January 27, 2009
Case Study

6 Steps to Optimize Website Conversions: 43% Boost

SUMMARY: You’re a start-up with one goal – to convince people to sign up for a free money-management service. And, as is often the case with start-ups, your money is tight.

See how the founder jumpstarted website conversions within the first few months of launching his service through optimization. Six steps to big visitor-conversion boost.

Nikhil Roy, Founder and CEO of Rudder.com, created a service that helps people organize their personal finances and better manage their money. The software he and his team developed sends a summary of personal bank account and credit card balances and transactions daily through an email to a registrant’s inbox or iPhone. The service includes a reminder tool that lets registrants know the amount of funds they have left in their accounts to pay bills.

Roy had already invested in PR by hiring someone dedicated to getting coverage for Rudder. The company got significant press coverage from major news publications, such as The Wall Street Journal, CNET.com, PCMag.com and CNBC's "On the Money." The company's PR efforts, along with some online ads, drew enough visitors to the homepage.

“Most of our traffic is direct,” Roy says. “That tells us [people] already knew about us before they got to us … Even the search traffic, most of the keywords are our brand name.”

The challenge was converting visitors at a rate his marketing team found acceptable.


Roy and his team launched the Rudder website last September, and began optimizing it in October. Here are the steps they took:

Step #1. Analyze the existing site

The team ran the existing homepage through a software program that analyzed the site. The software identified shortcomings based on known best practices in website optimization, and came up with hypotheses of what was worth testing. Here are the hypotheses the program identified to lift conversions:
o Static layout with all features visible
o New benefit headline
o Larger sign-up button above the fold
o Links to “learn more” pages
o More prominent security message
o More prominent “compatibility with MacMail” message
o Thumbnail pop-ups tied to features

Roy and his team then created three test sites. Variations A, B, and C combined a mix of the different hypotheses.

Step #2. Make the homepage a landing page

The Rudder team designed the homepage to look like a landing page. They decided that a homepage as landing page would work well for Rudder because the site has a single purpose: Get people to register for the free Web-based service.

The homepage design included all the relevant features of the service in one static layout, as the hypothesis suggested. Three variations were created:

Variation A:
-Included list of Rudder’s four main features next to enlargeable thumbnails.

Variation B:
-Included list of the Rudder’s four main features next to checkmark-shaped bullet points.

Variation C:
-Included the four features in the description of the links and in the prominent sign-up button.

Step #3. Create a headline that communicates value

The headline on the original homepage read: “It’s Finances in your inbox. Save a trip to your bank’s website!” Roy and his team decided the original headline didn’t express the value proposition clearly enough to convince people to try the service. They wrote two other headlines for the three variations:

Variation A headline:
“Take Control of Your Finances Right Now”

Variations B and C headline:
“Take Control: Experience a better way to keep track of your finances – for free!”

Step #4. Put a large ‘call-to-action’ button above the fold

Each variation also got a ‘call-to-action’ button above the fold – high enough on the page so visitors didn’t have to scroll down to see it. The original site had a smaller “Join Now” button on the lower half of the page.
Different sizes of the buttons were created:

Variation A:
A medium-sized “Get Rudder Now – it’s Free!” button.

Variation B:
A larger-sized “Get Rudder Now: It’s Free and Secure!” button.

Variation C:
A medium-sized “Get Rudder Now – it’s Free and Secure!” button that was placed even higher on the page than Variations A or B.

Step #5. Run the test

Roy’s team ran the A/B/C split test for three weeks in October and November. Visitors to the website saw one of three homepages, which were served in random fashion.

Step #6. Keep best of the best

Optimizing the homepage proved its worth: All three variations outperformed the original site.

Variation B produced the best sign-up rate, though. So, Roy’s team adopted that page as the homepage. Here are the test results:

- Variation B increased the conversion rate by 43.5%
- Variation C was the runner-up with a 31% lift in conversions.
- Variation A was least effective, but it still had a 15.8% lift in conversions.

Average conversions before and after:

o 16% monthly average conversion rate before optimization.
o 25% monthly average conversion rate after optimization.

“It was surprising,” says Roy. “I wasn’t expecting that big a bump.”

About 25,000 people have signed up for the free service since it launched four months ago, Roy says.

“One thing we noticed is about 50% of people that sign up for our service never go beyond the [optimized] landing page,” Roy says. “They are convinced right at the landing page that this is something for them and they go for it.”

Useful links related to this article

Creative samples from the Rudder test:

WiderFunnel Marketing Inc. – helped Rudder with the landing page optimization test:

Rudder’s press page:


Improve Your Marketing

Join our thousands of weekly case study readers.

Enter your email below to receive MarketingSherpa news, updates, and promotions:

Note: Already a subscriber? Want to add a subscription?
Click Here to Manage Subscriptions