As the world’s leading retailer—supporting approximately 10,500 stores and clubs and 2.3 million associates worldwide — it’s no surprise Walmart has a vast amount of data to process. With trillions of transactions come challenges around data centralization, preparation, and consumption. The Walmart finance organization has chosen to standardize on Microsoft Power BI to create an agile, scalable, and cost-effective reporting and analytics solution.
“With Microsoft Power BI, we've established a “semantic model library” covering massive amounts of financial data. It gives associates easy access to the full breadth and depth of Walmart data—all the way from executive leadership to individual analysts—and enables them to powerfully analyze the data in a very quick manner.”
Micah Dail, Director, Finance Data & Analytics, Walmart
Centralizing, democratizing, and optimizing data
Democratizing data and making it available to end users is a massive challenge when dealing with trillions of records of data from multiple underlying sources. However, like with so many challenges an organization faces, the solution was found by looking within and leveraging the talent of Walmart’s associates. In this case it was Mason Whitehouse and Micah Dail, Director, Finance Data & Analytics at Walmart, with 14+ years of combined experience working with Microsoft Power BI. The two knew Power BI could become a key component to Walmart’s data analysis, reporting, and standardization success.
With the support of Finance leadership, Dail and Whitehouse set out to revolutionize self-service access to financial data leveraging a scalable and cost-effective business intelligence platform that is as simple as dragging and dropping fields and applying filters. The result is a Power BI “Semantic Model” that gives analysts an easy way to gain actionable insights into daily journal entry transactions within the U.S. The automation of reports using this model drastically reduced the amount of time analysts spent on gathering data, providing the ability to drill-down to the most granular level of data available. The centralized nature of the data model is eliminating reporting silos and is removing the need for non-value adding reconciliation between disparate reports.
As Dail puts it, “We can’t have spreadsheets flowing around the business with different numbers. With Power BI, we have a single-source reporting model that allows associates to drill down into the most relevant details using their preferred method and enables them to manage their part of the Walmart business more effectively.”
Moving data from multiple sources and translating into rich BI
A large and complex organization like Walmart has datasets that span hundreds of billions of records from a collection of systems and disparate technologies.
Walmart’s Finance Data & Analytics team has worked to unify disparate data sources, including internal and external technologies, through a series of layered steps. Raw data is curated, cleansed, and enriched in a data warehousing solution before being presented within Power BI.
Empowering associates across the enterprise with single-source reporting
To effectively summarize billions of records in a way that helps leadership digest the information, Walmart’s BI solution presents aggregated results using Power BI’s in-memory layer. In-memory data is extremely performant and enables leadership to quickly consume and interact with high-level views of their business.
As analysts dig deeper and into the more granular levels of the data, unbeknownst to them, the model seamlessly switches to DirectQuery mode and efficiently queries billions of rows of data to fetch the underlying results. Both In-Memory and DirectQuery modes work together to enable users to drill from a summarized view down to the underlying transaction details within a fast and seamless experience.
Furthermore, instead of clearing and then reloading massive amounts of the same data into the model each day, Walmart leverages Power BI’s incremental refresh option which only loads the most recently changed data from the underlying database. This process significantly reduces the amount of time needed to refresh the model with the latest data.
When interacting with the data, users can leverage the interface they are most comfortable with, whether through the Power BI web client, Power BI Desktop, or in Excel. “The flexibility in which a user can interact with their data in Power BI has proven to be a critical enabler of our finance transformation process by helping the team transition from legacy Excel-based systems to Power BI,” says Patrick Keehn, Senior Product Manager, Finance Data & Analytics at Walmart. And with a Power BI Premium workspace, other reporting tools downstream of Power BI can leverage semantic models as a data source because of Power BI’s open-platform connectivity and compatibility with external technologies.
Furthermore, Power BI Premium enables a vast ecosystem of BI tools that can be used to aid development such as Tabular Editor, a third-party tool which enables advanced semantic modeling features beyond what is available with Power BI Desktop alone. Power BI Premium also simplifies deploying, managing, and sharing models.
”One of the key criteria in selecting our solution was knowing that Power BI works so well with a vast number of partners—not necessarily just Microsoft products,” says Dail. “At Walmart, we use a myriad of different vendors in our technology stack, so having a tool like Power BI that can bring everything together was critical.”
Regaining valuable time
“With Microsoft Power BI, we've established a ‘semantic model library’ covering massive amounts of financial data,” says Dail. “It gives associates easy access to the full breadth and depth of Walmart data—all the way from executive leadership to individual analysts—and enables them to powerfully analyze the data in a very quick manner.” The Finance team eliminated redundant reporting efforts and made consumption easier and faster so that associates could utilize the time previously spent on collecting data on more strategic efforts.
Exploring future business intelligence endeavors
Going forward, Walmart plans to leverage existing models as best practices while it expands its model library to include all the different facets of the business. The company’s goal is to enhance overall operations across the enterprise by empowering associates with the data it needs to be able to do so.
Find out more about Walmart on Facebook, Twitter, and LinkedIn.
“One of the key criteria in selecting our solution was knowing that Power BI works so well with a vast number of partners—not necessarily just Microsoft products. At Walmart, we use a myriad of different vendors in our technology stack, so having a tool like Power BI that can bring everything together was critical.”
Micah Dail, Director, Finance Data & Analytics, Walmart
Follow Microsoft