For all the hype, the promise of the data driven company hasn’t played out for most firms. Ninety-four percent of businesses suspect that their customer and prospect data is inaccurate—and not because they haven’t collected enough of it. Worldwide, researchers estimate that organizations generate at least 7.5 septillion (that’s 7,500,000,000,000,000,000,000) gigabytes of new data every day. Nor has the potential of big data failed because companies can’t act on it. The analytics and business intelligence software market is a $24.8 billion industry and growing at 10.4% each year.
The problem is the data and insights aren’t accessible when you need them. Too often it’s siloed—making it difficult to access and hard to keep up-to-date across all of your tools. And because you can’t keep it up-to-date, data decay inevitably sets in. Many companies have invested in building out a data team to make clean, consistent data available to executives for reporting (which is a massive challenge on it’s own). Meanwhile, the people interacting with the data every day know what they have is junk. So, instead of being data-driven, these people are being data-driven crazy. So much for data being the new oil.
Where did it all go so wrong? We’ve identified the biggest barriers to getting the most out of your data. Plus, we share the steps you can take to turn your organization from data-driven crazy to data-driven company.
The 3 biggest barriers to becoming a data-driven company
1. Treating the symptoms, not the disease
Many aspiring data-driven companies get caught in a cycle of fixing the same data errors over and over again while neglecting to address the root cause. It may seem obvious, but it’s an easy trap to fall into. Renowned data expert Tom Redman puts it this way: “The hard part is you have to look in the mirror and you have to admit yourself, ‘Oh my goodness, I had great intentions, but I spend a third of my day cleaning up bad data. It never occurred to me to eliminate the source. I have been contributing to the problem.’”
Once you swallow your pride, finding and eliminating the root cause of an error is much easier than correcting that same error every time it occurs. Think of this as the rule of compounding data quality.
2. Competing data definitions
Another key barrier to becoming a data-driven company? When departments align on the same metrics but define those metrics differently. One’s “lifetime value” includes the cost of acquisition. The other doesn’t. They’re on the same page, but reading things totally different, and drawing the appropriate—albeit divergent—conclusions.
When departments devise well-intentioned approaches to fixing their data discrepancies but don’t coordinate, it goes something like this. “One department decides they need to modernize and automate their client-oriented processes,” says Redman. “They think through what they want to achieve, re-define their data, and re-design their processes. Cut to another department—they also decide they need to modernize, so they too come up with a new approach for their needs. Pretty soon the company has 27 different approaches to client data.”
If you don’t have a unified data approach—which typically needs to come from the top down—you can quickly end up worse off from where you started.
3. Expecting new technology to solve the problem on its own
While new tech can be a wonderful tool for getting your data in order, putting your data into a shiny new container won’t magically clean it. “There’s an old saying that goes if you automate a process that makes junk, you just make more junk,” says Redman. “If you want good data, you’ve got to start creating it correctly.”
Becoming a data-driven company takes more than just cool tech—you also need to put in the work of creating a unified data policy, agreeing on (and defending) your definitions, and periodically cleaning up old errors.
4. Creating a data swamp
Often, when you capture everything, you achieve nothing. As Ross Mason explained in a recent Syncari webinar, “Too often when people start down the path of unifying their data in a data warehouse or data lake, they dump way too much into one place, hoping to glean future insights out of this store of information. Unfortunately, it ends up creating a data swamp that nobody goes near because it’s too complicated—nobody understands how to get in and get what they need.”
The solution? Break things down into manageable chunks. Narrow the scope of your data warehouse and who’s allowed to interact with it, and even go as far as naming it something specific and publishing a manifesto for what it’s for and why it exists.
How “data provocateurs” can realize the dream of the data-driven company
But there is hope. Redman is a firm believer in the power of “data provocateurs” to prompt their departments and companies to change. “I define a data provocateur as someone who is the first in their department to address data quality properly,” he says. And a data provocateur can be anyone—they don’t have to hold an advanced degree in data science or even have data in their job description. They just have to recognize that there’s a problem and take the initiative to start coming up with solutions.
To troubleshoot your data problems, Redman recommends that aspiring data provocateurs try the following:
- Figure out who is using your data.
- Find out what they need.
- Determine whether you’re delivering or not.
- If you’re not delivering, fix your processes.
- Regularly reassess.
It’s not complicated—but that doesn’t mean that it’s easy. Finding root causes and eliminating them is better for your data health in the long term, but it does require more work up front. Luckily, data provocateurs don’t have to go it alone.
“You may not do everything but, at least interrupt the status quo enough to get the company on the right path,” says Redman. What data provocateurs do is prompt the rest of the organization to take a critical look at their data habits and provoke them to change. “To me, these are the real heroes in data quality,” says Redman.
Often, all it takes is one brave soul to get the ball rolling.