Establishing a single source of the truth!
Large Electric Utilities have big data-management challenges related to everyday activities such as:
- Which plants do we operate, and when, to meet customer demands throughout the day?
- How do we assure we have enough power to meet demand?
- How do we assure we have enough capacity in the wires to deliver the power?
- How do we establish rates to assure that customers pay a fair and appropriate amount?
- How do we work with regulators to show them we’re managing data appropriately and making good decisions?
While most utilities rely on sophisticated tools for modeling and analysis, it’s common to find utility staff using Microsoft Excel when preparing reports and making business decisions. Sometimes, it’s only used for data entry and reporting, but more often than you might think, the actual modeling and data manipulations are also performed in Excel spreadsheets.
xOverTime has been working with one large electric utility for over 6 years, helping them to improve their business data management, the design and organization of their spreadsheets, but also in implementing xOverTime to assure better data governance. Over the years, we have established a mantra for approaching a new project: “Separate the Inputs, Algorithms, and Outputs”. Data associated with spreadsheets often gets jumbled, and often lost, in a workbook, so it’s important to identify:
“What Data is coming in, and where and how? Where are the calculations being performed? What Data is going out, and how will it be used?”
A common problem can occur when, for example, Person B collects input data from Person A in order to perform a calculation and generate results they are responsible for. Then, Person C collects the same input data for their own calculations, but from Person B, instead of going to Person A directly. Suddenly, it appears as if there are two versions of the true input dataset, one from Person A, and the other from Person B. Person A may not even know Person C is using their data. So, when Person A updates the dataset, and Person B happens to be out for the week, Person C won’t know they’re not using the right data. In this organization, there is not a single true source of the data that Person A provides. Data mismanagement like this is way too common and can lead to bad decisions based upon data that is not up to date.
When implementing xOverTime, we want one source, or “owner”, for each data element. When the owner updates their data, everyone using it remains assured of working with the true, current version of the dataset. Additionally, by using xOverTime’s patented naming convention for data in a database, all users have an intuitive way to store, find, and share the data without compromising the source or structure. Having one owner for each element, and each component of the process, improves data-governance, and it all starts with identifying, for each workbook: What are the inputs? What are the outputs? And what are the algorithms used?