5 Essential Practices of the Data-Driven Organization
Andrew Edwards | January 30, 2012
An organization designed around intelligence gained from the data it collects about its customer interactions will have these five characteristics.
Analytics may be multi-channel these days - mobile, digital, social, email, and yes, even "web analytics" as once it was known. Analytics also means audience measurement via surveys and panel-based analytics as part of a site-ranking methodology. Whatever its latest form, analytics only helps when integrated into the creative and decision-making processes of the organization.
Too often analytics itself works in a small building off on the edge of the corporate campus. Occasionally a paper airplane flies into the executive window with a message from the analytics team. And the executive, on his way to a meeting, steps on the folded paper. It sticks to his shoe and by nightfall the janitor has swept it into a dustbin. So ends the influence of analytics in the non-data-driven organization.
Try to imagine an organization that is designed around intelligence gained from the data it collects about its customer interactions. This organization will have at least these five characteristics:
- Define what it wants to measure. Analytics deployment experts will work closely with marketers and business owners to understand the key business drivers for the organization. These drivers will be closely aligned with actual business success. For instance, branding will want close interaction. Services will want leads. Sites that rely on advertising will want volume plus careful segmentation. The analytics experts will turn these needs into reporting structures that can answer important questions - going well beyond the baseline reporting that comes before customization. As important as deciding what to measure is deciding what not to measure. The organization will avoid overcrowded report structures; will have realistic goals about what is measurable and why; and will make sure expensive tools are not overburdened with whimsical profiles that never get viewed.
- Deploy analytics tools expertly. That clamor you hear is the sound of the poorly chosen agency or miseducated IT team grabbing their pitchforks and torches as they try to kill this initiative - they may believe their jobs depend on "controlling analytics." But in the data-driven organization, they are quietly asked to go back to doing what they do best - create great content; make sure systems function. Perhaps the most common difference between the properly aligned team and the data-flunking team is the way expertise is deployed. The data-driven organization will not blame the tool. It will not permit content creators to claim they "also do measurement." They will audit their tags for completeness and functionality. They will have a neutral third party - expert specialists in analytics - deploy the tool with clarity of purpose, and without bias to anything but the truth about content success, campaign success, and ROI. They'll know that any other road leads to costly waste, bad politics, and ultimately, no insight.
- Analyze results and make recommendations. At the successful organization, reports won't gather dust. They will be shared with the right people in the organization - people who can make decisions and direct action based on the findings. The findings will be clarified by people who understand how the data was gathered, can vouch for its accuracy, and put the data into context. In the same meeting, an action plan will be devised. What content will live - or be retired? Which partners seem to drive the most successful visits? What campaigns pulled not just the most visitors, but the most desired actions? Resource reallocation may be in order. Cuts in a place least expected. Additional effort toward an area that measures well - perhaps one that had been undervalued. The ability to reprioritize and react to data is one of the most important characteristics of the data-driven organization.
- Create changes based on data. The developers may say they've got the best new wireframes this side of paradise, but somehow the numbers indicate rethinking the existence of that entire module. The content provider may insist on real estate where its click-throughs don't pay the rent. The data-driven organization knows how to say "no" to the ineffective; and to insist on creative, content, and campaign changes based on what came out of the analytics. Was that partnership successful? Yes? Why? The data will reveal why - and the next partnership had better incorporate some of the characteristics of the winner. The organization won't have time for slight tweaks to a tanking campaign - it will demand wholesale change if it's needed. It will stand by the numbers, and insist its agencies and other content managers stand with them.
- Measure again - and again. Done with that measurement exercise? Here's the next. Analytics is a cycle of definition, measurement, planning, informed improvement, remeasurement, definition, measurement…forever. The organization knows that optimization is never one-and-done. It should not be enslaved solely to IT or developer "release cycles." It should not come into the picture too late to have an impact. It should be as close to real time as possible without anyone twisting an ankle by being in too much of a hurry. Here is where you ask: how'd it go with those changes? Helpful? Better? Even a little better? OK, let's make it better still. Incrementally, perhaps…but always on the road to Betterville. The data-driven organization lives in a cycle of measurement and improvement.
Do you recognize your organization in the above? If so, you're in a great place. But chances are you'll notice where your team is off in some other direction - they're not on the proven path to improvement. And with soaring emphasis on digital enterprise, how long can you afford to be part of that team? Making the necessary changes may not be easy. Failing to take advantage of data is worse.
If your company is going to thrive in the digital marketplace, it will be as a data-driven organization.