of A/B Testing, let me give you a quick breakdown of how it works:
A/B Testing is a strategy in marketing in which two versions, A and B (the control and the treatment) are tested against each other. The goal is to identify changes that increase the chance of what you want to occur, to occur.
There are many online A/B testing tools (including Wishpond) which allow you to send half your page’s traffic to the original (A) version, and half to your new, treatment (B) version. The test is run until one variation is clearly more successful (with a 95% confidence level).
Original Landing Page Example
For the purpose of this article I’m going to be testing the example landing page below. Because I won’t actually be driving traffic to this page, or any of the tested versions, all statistics I’ll be quoting will be from reputable case studies. I’ll add up the possible increase in conversions in my conclusion (you can play along at home if you like, as well!).
Not the most exciting thing I’ve ever seen in my life, I’ll admit. But then again, I’ve seen worse. It has all of the primary lead-generation variables: USP, subtitle, image and a CTA. Let’s say it’s converting at a rate of 5% (the low end of average).
A/B Testing the Unique Selling Proposition (USP)
The landing page’s current headline (“making business, personal”), is not really a USP so much as a slogan. The difficulty with this is that many brands consider their slogan to be a part of their business identity - especially if they have been running for many years. Sidelining can be emotionally very difficult.
Get over it. This is business. For example, let’s take McDonald’s, whose slogan “I’m lovin’ it” is perhaps the most well-known in the world. Their landing page’s current headline? “The menu you love, plus so much more”. This is a USP. It’s a unique sale’s point built on offering what people already know and like, with the addition of new and exciting options.
Anytime your business becomes more impressive than another, use the power of a comparison value proposition to encourage conversion.
A/B Testing the Image
The existing image, of a professional-looking group, is by no means a bad one. But, of course, you know, as an informed digital marketer, that a picture of a smiling woman has proven to be the most effective for customer engagement.
In fact, in 2003 a Harvard student worked with a South African bank, sending 50,000 letters offering short term loans. They varied the interest rate as well as included psychologically-influential cues. It turned out that having a picture of a wholesome, happy female in a corner of the letter had as much positive impact on the response rate as dropping the interest rate by four percentage points.
Your landing page’s image is one of the most influential, but frustrating, variables we can test. I’ve found that a smiling woman is the most reliable - but that other images (of groups of people, abstract images or inanimate objects) may increase conversions more based on your business’ audience.
Once you’ve got a great format and text, rotate a few images through your testing cycle to find which gets the most response. They’re one of the easiest things to test, but one of the more difficult to anticipate.
For more information on A/B Testing your image, check out my articles The Psychology Behind a Successful Facebook Ad Part 2: Images, and How to A/B Test your Facebook Ad to Maximize ROI for some best practices.
A/B Testing the CTA
I’ve written before that your landing page is a dance with the visitor and your CTA is leading. I love this metaphor. If your CTA is too aggressive or demanding it’ll step on your page visitor’s toes and they’ll find a new partner.
Remember to keep your CTA’s appealing, rather than demanding. Use ‘you’ or ‘my’, ‘free’ and ‘get’. Tell people what they stand to gain, not what to do.
Here are 5 appealing CTA formulas:
Start your/my free [trial period]
Get your/my free [focus of landing page]
Increase your/my [beneficial result of your service/tool] today!
Learn more today
Try [service/tool] for free
I hear different results from using ‘your’ vs ‘my’ in your CTAs. Some businesses have found the small change of ‘Start your free trial’ to ‘start my free trial’ increases their landing page CTR, some find the opposite. As a result, I’m not going to tell you which one will work for your business. You’ll have to test it yourself!
A/B Testing the List of Benefits
The list of benefits is something that may not necessarily increase your page’s conversions, but nor is it likely to hurt them. We include benefit lists to ensure that if a USP headline, subheader, or traffic source (like an ad or social media link) hasn’t already sold your lead, they get that little bit more encouragement they need to convert.
Your list of benefits could be anything your service, product or tool offers beyond the USP. This list gives more information that people may need, like how your service works, or the steps they need to take to get the result you’ve quoted them.
Remember to keep your list of benefits short and sexy. Include a maximum of five, and draw attention to them with icons, small images, or a clear and delineated box.
A/B Testing the Trust Symbol/Customer Testimonials
Implementing trust symbols or customer testimonials pretty much improves landing page conversions across-the-board. Blue Fountain Media found adding the VeriSign logo to their page increased conversion by 42% and sign-up-form entries by 81%.
Basically you’re telling your landing page visitor that you’re trustworthy; that you’re not trying to cheat them out of their hard-earned cash; that other people have put their faith in you before and won out.
Unless you have a trust symbol from a seriously influential and recognizable authority, I’d recommend you use customer testimonials over trust symbols. Not only do landing page visitors like to see that you have customers, they also trust them more than they do you. Use direct quotes from the most well-known brands you’ve worked with (as their business profile will increase yours).
A/B Testing the Colors
Sometimes it’s the smallest details that have the largest effect on your conversions. It’s changing the color of your CTA button from light green to yellow ( 14.5% conversion increase). Or contrasting the color of two links within a single image (60% increase in conversions).
To get an idea of how color can affect your business persona, elicit an emotion or encourage an action, read The Psychology Behind a Successful Facebook Ad Part 1: Color. Or, for the people who ate lunch at their desks today…
Here are the psychological impacts of five main colors:
Blue : Blue is, across both genders and all age-groups, most people’s favorite color. It is said to create the sensation of trust and security. Lighter blues are calming while darker blues denote professionalism and sincerity.
Green : Associated with wealth as well as environmental subjects, green is the easiest color for the eye to process. Green signifies positive action (think,’green means go’) and affirmation.
Purple : Associated with calm, femininity, and wealth, purple is the second most popular color among women, at 23%. On the other hand, purple is the favorite color of 0% of the male population.
Red : The color red is associated with passion, excitement and urgency. It’s a dangerous color in marketing, as many people associate red with negativity and mistakes. However, it attracts the eye better than any other color and gives the impression that time is passing faster than it is (as it causes our heart to beat faster) causing us to act when we otherwise wouldn’t.
Orange : Eye-catching, bright and sunny, orange is one of the most popular colors for landing page Calls-to-Action. While a good tone and amount of orange is seen as warm and inviting, too much has been associated with naivete and a lack of professionalism.
Our Variation/Final Edit:
What do you think? Does your gut say the landing page we’ve created would convert better than the original?
Let’s do some quick arithmetic:
A/B Testing the USP : Possible conversion rate increase of 127%
A/B Testing the image : Possible conversion increase of 98%
A/B Testing the CTA: Possible conversion rate increase of 161%
A/B Testing the list of benefits : Possible conversion rate increase of 91%
A/B Testing the trust symbol : Possible conversion rate increase of 72%
A/B Testing the colors: Possible conversion rate increase of 21%
Total possible increase in landing page conversions: 570%
This means, in short, that if our landing page was converting at 5% in Version A, our new landing page could be converting at 28.5%. Now think about this, if your ecommerce site is seeing traffic of 10,000 visitors daily, and each conversion’s average purchase is valued at $35, you would be making $17,500/day with the first landing page, and $99,750/day with the variation.
A/B Testing works. And, unless you’re going to engage with a landing page tool, you need to be doing it yourself. Often.
The A/B Testing process is continuous - not necessarily because there is not landing page optimization, but because your page won’t stay optimized. And honestly, it’s very likely your landing page will never be 100% optimized for conversions anyway. There’s always small steps you can take, tiny variables to change, that will affect the page’s conversion rate.
Have you had successes (or failures) with A/B testing? Start the conversation below.