How GoToMyPC Spiked Lead-Generation and Conversion Rates

GoToMyPC, from Citrix Online, a division of Citrix Systems, offers a subscription-based remote access service for individuals and businesses. It allows users to instantly and securely access their PCs “from anywhere.” Subscribers can check email, open files, use applications, and access their networks from any Web browser with a simple-to-set-up, two-minute download.

Like many software companies, Citrix Online acquires almost 90 percent of its free-trial users from online marketing activities. I had the privilege of working with Bernardo de Albergaria, Citrix Online’s marketing VP. We talked about our work at a recent trade show. Here’s a bit of what we shared.

The goal is to get as many visitors as possible to complete a free trial. The premise is if they try it, they should buy it. The company’s main metric is a ratio of new visits to users in trial.

The trial flow requires several actions:

  • Input personal/profile data
  • Choose a password
  • Input billing info
  • Download the software

Citrix Online developed in-house A/B testing capabilities, where it can create ad-hoc flows and shuffle placement/sequencing of flow elements. A test is always running, involving approximately 20-30 percent of available traffic.

The company has been testing and tweaking its landing pages for improved conversions since GoToMyPC’s launch in 2001. Back then, its visit-to-trial ratio was 0.3 percent. With a few revisions and incorporation of the first part of its registration process, it quickly boosted this to 0.7 percent. Then the ratio jumped to 1.1 percent with its first purpose-built landing page.

GoToMyPC increased the conversion rate over 400 percent in 2001. De Albergaria offers some direct marketing tips:

  • Headline. The headline was static, bold, and optimized through text link and email subject line testing.
  • Registration form. The form was designed to be above the fold, with a strong call to action. A reminder “two-minute setup” message reduces fears of a long install process.
  • Copy format. Citrix Online used a short description, with benefit/usage examples instead of feature descriptions (see “The Way Customers Want to Buy” for a deeper explanation). This was then optimized through email.
  • Press quotes with logos. The quotes provide third-party endorsements from recognizable brands to increase trust.
  • Ugly burst graphic. The graphic creates a special offer image.

In 2002, these revisions were improved upon. Citrix tested combining the registration and credit card forms; it found that by not asking for credit card information on the landing page, the form performed 197 percent better. It learned providing password instructions in boldfaced type improved results by 15 percent. It also added links to actual press quotes and logos that increased trust, provided more information, and reinforced the third-party endorsement. It also refined copy and tested numerous images to see which performed best.

Since then, the company has tested taking the same page components and applying a different design to them: a “light” landing page with reduced copy and imagery; a completely new layout; Flash and animated graphics to illustrate benefits; cobranded pages; and both heavy and minimalist copy. It’s tried having the registration form on the top of the landing page, even some radical detergent, or super flashy, design pages.

Citrix Online’s approach has been exceptional. It’s seen amazing increases in conversion rates and learned a lot about what motivates the average visitor to start a trial.

Now, it’s hitting some plateaus with current efforts. The return on investment (ROI) of incremental tweaking is declining. It’s still testing many other elements: pricing, flow, acquisition creative storyboarding, billing methods, and so on. However, Citrix decided it’s time to customize the prospect’s sales flow by user segment.

It’s developed six personas using persuasion architecture for testing. These personas are as diverse as David Tudor, a Competitive Workaholic Backer-Upper; Paul Berger, the Freedom Seeking Technophobe; and Evan Lambert, the Techno-Controlling Doubter (obviously, these personas have a lot more detail we can’t share here). We completed wireframing the experiences each persona will receive and wrote copy for all the pages.

De Albergaria and his team are busy implementing the new pages we created and are starting to test copy and creative based on each persona’s individual needs. When the new pages are launched, they’ll already be somewhat optimized.

De Albergaria shares a few additional tips:

Leonardo da Vinci said, “Simplicity is the ultimate sophistication.” This is definitely the case with your online efforts. Whatever you can do to simplify your copy to make it easier to read and comprehend, do it. Whatever you can do to simplify your process, test it. Just remember K.I.S.S.

It’s imperative to get a user to start the trial during their first visit. It’s difficult to change their mind after they’ve abandoned the site on their first visit. The reason we’re working on a persona approach now is so that we can answer all the questions each individual user segment may have and qualify our visitors immediately. We don’t expect each user segment to perform equally, but if we find two or three of the six that we can optimize for, we should see some tremendous gains from our efforts. Not all prospects are created equal.

Most important, never try to address what you think might work without actually testing it first before the court of last resorts: your visitors. You’d be surprised to learn what you think is going to do great may drop your results, and the stupidest idea may give you an incredible boost.

De Albergaria now faces a new challenge familiar to many. The company has released a new product, GoToMeeting. The company must calculate what its baseline conversion should be. Next, I’ll discuss how, based on a number of factors, we arrived at the right number to plug into the spreadsheet.

Related reading