by Adam T. Sutton
, Senior Reporter
The goal of testing a landing page is to increase conversion rate, but not always just that exclusively. The best marketers know that they must also learn from tests if they want to push results higher.
Brian Smith, Director of Marketing, AwayFind, strives to learn from his tests. After joining the email notification and sorting service last year, he wanted to test its homepage.
"Our primary driver of traffic is our PR efforts," Smith says. "Our homepage is effectively our primary landing page, and we need to convert that traffic into premium users."
Smith encountered resistance. Coworkers told him the homepage had already been tested and optimized, and that other projects had the company's developers tied up.
He persisted, though, and got the resources. Now he had to make sure the test went well and revealed something about the audience. His team followed these five steps:
Step #1. New copy for the homepage
Having just joined the company, Smith wanted the test to have impact. He targeted the homepage
, which offered visitors a free trial. He hoped to increase its conversion rate.
Looking at the page's copy, he noticed an emphasis on features. Here are a few lines:
- "Let us find your urgent messages"
- "When you receive an urgent message, AwayFind will notify you"
The page also mentioned a feature that delegates emails to another person. Smith thought the homepage wasn't the right place for it.
"It is a very interesting and compelling feature, but I realized that what the data showed is that people weren't really using it."
Benefits, not features
Smith saw an opportunity to test a change to the homepage's copy. Instead of emphasizing features, he wanted it to emphasize how AwayFind's service helps customers.
For example, here's the sub-headline of the new copy:
"Get AWAY from your inbox -- let urgent emails cut through the clutter and FIND you…instantly."
Step #2. Clarify the layout
Smith handed the changes to AwayFind's design team. The new copy was longer, and the team thought it looked a bit bulky on the page.
"Keeping this layout would make the heading four lines, which would overpower the page visually and be difficult for the reader to digest," says Keith Ludlow, Director of Design, AwayFind.
The team made the following adjustments to the test page
- Shortened headline into a single line
- Split the remaining copy between a subhead and a smaller block of text
- Changed the subhead text from blue to white to create "a bridge between the headline and the smaller text below"
"By easing the reader through the text in stages, we were able to fit more copy in less space," Ludlow says.
Step #3. Review test results
"We would have been happy with a 15% or 20% increase," Smith says, "but the result really knocked our socks off."
The new homepage achieved the following:
- 42% increase in clicks to the sign-up page
- 91% increase in registrations for the trial
"I started almost 20 years ago in direct mail, and I have been testing for a long time. I can think of maybe a half-dozen or dozen occasions where I have seen results like that."
Everyone was happy, but Smith wanted to learn more. He could not be certain of what caused the increase. Was it the new copy or the new layout?
Step #4. Create a follow-up test
The team created a second test page
that used the new copy in the original format. It was the same page that Smith handed to the design team earlier, the one with bulky blue text.
The team planned to test the second page against the original. Smith realized it was unlikely to show a higher conversion rate than the first test, but that wasn't his goal.
"We wanted to know what really caused the change, so we went back and retested."
Step #5. Review results for insights
The team noticed interesting results. Compared to the original homepage, the second test earned:
- 23% increase in clicks to the sign-up page
- 53% increase in registrations
Comparing the results of the two tests, the team estimated that the new copy accounted for 54% of the increase in clicks and about 58% of the increase in registrations in the first test. The remainder was attributable to the change in layout.
"The copy came out slightly ahead but not by a huge amount," Smith says.
Learning from the results
The test did not reveal any earth-shattering insights (although the results were great). Looking at the data, the team could make the following hypotheses:
Want to learn more about this test and other companies’ optimization efforts? Brian Smith will present at Optimization Summit 2012 in Denver, June 11-14.
- Copy should emphasize the benefits of popular services rather than features
- Changes in copy can have more impact than changes in layout
- Even a page that has been thoroughly tested has room for improvement
- Original homepage
- Homepage test #1 - new copy, new layout
- Homepage test #2 - new copy, old layout
Related ResourcesLanding Page Optimization: Radical redesign leads to 3,566% increase in conversion Offline and Online Optimization: Cabela’s shares tactics from 51 years of offline testing, 7 years of digital testing Copywriting: How to improve headlines on landing pages and blog postsPersonal Development Plan: Free resources to help you start testing (and help your career)Landing Page Wireframe: Why focusing on ‘one variable at a time’ doesn’t workCommon Landing Page Mistakes: Form fields that stop selling value