How Priorities, Policies, and Processes Can Make Your Data Strategy More Efficient
The prospect of embarking on an effort to improve master data is daunting to many businesses. Just the thought of trying to gather all the data in all the systems and build out a common view of it seems almost impossible to achieve.
Part of the issue is that organizations try to do everything at once and get bogged down in a tangle of interdependencies, which each seem critical. Soon the apparent inability to move on one aspect of the project before completing another aspect of the project devolves into an inescapable Catch-22.
Plenty of leaders just end up throwing their hands in the air, scrapping projects, and declaring the work and the funds sunk costs.
It doesn’t have to be this way, though. Taking the time to evaluate your efforts using our framework will help ensure that your team’s time and efforts are properly focused, that you’re getting the desired value out of the project, and that you can reduce complexity and interdependencies to make your project run more smoothly. This relies on three key pillars for master data:
Before starting down any technical paths, it’s important to lay out the goals your business hopes to achieve as the result of its master data efforts. Is the goal to improve your direct customer interactions? To reduce product returns or chargebacks? Is the company trying to improve its sales territory planning, or produce better financial reports?
The goals must be true business goals – goals that, if achieved, could impact the top-line revenue or bottom-line profit of the business unit or organization. It’s important to resist the urge to consider data-related KPIs as goals, as these are too narrow and don’t reflect the overall strategy of the business.
The business goals ought to be the driver of every decision you make with regard to your master data project. Once you understand what your goals are, you can start to prioritize the work that needs to be done.
Let’s say that your goal – one that many Dun & Bradstreet clients share – is to improve direct customer interactions. In this case, the data that is going to be most instrumental in doing this is your actual customer data: business names, addresses, contact information (executive names, phone numbers, email addresses, and so forth). You’ll probably also want to have a good sense of your customers’ purchasing history – where they’ve purchased, what they’ve purchased, and how many dollars their average purchase is worth.
Those data sets need to be your priorities – not your inventory data, product/SKU data, or payroll data. Allow your goals to drive your priorities, and allow your priorities to drive your work.
The benefit in limiting the scope of the data that you focus on is speed. By focusing on a handful of data sets rather than every data set in the company, you can start to deliver value from the improvement based on those data sets, building confidence within your organization that similar endeavors with other data will be worth it, not to mention gaining the benefit of having reached the goal itself.
Make sure your priorities are well-documented, communicated to all key stakeholders, and agreed upon. Then move on to developing policies that will inform how you reach your goals.
Once you’ve identified the priorities that align most closely with your business goals, the next step is to define the policies that will frame the work that you do and how you measure your success.
Policies can cover a broad range of aspects about your enterprise’s data – from how frequently it ought to be updated to what percentage of missing values is acceptable to which reports need to be available to whom. Policies may dictate the speed at which data should be accessible, identify situations in which manual intervention is required, and outline the expectations for compliance with appropriate legislation.
You may find that policies already exist within your organization that can’t be changed (or shouldn’t be changed). That’s fine; these broader existing policies can provide boundaries and guidelines for further refinement. Regardless of whether you’re starting fresh or working within existing policy constraints, developing and refining your policies can be thought of similarly to building a set of requirements; when you’re finished outlining your policies, you will have documents that spell out the expectations you have, and a way to measure your success against those expectations. If one of your policies, for example, is that 99.5% of customer addresses must include a postal code, then you have a benchmark against which to measure your adherence to that policy, regardless of which processes you use to try to implement it.
It’s important to remember that policies aren’t processes – they aren’t intended to explain how to do something, just what it is that needs to be done.
Processes are the steps – whether automated or manual – that your organization will use to implement the policies you lay out. Process definitions may include which third-party data vendors you’ll partner with, which software packages your organization will use, which database platforms you’ll leverage, the way that you’ll meet the specific policy requirements you’ve laid out, and even the processes your team will use for doing things such as data stewardship, backup, or other tasks.
Too often, organizations start by trying to define the processes they’ll use without first determining whether those processes support a high-priority effort or whether (and how) those processes ensure that the intended policies are carried out. This cart-before-the-horse approach quickly results in a breakdown – no one is sure which processes are most important, and no one, even if they think they are sure, can explain why.
Another stumbling block that organizations encounter when developing processes is that they fail to consider the humans who are actually performing, overseeing, or receiving the output of the processes. Developing a set of processes that can’t reasonably be accomplished, evaluated, or interpreted by your team doesn’t lead to success – it leads to failure, frustration, and another cycle of development. Circumvent this problem by keeping the affected teams in the conversation while these choices are being made. They’ll often offer some of the most important observations for making the processes effective!
Defining your processes may require going several levels deep into details, decomposing and further decomposing a high-level process. That’s okay: not every process will be trivial. But in the end, the result will be a set of processes that your team will look to implement, using partners, platforms, software, and other resources. And most importantly, each of those processes can be tied back to a policy, and each of those policies will be tied back to a prioritized effort toward defined goals.
Setting Yourself Up for Success
The benefits of having a well-defined and achievable strategy for master data far outweigh the challenges organizations perceive.
The “three P’s” framework can help organizations get to value with their master data efforts more quickly and effectively, but as with most challenges in the world of data, doing this once isn’t enough. After every iteration, go back to the start of the framework and reevaluate the business priorities. During the time your team has been focused on standing up your initial master data implementation, internal and external forces may have shifted the priorities of the organization, and the next set of efforts may need to reflect this new reality. A primary focus on priorities ensures that your organization and your implementation stay responsive to the changing business landscape.
That’s important, because getting a handle on your master data is already a complex and challenging undertaking, made more difficult when organizations fail to properly set goals and priorities prior to establishing policies and processes. By using our simple organizational framework, companies can improve the efficiency of their projects and successfully attain business value.
Learn more about Dun & Bradstreet master data – foundational data that unites systems, teams, departments, and businesses. Our data strategists and consultants can support your efforts.
George C. L’Heureux Jr. is Principal Consultant for Data Strategy at Dun & Bradstreet. Learn more about Dun & Bradstreet master data – foundational data that unites systems, teams, departments, and businesses.