By Nilly Essaides
Master data management (MDM) and governance may not sound sexy. However, they are rapidly becoming a key business imperative as the quantity and diversity of data grow, and companies face a rush of new software implementations. Without accurate data, it’s difficult to forecast and plan. The same goes for finance. If it cannot trust the data it uses, it cannot provide management and business leaders with reliable analytics and insight.
It’s no wonder that MDM improvement ranked as the number one technology project in The Hackett Group’s Finance Key Issues Study for 2017. If data is not captured and stored consistently and accurately from day one, it cannot be mined and analyzed later. We found that finance is often leading the charge toward better MDM and data governance because it’s sitting on sensitive data internally, and also releases some critical data externally.
Nor is MDM something that should be handled in the IT department alone. Data definitions have to be developed within the context required by internal “consumers” or customers. The Hackett Group’s 2016 Master Data Management and Governance Study found that successful MDM companies ensure collaboration between various data stewards in their MDM projects, including business owners and IT. They have formal MDM strategies and they dedicate full time staff to MDM.
What’s Driving MDM Awareness?
Three factors are putting MDM and data governance at the top of executives’ minds:
- Companies are collecting more data in different formats (internal/external, structured/ unstructured). It is imperative for data to be coded correctly upfront using standardized definitions so it can be retrieved faster and more accurately. At the same time, the speed of information is overloading existing database infrastructures. To handle the quick pace and quantity of incoming information, data warehouses need to be modernized and definitions standardized and simplified.
- There are more implementations of EPM, CRM and other applications. As many companies learned the hard way from past ERP implementations, getting data issues sorted out first helps avoid a world of problems down the road. There’s also growing trend toward the implementation of cross-enterprise applications. Source systems must learn to talk to each other – or at least communicate clearly with a central repository.
- Integrated planning and demand for data-driven insight requires more than just first-order data cohesion. Companies are increasingly in need of tighter definitions and governance around second-order data like Key Performance Indicators (KPIs). When departments report to management using different KPIs, it creates multiple pictures of performance.
How to build a successful MDM Strategy
The imperative to develop and implement effective master data management and data governance programs is only going to grow more urgent given that big data and new software implementations are the hallmark of the digital era. Here are important steps companies can take to ensure their MDM strategy is successful:
- Identify a business case for MDM. Data is a unique and valuable asset that drives business performance. It needs to be stored and managed carefully so it can be retrieved and analyzed to help make smart business decisions. This is a business issue, not a technology issue.
- Clearly define program scope. Don’t attack everything at once. Define a certain area of data to concentrate on, beginning with what elements to include; then build consistent definitions around it and describe all related processes. Use a targeted pilot to build momentum. A broad, enterprise-wide project can easily collapse under its own weight.
- Create a joint data stewardship model and assign full-time resources. MDM is not the responsibility of one department. It needs to bring the right people to the table to merge technical and business acumen and focus on the needs of the data consumer so that the data is useful in the business context. While this is a team effort, it must have a dedicated process leader who is in a position to see the full picture.
- Formalize a set of policies, procedures and communication processes. The policies will ensure consistency in definitions as well as change processes. They will clarify who needs to sign off on changes and how to communicate these along the way.
- Define value metrics that measure the effectiveness of the program. KPIs such as “percentage of data accuracy” are not enough. The metrics need to show how accurate data contributed to the business process – for example, how it helped close purchase orders, not just how long it took to close a vendor order.
Cleaning up the company’s master data management and governance framework is a precursor to any big data project, system integration or new implementation effort. Without first coming up with data definitions, a process for how data is stored, who can make changes to definition and who must approve those changes, the data warehouse becomes useless and analysis meaningless. Finance must ensure that data it uses is solid and consistent before it can provide management and business leaders with insight and advice to drive smarter decisions about enterprise performance.
Source: Generation CFO LI Group