The Risk of Conversion Rate Optimization

What are you doing to increase conversion?

• Analytics-based optimization
• A/B testing
• Online surveys and offline focus groups
• Site search functionality overhaul
• Total site redesign
• Insert your own tactic du jour here

Which tactic requires the most effort and resources? Why did you choose one tactic over another? Can you articulate the risk of your efforts? Have you tried all the above, only to experience marginal improvement?

You’re not alone.

A recent ClickZ Features case study detailed how marketers for Universal Orlando have been “tweaking” the site to gain only small conversion growth. Executives wanted large conversion growth and were seeking optimization techniques to reach their goal. (Disclose: my firm worked on the Web site.)

The article quotes Algernon Callier, VP of brand and interactive marketing for Universal Orlando, “When we looked at the idea of redesigning the site, this had come after a good three years of making minor improvements to our existing Web site. We did facelifts, redesigned key pages, and experimented with traditional conversion tactics. We got all the low-hanging fruit.”

The Efforts Don’t Match the Results

Shop.org‘s recent “State of Retailing Online” shows the average overall conversion rates declined from 2.6 percent in 2004 to 2.4 percent in 2005.

This is happening despite many of our noble efforts. Since 2004, technology has improved, Web analytics are more prominent; customers are more comfortable buying online, and we’re committing more resources to optimization, trying everything from A/B testing to total site redesigns.

Click on thumb nail to view full-size image

The report also reveals some other interesting morsels. When comparing some common optimization tactics and their effectiveness in 2004 and 2005, many techniques are becoming less effective, not more. And of those techniques that are increasing in their effectiveness, the increase is minimal.

We should be getting better at optimizing, since we’ve been doing it longer and have more experience. As a byproduct, our conversion rates should be trending upward.

That’s not what’s happening.

Optimization Tactics Aren’t Without Cost

Optimization isn’t without risk. Many online practitioners dive directly into optimization tactics without any thought to overarching strategy.

A/B testing may not be a high-risk technique, but would you willing to commit human and financial resources to it if you knew another tactic would produce greater results sooner?

To some, especially those bored with their current sites, a site redesign seems like a no-brainer. Yet many redesigns throw the baby out with the bathwater. A site redesign may be the riskiest of all tactics, especially if you break what’s currently working on your site in the name of having something newer, flashier, and prettier.

When effort begins to outsize expected results, it’s time to evaluate your optimization regimen. Stop throwing these well-adopted tactics at the wall to see what sticks. Optimize your optimization.

Signs You Need More Than Optimization Techniques

• Marketing in silos. Many larger companies have too many chefs cooking up sometimes slightly, sometimes drastically different marketing experiences across too many channels. Optimizing successes and failures aren’t communicated well within the company, not to mention efforts aren’t synchronized for a more consistent customer experience.
• Giving up too soon. Many companies cast a technique into the river. If they don’t bring up a fish after a few casts, they’re abandoned.
• Optimizing too many wrong assumptions. Many companies correctly identify where problems occur but incorrectly try to fix things with a hammer when they really need a wrench. Are you changing the color or shape of a button when the culprit might be a lack of a point-of-action assurance?
• Thinking too small. Are you trying to optimize a single pipeline at the expense of others? Are you applying too much effort to selling today? You can convert today and load up several pipelines for future conversions.

Methodology vs. Techniques

If you’re worn out by your optimization efforts, consider this: As a visitor arrives at your site, thousands of variables affect her ability to convert. To that, add the fact you have many visitor types with many different needs. Multiply this by each product or service line you offer. These variables also change.

What are the odds of finding the ultimate set of variables for maximum conversion on any given page? How much creative time, effort, and resources are you willing to experiment with until you find out?

Talent can find some wins when optimizing, but alone it can’t account for, test, track, and execute successful implementation with that many variables at play.

Online marketers need a robust, efficient process to make these variables manageable. Marketers need a predictive model of customer behavior, a model that can help prioritize and measure optimization efforts and steer them away from fruitless efforts.

Said Callier in the above-mentioned Features article, “The approach takes into account human decision making styles. It’s beyond demographic, beyond male/female, young/old. It gets into the psychology of how humans make decisions and how they get comfortable making decisions.”

The Universal Orlando site was built on persuasion scenarios that manage these variables by design and make them accountable to a few more stable variables, normal human psychographic behavior, and customer and business goals. The results? The new site beat the old on 30 of 40 qualitative and quantitative measurements, “including likeability.” According to the case study, “Online ticket purchases, a key metric for the resort, are up almost 80 percent year to date.”

Are you still throwing stuff against the wall to see what sticks and wondering why you aren’t extremely satisfied? There’s a way to clean up the mess.