Our Top 10 Split Tests of 2015: The Biggest, Strangest, and Most Interesting A/B Test Results We Saw Last Year

Top 10 A/B Tests of 2015

When a select group of LeadPages® members run an A/B test, they’re not just improving their own marketing. In a way, they’re helping the world.

I’m talking about those intrepid marketers who allow us to feature their split-test results on our blog. By running interesting tests, they allow everyone to learn more about what kind of landing pages, messaging, and design are (and aren’t) working right now.

We reported on nearly 100 of those A/B tests in 2015. Some of the results confirmed what we’d have predicted, while others were surprising—even a little mystifying.

Here are 10 of the most eye-opening A/B tests we saw this year. We hope the results inspire you to do some testing of your own, and with that in mind, we’ve created a new PDF for you, called “25 Easy A/B Tests to Help Uncover Your Landing-Page Success Formula.” Get a free copy below:

Download Your "25 Easy A/B Tests" Guide

If you have LeadPages™, you don’t need to download this template – it’s already available to you inside your LeadPages account. Just log in and you’ll see how super easy it is to customize this page in seconds with no technical knowledge or skills, make it mobile responsive, integrate it with your email service provider or CRM, run A/B split tests, and publish it to Facebook, WordPress, or your own server.

In no particular order, these are the top 10 tests that really got us thinking this year:

1. A Baby-Related Conversion Boom

test1

The test: Justin Verrengia tested the impact of his webinar presenter photo. One was a traditional headshot on a white background, while the other showed Justin standing outdoors, holding a baby in his arms.

The result: The baby was a winner, improving the signup rate 66.12%.

What it tells us: Professional polish is important, but so is a sense of humanity—especially in a personality-heavy format such as a webinar. Commenter Piers brought up an additional point: “It was visibly showing the speaker as a parent that made him more relatable to a sizeable chunk of his demographic.”

Supporting evidence: Though no baby was involved, acupuncturist Jeffrey Grossman got a 28.56% conversion-rate lift this year when he tried a more informal photo over a standard headshot.

2. Can You Be Too Confident?

The test: A real-estate firm tested two kinds of messaging in its LeadBox™: one that sought to allay visitors’ worry that they’d be contacted too often (“Want to see this community? Afraid that we’ll bug the heck out of you? Don’t be. We will follow up with you but we will not pester you. I promise!”) and one more straightforward version (“Want to see properties in this building? Fill out the form and our expert will get back to you right away”).

The result: The straightforward version won by an impressive 152%.

What it tells us: It’s worth taking a confident approach in your messaging, speaking to what visitors want rather than highlighting potential drawbacks. After all, many visitors might never have thought to worry that they’d be “pestered” by the agent until the LeadBox™ brought it up.

2test

Supporting evidence: A cooking site experimented with replacing the typical “100% Secure” submit-button copy with “It’s Free”—and saw a 20.77% conversion-rate boost.

3. The Power of Negative Thinking

3test

The test: On a landing page aimed at pilots, Doreen Nyback put 2 different spins on her headline. Version A read “Here’s What Every Pilot Should Know Before Buying a Single-Engine Airplane,” while Version B promised, “You’re About to Learn the Secrets That Most People Will Never Know About Buying a Single-Engine Airplane!”

The result: Version B won, giving the conversion rate a relative 388% increase.

What it tells us: Offering something that “most people will never know” suggests a greater advantage—and a rarer benefit—than “what everyone should know.” It’s a time-tested copywriting technique, and tests like this suggest there’s plenty of life left in it.

Supporting evidence: In another split test this year, a headline with a similar pinch of negative language (“Learn the 14 Best-Kept Secrets of Business Coaching Franchisors: What They Don’t Want You to Know About Buying Their Franchise System”) performed108.11% better than “Everything You Need to Know BEFORE You Purchase a Business Coaching Franchise.”

4. A Numbers Game

The test: Are people wowed by big numbers, or by specific big numbers? In this A/B test, one headline promised to reveal how a marketer got 21,559 leads and 823 new customers signed up, while the other framed the same numbers as “over 20k” leads and “more than 800” signups.

The result: People wanted the nitty gritty details, it turned out—the specific numbers performed 140% better.

What it tells us: At every step of the way, your page visitors are consciously or subconsciously gathering evidence about your credibility, and giving exact numbers instead of ballpark figures can make your results seem more “real.” Even seemingly minor decisions like this can have a substantial effect on your bottom line.

