But in conversion optimization, examples are especially hard to come across. Unless you know they’re happening or want to spend an inordinate amount of time Googling, you’ve got to resort to your imagination to devise new ideas for experiments to run on your website.
While I’m sure your imagination is very vivid, we figured we could do some of the research legwork for you. Below are nine fascinating case studies to inspire your next conversion optimization test.
Most successful blogs include a call-to-action at the end of their blog posts. It’s usually full-width — large enough for people to notice the offer and hopefully convert on it.
But are people noticing that CTA, or are they learning to tune them out?
Here at HubSpot, we were curious if our readers we developing static CTA blindness — so we decided to run a test to see if we could increase our CTA clickthrough and conversion rates without being a huge, annoying marketing jerk.
To accomplish both of those goals, we decided to try using slide-in CTAs that would appear as soon as the right sidebar disappeared — usually half-way to three-quarters of the way through a post. Here’s an example of the slide-in:
To test this out, we added slide-in CTAs to 10 of HubSpot’s highest-trafficked blog posts. A month later, we looked the following stats for the slide-in CTA and the static CTA at the end of the post:
In this test, the slide-in CTA had a 192% higher CTR and generated 27% more submissions — mission accomplished.
It’s one thing to talk about a conversion "best practice" (placing something above the fold) — and it’s another to see it debunked in real life.
ContentVerve.com was the one to debunk it all. To test this best practice, they put a CTA alllllll the way at the bottom of a very long landing page.
The result? 304% more conversions on the below-the-fold CTA.
Take that, best practices. This is a great reminder that you should always test out tactics like this yourself, no matter how often they’ve been recommended.
On the flip side, sometimes A/B test can uphold long-held best practices — that’s exactly what happened in this test by Optimizely.
Before they began experimenting, Optimizely was running a few PPC ads with several different types of messaging to one landing page. The landing page did not use the same terminology as the ad — instead, it read "Try it Out for Free." Below’s an example of what this whole process roughly looked like.
So Optimizely decided to test if aligning the copy on the landing page to the ad resulted in more leads generated:
And the treatment worked: It resulted in a 39.1% increase in leads generated.
So if you’re looking to optimize your cost per lead in your own PPC campaigns (or increase the effectiveness of any other type of referral sources), try aligning your landing page copy and referral source copy.
Usually, the answer to getting higher quality leads is to increase the number of form fields — that way, only the truly committed (and usually most qualified) candidates will take the time to fill out the form.
And the two changes worked — Iron Mountain got 140% more qualified leads with this new form copy and design than they did with the old.
Truthfully, this change makes sense. People who are going to take the time to fill out the right information to receive a quote specifically are probably going to be better fits for your business.
Makes you want to learn some code so you can start tweaking forms yourself, huh?
Most marketers think that the more social media share buttons, the better your conversion rate will be, right?
Taloon.com actually found the opposite to be true. They wanted to get more people to click "Add to the Cart" on their product pages and wondered if removing share buttons would help increase conversion rates.
Spoiler: It did. Removing the social share buttons increased clickthroughs on the main call-to-action (add to the cart) by 11.9%.
The lesson here is that social proof can work if there are already people sharing something — if there’s no one sharing your page, it could hurt conversions.
Most blogs look very similar — top navigation, body of the post on the left, and a right-aligned sidebar with extra information. What if this typical format was turning off people from converting?
IMPACT thought it might, so they decided to put it to the test. After removing the sidebar, they used a standard CTA on one post and an in-post form on another — both for the same offer.
In this new blog design, the in-post form performed way better: It generated 71% more leads.
Maybe this is something you should try on your blog, too. Full disclosure: We’ve tried a similar test on this blog, and it didn’t show the same results.
While most conventional landing page copywriting advice will tell you that straightforward copy is the way to go.
For Copyhacker and JDC Repair, that didn’t prove to be true. JDC Repair is a while-you-wait iPhone repair store catering to image-conscious teenagers (and their parents) who’ve broken their iPhone screens — and this audience could be the type to appreciate some coy copywriting on a landing page to schedule a repair appointment.
So they decided to test very clear, value-driven copy with one that’s a little more fun:
Image credit: Unbounce
And the fun one won: 18% more people scheduled a repair appointment.
Again, another lesson in "you should always test things out for yourself." Maybe your audience loves fun copy, too!
A few years ago, 37Signals was trying to make drastic changes to the page layout for Highrise (which has since been spun off into its own company). They first tested a longer page design with more descriptive copy. The result: 37.5% increase in conversions.
The next test blew the first out of the water. They made the page shorter than the original and added a lovely photograph of their smiling customer in the background.
This time, the page got 102.5% more conversions. Even after doing further testing using different photographs of different customers, the results held up: Photographs of smiling people worked.
Who knows if this’ll work on your landing pages, but at least you have some data to help you debut your modeling career on your company’s website. 😉
This story comes from the Sidekick team at HubSpot. They’re constantly testing to see how they can use their homepage to generate new active users — and this test helped them do just that.
Previously, their homepage included a list of all the features Sidekick has:
But they were curious to know if those detailed actually mattered — for a product as low-touch as a Chrome extension, do people need to the feature laundry list to convert into active users?
So they decided to replace the feature list with a few user testimonials. Here’s what that treatment looked like:
Sidekick ran this test until it was statistically significant, and found out something very surprising: The testimonial beat out the feature list by 28%.
Their theory on why this change took place? The former didn’t make people curious enough to click through to the Chrome Extension installation page. I’d also venture a guess that people could have wanted more social proof before downloading a new tool into their browser.
Regardless of the reason, this worked — and it might work for your company, too.
What awesome experiment have you seen companies do? Share your favorites with us in the comments.