If you haven’t heard of Fiverr, they’re a global online marketplace that’s raised over $100 million in funding / kind of a big deal. They’re also a superstar Unbounce user. But that’s not why we’re featuring them here.
We’re going to show you how they A/B tested the **** out of their Unbounce landing pages and increased conversions by 457%. In 10 months. With one marketer.
Or does one?
Yoav, the Growth Specialist at Fiverr who ran these tests, takes his A/B testing seriously. All of his tests have achieved a 99% confidence level, with over 500,000 unique visitors per month from Google and Facebook pay-per-click ads — across 400 A/B tests. Yes, we’re talking 400 landing page variants.
Sounds like a lot, but when you have the power of Unbounce’s landing page builder, all you need to know how to do is drag and drop. In minutes, you too can create a landing page design to test.
A/B Test 1: The Page Layout Test
The treatment had a 23.5% increase in conversions, just from pushing down the content a bit to add some whitespace (okay, technically the space isn’t white here, but the idea is having breathing room). This is probably one of the easiest A/B tests you can run.
A/B Test 2: The CTA Position Test
Make your CTA obvious. Prominent, even. Fiverr increased registrations by 27.3% just by moving the button to its own line, centering it and widening it.
A/B Test 3: The Actionable Copy Test
Look at the control’s headline and subhead. Pretty bland, doesn’t seem like it’s talking to anyone in particular. It’s just shouting into the void, “Professional logo designers! Custom logo designs starting at $5!” You wouldn’t yell every sentence at someone in real life — why do this?
The copy’s makeover on the treatment is more conversational and includes more benefits phrased in an easy-to-understand way. Result: a 29.81% increase in signups. #winning
A/B Test 4: The Ultra-Specific Copy Test
So that last test was pretty impressive, eh? But Yoav ran another test after that test: he replaced the vague-ish “easy” in the second line with the ultra-specific benefit of a custom logo that’s three clicks away. He didn’t even have to change the headline or CTA to get this 15.41% increase in signups. Crazy.
A/B Test 5: The Ultimate Validation Test
Here’s the thing with A/B tests. Normally, you should only test one hypothesis at a time — testing two different headlines should be a different test than, say, testing different button colours. Otherwise, you won’t know which change is responsible for the positive (or negative) results.
But Yoav wanted to validate three things he’d learned from previous tests specifically that:
- providing more information in the form of a “How it Works” section would keep the user on the page and decrease the bounce rate
- pushing the Gigs further down the page would get more users to sign up (since he didn’t need to get them to actually book a gig at this point)
- using a photo of a woman looking directly at the CTA, with calm background colors, would increase conversions
By combining these three changes that had tested positively into a new design, Yoav got a 57.74% increase in registrations. Wowza.
A/B Test 6: The “Learn More” Test
The thing with CTAs is you should have one per landing page. The more CTAs that link out of your landing page, the more distracted your reader will be and the less likely they’ll be to take the action that you want. If you really want your user to sign up, but add other buttons that say “Learn More,” you’re not helping them focus on that one goal. This is a key concept in conversion rate optimization (CRO) and is often overlooked because for some reason humans just love having 20 links on a page.
You want to optimize for the most valuable conversions. In this case, having more information available through the “Learn More” seemed to have helped leads become more convinced that Fiverr was a solid choice, as evidenced by the 22% increase in orders and 17% decrease in bounce rate — these leads were more valuable because they were more informed. But the fun doesn’t stop there. This then boosted their Quality Score and consequently lowered their cost-per-click in their PPC campaigns. That’s like a win-win-win-win.