5. Benefits, Not Features?

5test

The test: A LeadBox™ aimed at chiropractors sported 2 different headlines. One simply promised “Free SEO & Marketing Tips,” while the other encouraged visitors to “Discover How Your Practice Can Be #1 on Google.”

The result: “Free SEO & Marketing Tips” performed 26.79% better.

What it tells us: Conventional wisdom holds that your copy should focus not on your product and its features (say, free SEO and marketing tips), but on the benefits to customers (such as ranking #1 on Google). Why didn’t this hold true here?

For one thing, the winning LeadBox™ had the power of “free” on its side. But it may have also better reflected where visitors were in their businesses. The less successful headline may have worked well for visitors who had a goal (ranking higher on Google) but weren’t sure how to accomplish it yet. But if those visitors already knew what they needed to do to achieve their goal—get better at SEO and marketing—the more direct headline likely made a stronger, quicker connection.

6. The Third Form Field’s a Charm

6test

The test: Tom Challan used 2 LeadBox™ variants to give away his lead magnet, How to Crush Your Fear of the Phone. One had just 1 form field, while the other had 3.

The result: The LeadBox™ with 3 form fields performed 120% better.

What it tells us: Most marketers we’ve seen find more success when they take a minimalistic approach to their form fields, so this was a little surprising. Tom’s results are a reminder that even basic principles are worth testing for your audience. (My pet theory about why this test turned out as it did: when a lead magnet is actually about overcoming your fear of the phone, a request for your phone number may have seemed more natural.)

7. No Call to Action, Plenty of Action

7test

The test: Jazz Piano School performed one of the most creative A/B tests I saw this year. One version of their landing page featured a typical call-to-action button . . . while the other version had none. Instead, visitors had to click on the large main image to trigger the LeadBox™.

The result: Dropping your call to action seems pretty crazy, but it paid off: the no-button page performed 35.69% better.

What it tells us: Human curiosity is a powerful thing. In this case, it appears more people were moved to play around with a possibly-interactive-looking image than to respond to an explicit call to action. Especially if you’re in a creative field, it’s worth thinking about how you can add elements of play and discovery to your site. (And being bold enough to run against-the-grain A/B tests if you suspect they’ll pay off.)

8. Quirky Copy

8test

The test: An educational marketer tried 2 different LeadBox™ buttons: one reading “Send Me the List Now,” the other “Why Not … Send It Over.”

The result: The quirky, informal approach worked well here, boosting the opt-in rate 34.79%.

What it tells us: Finding the best voice for your audience is more important than adhering to general copywriting principles. In this case, laidback button text seemed to connect better with the company’s high-school and college-age audience.

9. 1 Word, 129% Conversion-Rate Gain

The test: One headline in this landing page test talked about “Xbox 360 consoles,” while the other variation described them as “Xbox 360 gaming consoles.” That was it.

The result: “Gaming” improved the conversion rate by 129%.

What it tells us: This one’s a bit of a head-scratcher. Personally, I never would have predicted that adding one simple, fairly mundane word to a headline could produce such an impressive improvement—but that’s why A/B testing is eternally interesting. It could be that adding “gaming” added extra clarity, or, as a commenter suggested, that the extra word created a more reader-friendly line break.

10. The Biggest Boost of the Year

10test

The test: A business-software company tested two different landing page backgrounds: an appealing island-vacation scene and a photo of two professionals leaning over a table filled with charts and graphs.

The result: In the most dramatic conversion-rate increase we saw in our community’s split tests this year, the all-business photo improved the opt-in rate 465%.

What it tells us: Pretty images alone aren’t necessarily going to get people to opt in. Your imagery absolutely has to be relevant to what you’re offering—even if that means taking a relatively buttoned-up approach. Otherwise, you risk looking unfocused or less than professional, and your conversion rate could take a huge hit.

Inspired to run some tests of your own this year? We’ve got just the place to start. Before you go, grab a free copy of our new infographic, called “25 Easy A/B Tests to Help Uncover Your Landing-Page Success Formula.”

Download Your "25 Easy A/B Tests" Guide

If you have LeadPages™, you don’t need to download this template – it’s already available to you inside your LeadPages account. Just log in and you’ll see how super easy it is to customize this page in seconds with no technical knowledge or skills, make it mobile responsive, integrate it with your email service provider or CRM, run A/B split tests, and publish it to Facebook, WordPress, or your own server.

Did you run any A/B tests last year? Share your most interesting results in the comments!