Big data continues to be the new frontier for many businesses, but an organization can’t just collect all the data it can and expect it to do work for them. They need to carefully manage that data, and apply it in ways that make sense. Without a solid plan, data management can be messy, expensive, and worst of all—ineffective.
No matter how much data an organization can potentially generate, the end results are going to depend on how much it’s spending, and what they do with the data it procures. Too many companies are paying ridiculous fees and investing excessive time in their data management process, which is why a new philosophy—lean data management—is starting to gain popularity.
The core principles of lean data management
According to Clarkston Consulting, choosing to operate with lean data management is analogous to “going green,” as the primary goals will be operating as sustainably as possible while relying only on the resources truly needed for the task. Here are four core principles of how to do it right:
1. Ask the right questions.
First, the organization needs to be asking the right questions up front. No matter what kind of data it has, or what the intentions are, the end results are only going to be significant if the organization is asking questions that the data can speak to. Conclusions don’t magically populate out of the information gathered; they only start to take shape when put in the reference frame of an initial inquiry, such as “how much time do our customers spend using our product,” or the more complex, “are our customers passionate about the brand?” The sexiness and enormity of big data makes it tempting to ask too many the questions, but it’s better to narrow the target.
2. Only collect data for what you need.
According to Newforma, constraints are one of the most effective ways to keep project budgets under control. This, too, requires a narrower approach. Instead of collecting as many data points as the organization can gather, focus on gathering only the data needed to achieve the end goals. Sometimes that means closing out data that’s no longer useful, eliminating survey questions, or cutting software tools that are no longer needed to track certain metrics.
3. Reduce storage and operating expenses.
Data storage and management get cheaper every year, but they still cost money. An organization will pay a subscription fee for cloud storage platforms and analytics software used to monitor results and form conclusions, not to mention the salaries of analysts and data experts to work with those platforms. If an organization isn’t careful, it could hemorrhage money in this area. Don’t spend more than truly needed in any one area, and shop around for less expensive options. Even if the organization only saves a few hundred dollars a month, that will make an impact.
4. Continually measure the impact of your data
Everything in a business should be measured and scrutinized in terms of its return on investment (ROI), and data management is no different. The organization should be working to understand how much it has invested in big data, and what value it is able to get out of it—such as through improved marketing campaigns or increased client retention. An important question to ask: “Are you making more than you spend on these solutions?’
Is Lean Data Management Right for You?
So should your business start adopting lean data management practices? Some of these rules are general, and could apply to almost any business—or any department within that business. For example, reducing the expenditures for a given project and focusing on the return on your investments are always going to be valuable strategies.
However, when applied specifically to big data, these principles only truly start to show their value when dealing with significant volume. If an organization in only collecting data on a few metrics, or a few variables such as customer satisfaction or close ratios for sales, any effort it mounts to save costs will be marginally effective, at best. Instead, it’s the companies working with high volumes of data, on a regular basis, that need to start thinking about these lean fundamentals, and how to operate more efficiently.