By Brooke Larson, Strategic Account Manager
We all know that we should be constantly testing on our website. This is important not only to keep up with the competition, but to stay relevant to our customers, to maximize our conversion rate and to apply data insights into actionable changes. So, the question becomes, “what does that look like and who should be involved?”
“Survey results show that companies who shift to a culture that embraces testing to make decisions increase conversion 100%.”*
Additionally, consider this recent quote from David Edelman, principal at McKinsey & Co., “If you don’t test, you don’t grow,” Edelman said. “If you don’t test, then you don’t come up with new things. Marketers should be getting out 5, 10, 15, even 20 tests per week into the market. That test-and-learn ethos will result in tighter teams.”
So this adds another dimension to the conversation; beyond what to test, what is the organizational structure needed to support a true testing culture?
What we typically see is that the answer to this evolves as companies move through the analytics maturity curve. Initially companies have a silo approach to testing and it generally starts with a marketing resource either part or full time.
Ultimately, to produce data-driven tests that are revenue impacting at the velocity that Edelman sites, it is imperative to engage the entire company. The cross-functional engagement is the best way to gain visibility into all of the key areas to test on the website.
In order to achieve this, two things have to happen. First, you build this cross-functional approach to testing so you are looking at the entire customer experience, end-to-end and engaging the widest array of feedback on areas to test. Additionally, the way you think and talk about data has to evolve, too. Both are elements of the maturity curve and both are vital to success.
Organizational Structure
To truly be a proactive organization that tests based on data driven insights, you have to reverse the typical organization inertia. In a typical ecommerce shop, you have an eCommerce Director and underneath you will find an eMarketing Manager, Site Manager and a Business Analyst. The Managers hear about a problem or see something in their weekly dashboard and they ask the analyst to dig into the anomaly to help the Executive team understand what happened. If you notice everything about this process is backwards looking. How do we better understand what has already happened?
What we see as the key to driving real results from testing is a complete reversal of this approach. In this new world, you have a strong Business Analyst as the gateway for the test plan. By looking at trends and data, the analyst is finding areas that should be investigated and explored through further testing. Ergo, the analyst and the data drive the testing initiatives.
What to test?
One of the most powerful components of testing is the end goal of targeting content to smaller groups of visitors on your site. This becomes the cornerstone of your personalization strategy. A test may not show a clear winner from an overall conversion rate perspective, but as you start to drill into your segmentation and look at the results, you will likely see strong differentiation and preference by group.
For example, we saw with a recent client that an overall A/B test didn’t warrant a clear winner from an overall conversion rate perspective. However, as we dug into the data it was clear that in the evening, the preference for a more bulleted landing page with less text was winning against the longer and fuller body of text on the other experience. Overall, we didn’t see a preference. BUT, at night you could see this differentiation in behavior. Perhaps, the night owls are tired and feel overwhelmed by too much text? Based on this, we build a hypothesis, put the results into action and continue to test to see what else emerges.
Remember that as Avinash Kaushik says, “80% of the time you are wrong about what a customer wants or expects from a site experience.”
We have deep skillsets around setting up tests and testing roadmaps from both a technical and business perspective. We help clients no matter where they are on the maturity curve. Call us and discuss your current challenges and we can provide further direction.
* Adobe 2014 Digital Marketing Optimization Survey. This year’s study had over 1000 respondents globally—60% from North America, 27% from Europe and 13% from Asia.
