What good is analytics data unless it’s used to improve your site? More companies are taking Web analytics more seriously, but few put analytics findings to work to improve site performance.
Far too often, Web analytics data is used as a report card: “How did we do last month?” To realize the Web analytics return on investment (ROI), organizations must use the data to formulate changes to their sites that will lead to improved conversion of key site goals.
Think back to junior high science. A likely assignment was to create a hypothesis, test it, report on the findings, then use the results to make improvements. You can apply the same concept to your site to improve ROI.
Recently, I presented some analytical findings to a new client who sells consumer-oriented monthly services. The company uses one of the top analytics tools available but had only been looking at high-level visitor information, such as site visits, page views, and key page data. It had yet to formulate a picture of its overall site performance. We worked with the company to identify key products and analyze the purchase path for these products.
Identify Problems and Opportunities
Together, we identified a few key steps in the different purchase funnels that were not converting enough visitors to the next step in the process. We flagged those as key opportunities, then prioritized them based on the company’s focus and potential consequences of changes.
Develop a Hypothesis
We decided to A/B test a key page in one of the purchase funnels. We identified a number of hypotheses behind some of the problems, based on where people were going and what was converting successful buyers.
Our first hypothesis was that the message had to be simplified. The second was that people needed even more information on the product offering while still keeping the offer simple. Obviously, these are two different approaches, and two concepts, behind the reason for failure.
Define Your Success Metrics
In addition, we identified the test’s success metrics to define how to measure the outcome. We prioritized each metric so the determination of the winning option could be easily identified. The top measurement goal was to improve the conversion rate of visits to the main product page to a successful checkout. By focusing on the hypothesis and success metrics, we developed five different options to test against the existing page.
Analyze Your Results
Just last week, we presented the preliminary findings after a week of testing. All five options outperformed the existing page. Two designs clearly outperformed the others (the two hypothesis outlined above). They improved the conversion rate by approximately 0.5 percentage points. Initially, that doesn’t sound significant, but the initial conversion rate was 3 to 5 percent.
Reap the Benefits
We analyzed how the changes affected revenue, both monthly and annually. Though that 0.5 percentage point change sounds small, its cumulative total exceeds $1.5 million annually. Over the coming weeks, we’ll fine-tune the testing based on the two best-performing pages.
You don’t have to double your conversion rate to enjoy a significant benefit. Even small changes can have major outcomes. When testing, don’t be afraid to experiment. You may be surprised by what works.
Emily Ma, product director of Tencent’s advertising platform products department, was a keynote speaker at ClickZ Live Shanghai where she discussed the ... read more
In today's multichannel world how can marketers use data to ensure the experience a customer receives is relevant to them?
The terms that customers type into your site search function can help you to gain an understanding of user behaviour and can be used to optimise ... read more