Sergey Nivens - Fotolia
Enterprise data governance isn't just managing the data an organization company possesses, it's also key to managing the data supply chain, according to Charles Link, director of data and analytics at Covanta.
Link detailed his views on data management during a technology keynote at the Talend Connect 2020 Virtual Summit on May 27. Executives from other Talend customers, including AutoZone, also spoke at the event.
Covanta, based in Morristown, N.J., is in the waste-to-energy business, operating 41 facilities across North America and Europe. Data is at the core of Covanta's operations as a way to help make business decisions and improve efficiency, Link said.
"We're never just pushing data; we're never just handing off the reports," Link said. "The outcome is not data; it is always a business result."
Link said he's often observed that there can be a disconnect between decision-makers and the data that should be used to help make decisions.
To help connect data with decisions, "you really need both the data use and data management strategy to drive business outcomes," Link said.
Enterprise data governance strategy defined
Link defined data use strategy as identifying business objectives for data and quantifying goals. The process includes key performance indicators to measure the success of data initiatives.
An enterprise data management strategy, on the other hand, is more tactical, defining the methods tools and technologies use to access, analyze, manage and share data, he said.
At Covanta, Link said enterprise data governance is essentially about the need to have what he referred to as data supply chain management.
Charles LinkDirector of data and analytics, Covanta
Link defined data supply chain management as data governance that manages where data comes from and helps ensure consistent quality from a reliable supplier.
For that piece, Covanta has partnered with Talend and is using the Talend Data Fabric, a suite of data integration and management tools that includes a data catalog that helps enable data supply chain management. With Talend as the technology base, Link said that his company has deployed a central hub for users within the organization to find and use trusted data.
"There is now a shared understanding across business and IT of what our data means," Link said. "So now we trust the quality of the data we use to operate our facilities."
The chaos of data demands driving AutoZone
For auto parts retailer AutoZone, managing the complexity of data and overcoming data challenges is a foundation of the company's success, said Jason Vogel, IT manager of data management at AutoZone.
AutoZone has 6,400 stores and each store carries nearly 100,000 parts. In the background, AutoZone is moving data across its disparate data hubs and stores, making it available to the company's business analysts. Data also helps ensure that AutoZone customers can get the parts they need quickly.
"We have 20 different types of databases -- not instances, types," Vogel emphasized. "We have thousands of instances and Talend serves as the glue to connect all these systems together."
Vogel noted that AutoZone is looking to expand its real-time data processing so that it can do more in less time, getting parts to its customers faster. The company is also looking to expand operations overall.
"The only way to accomplish that is by moving more data, having more insight into how data is used and accomplishing it all faster," Vogel said.
Many organizations continue to struggle with data
AutoZone isn't the only organization that is trying to deal with data coming from many different sources. In another keynote at Talend Connect, Stewart Bond, research director of data integration and data intelligence software at IDC, provided some statistics about the current state of data integration challenges.
Bond cited a 2019 IDC survey of enterprise users' experience with data integration and integrity that found most organizations are integrating up to six different types of data.
Those data types include transaction, file, object, spatial, internet of things and social data. Furthering adding to the complexity, the same study found that organizations are using up to 10 different data management technologies.
While enterprises are managing a lot of data, Bond said the survey shows that not all the organizations are using the data effectively. Data workers are wasting an average of 15 hours per week on data search, preparation and governance processes, IDC found. To improve efficiency, Bond suggested that organizations better manage and measure how data is used.
"Measurements don't need to be complex; they can be as simple as measuring how much time people spend on data-oriented activity," Bond said. "Set a benchmark and see if you can improve over time."
Improving enterprise data governance with data trust
She noted that it's important to measure the quality of data, to make sure that organizations are making decisions based on good information. Talend helps its users enable data quality with a trust score for data sources, as part of the Talend Data Fabric.
"When you think about what Talend does, you know, you think of us as an integration company," Bemont said. "Quite frankly we put equal, and maybe even in some cases more, importance on not only just being able to have a lot of data, but also having complete data."