On average, people should be more skeptical when they see numbers. They should be more willing to play around with the data themselves.
Nate Silver
‘That which can’t be measured can’t be managed’ has a loose corollary: ‘measurements you don’t fully understand won’t allow you to manage fully.’ To leverage data to ask interesting business questions, it must be well understood, timely, accurate, and consistent. The management of that data, the most critical and most important to fundamental business functions, is the practice of Master Data Management. Your ability to ‘master‘ data (forgive the pun) has fundamental real-world impacts on your business’s bottom line.
Classic examples of master data include Customers, Employees, Vendors, Suppliers, Parts, Products, Locations, Contact Mechanisms, Profiles, Accounting Items, Contracts, and Policies (from ‘Master Data Management’ by David Loshin). It’s the data that runs the fundamentals such as supply change management and accounting.
To fully take strategic advantage of master data and leverage it to improve operations, your Chief Data Officer should work directly with the Operations team to ensure that master data is timely, accurate, and reliable. Good Master Data Management enables a better Customer Experience (as measured by Net Promotor Score) and will help your Product team to better tell the value story of your product suite – as well as remove barriers to customer loyalty such as billing issues and negative product stories. Failure to manage your master data results in risks and impacts such as bottom-line dollar losses, loss of market, and regulatory challenges.
The following is a collection of ways that Master Data Management can improve your company with some examples I’ve encountered in my career.
1. Cross-Functional Visibility
In the Operations Management class I teach, I often use a quote attributed to Marshall Field in the 1890’s: ‘Give the lady what she wants.’ This simple mantra demonstrates two things (both revolutionary at the time):
- Mr. Field had a command of his data. He knew that his buyers were largely women
- Mr. Field had a relentless drive to leverage that data to make better-informed decisions about his sales floor
Without data that can be seen across departments and well understood (with linked attributes and lineage), the organization will not be able to make smart decisions for ‘the lady’.
Real-world example: Failure to expose relevant parts of a patient record across disparate parts of a healthcare practice can result in misdiagnosis of problems. I once ran into this with diabetes diagnoses at a large health insurance company that I was helping on a management consulting contract. Diabetes often has comorbidities with mental health and related physical health issues (such as vision). By working as a team with better master data, we were able to achieve better results for the patient.
2. Data Accuracy Enhancements
In healthcare, accurate data in the patient record is literally a question of life and death. In retail, the supply chain relies on SKU data to be accurate and consistent for ordering and billing to work correctly, with poor quality data creating order delays, invoicing issues, and customer disappoints.
Master Data Management helps keep that data straight with correct customer address information, and the correct patient formulary and reduces the chance of delays due to improper location information.
Another example is a previous project I led with a related problem. A large local retailer received inventory into its warehouse, which was more often than not, in different unit quantities than what is used at its stores, resulting in a lot of labor breaking boxes with two dozen units into four batches of six (for example). To make matters worse, the data feed for that element was inconsistent and often inaccurate resulting in additional blood and treasure spent cleaning the data feed when the problem became too big to live with.
Solution & Results:
We developed a plan to Master the SKU information related to case count. That now accurate data could then be used to influence the supply chain to repackage the fastest velocity SKUs resulting in massive labor savings.
Failure Modes are the way your product, platform, (or… anything really) might fail. They are systematically reviewed by the cross-functional team you assemble for three things: severity, the likelihood of occurrence, and detectability. For example, a nuclear blast would score as very severe, low likelihood of occurrence (a debatable point), and very, very detectable.
3. Regulatory Compliance
Table stakes in healthcare are the ability to accurately and securely store data to that it’s available in a timely manner to providers and vendors with an approved use case. Examples include diagnosis data, formulary information, and prescription data. Utility companies must store and maintain usage data, meter details, and transmission line information. Master Data Management allows that data to be timely, accurate, and consistent. Failure to master that data can cause your friendly, local regulator to find a new home-away-from-home in your C-Suite.
Another example: a large utility company I worked at could not get a well-known Midwest landmark building to invoice through batch billing due to the unmastered Meter Type information resulting in loss of revenue and (more importantly) loss of reputation as the billing became a local joke.
This was early in my career and we were able to solve the problem through a lot of “unnatural acts of data heroism.” We were later able to apply Master Data Management techniques to solve the data quality issues on the front and many similar problems downstream
4. Information Confidentiality and Security
Confidential and restricted-use data retention rules are often a matter of make-or-break reputation as well as regulation. For example, HIPAA standards require specific protections for use and storage. Violations can incur fines and almost always a loss of reputation. Understanding the data definitions is critical to be able to apply those regulations correctly and to ensure that the treatment of the medical condition is accurate.
Example: At a large pharmacy benefit manager, I was privy to the results of poor Master Data Management when a manually manipulated file resulted in a patient getting (and nearly ingesting) the wrong prescription. The ability to manually touch that data should never have been allowed and the loophole was quickly closed. The key records were quickly mastered when the extent of the problem was discovered.
Understanding the strategic need for Master Data Management is critical to improving your company’s operations, creating opportunities for customer delight, and avoiding helicopter-parent-like regulatory oversight. Making sure that Master Data Management is part of your companies, strategy, vision, and growth plans is critical to success. Doing so will improve the cross-company visibility, improve data accuracy, help with regulatory compliance, increase privacy and security, and reduce unnatural acts of data heroism – resulting in bottom-line improvements.
Larry Odebrecht published this article first on LinkedIn here.
