Without realizing it, you’ve likely participated in a multitude of tests whenever you’ve visited sites like Google and Amazon. Why? These companies are rightfully obsessed with hard data, assessing their websites thousands of times per day in an effort to optimize conversion rates. While this process of “measurement and optimization” is integral to their success, it isn’t proprietary. Your business, too, can benefit from adopting similar methodologies.
Increased conversion rates result from analysis and testing. Put simply, you collect information, and then you adapt your website in response to data you’ve gathered. While “ideal” conversion rates vary from industry to industry, they share a commonality that was recently mentioned at a conference of the American Marketing Association. The keynote speaker, Michael Loban, VP of InfoTrust LLC, shared the following insight: “The only good conversion rate is the one that increases every day.” This statement underscores an important truth about data-driven optimization: Success is dynamic. When executed effectively, optimization should provide continual improvement. Your day-to-day adjustments to your site needn’t be dramatic, but minor changes add up to an increasingly effective online presence.
So how exactly do you optimize your site? You create experiments to test variations of your Web pages and then allow the data to determine a winner. The most predominant optimization test is known as “A/B.” It involves evaluating two pages, A and B, by comparing them to each other. Another popular measurement technique is multivariate testing, in which you test many different variations of the same page all at once. This can be as simple as testing different colors of a button on the same page and noting which color drives the most conversions. You could also try changing the wording on a button from “Submit” to “Join in the fun!” You might be surprised at what the data (and your site’s visitors) will tell you.
Google Analytics offers a simple and effective testing option for all of its users. Called “Content Experiments,” it enables you to test up to five full versions of a single page at once. This isn’t necessarily an A/B test, since you can test more than two variations at once, and it isn’t a classic multivariate test, since you can test up to five full versions of a page, but it’s effective. Google refers to it as an “A/B/N” test, which seems to fit quite nicely.
So, what should you start testing first? According to Marketing Sherpa’s 2012 Benchmark Survey, the most prevalent website optimization priorities include the following:
1. Increase overall conversion for a page.
2. Learn about customer behavior.
3. Discover the most appropriate wording on a page.
4. Determine the most effective page elements.
5. Find leaks in your conversion funnel.
At gyro, we strongly recommend a practice of measurement and optimization for all of our clients. But will cultivating a healthy preoccupation with data pay dividends for your business? You won’t know for sure until you test it …
Andy Gibson is a performance marketing analyst at gyro Cincinnati.
Follow Andy @apgibson16