Covestro Deutschland AG
Customer & challenge
Covestro is one of the world’s leading manufacturers of high-quality plastics. They are a global supplier across a variety of industries including transportation, construction, housing, sports, healthcare, electronics and electrical engineering. As of 2023, Covestro reached €14.4 billion in sales while employing 17,500 people across 48 locations. By leveraging digital technologies and processes, Covestro is producing much more sustainably today than it did ten years ago.
Driving this effort is extensive data analysis and simulations, driven by data from SAP. To continue this trend of sustainability, Covestro sought to improve accessibility of SAP data to an Amazon S3 data lake, where it could then be consumed by reporting and analytics tools. However, there was not a native way to transfer data from SAP to Amazon S3 directly.
“With the use of Xtract Universal, we have significantly optimized the availability of SAP data for data analytics and innovative applications. By providing SAP data more quickly, we can shorten the cycle of our analyses and respond more agilely to current developments.”
Dr. Jörg Janssen, Team Lead: Data Analytics, Covestro Deutschland AG
Solution with Xtract Universal
After testing several options, Covestro chose Xtract Universal for this data integration.
“I initially used the free trial version and was able to extract the first data from SAP and export it to our data lake within an hour,” recalls Andreas Bliss, Manager Data & Analytics Platform, Covestro Deutschland AG. “I only needed support when I encountered problems.” And even that support was excellent, says Bliss: “When I submit a ticket through the support portal, I get a competent response on the same day or, at the latest, the next day.”
Covestro utilizes Xtract Universal to transfer data from SAP ERP, SAP BW, and SAP IBP to their S3 data lake. After the data is transferred, it is processed further in Tableau (data analysis), Posit Connect (data visualization), Posit Workbench (data science), and other applications.
Requirement: Extracting data from SAP ERP, SAP BW, and SAP IBP into a data lake to make it centrally available for data analytics and innovative applications
Solution: Xtract Universal
Added Value: Cost-effective, automated SAP data transfer; SAP table integration in ten minutes instead of three hours, near real-time data availability
A major advantage for Covestro is the ability to use Xtract Universal for connecting not only to SAP tables but a variety of systems and objects such as SAP BW queries.
They also benefit from Theobald Software’s innovative interface, which is designed to be lightweight and easy to use. Bliss told us:
“After the spin-off from Bayer Group, we used the group’s data integration platform to transfer SAP data to a database for a transition period. Although the solution had a very extensive range of features, it was significantly more complex to operate than Xtract Universal. For example, integrating a relatively simple SAP table would take me two to three hours with a data integration platform, whereas with Xtract Universal, I can do it within ten minutes.”
Covestro also appreciates the value and low TCO of Xtract Universal. Andreas Bliss told us, “A different data integration platform would have been 15 times more expensive than Xtract Universal.”
“Developing a corresponding interface ourselves would also have involved more effort, as we would have had to operate, maintain, and service the interface,” adds Dr. Jörg Janssen. “Now we can focus on generating value from the data instead of dealing with the extraction.”
In the future, Covestro plans to also use Theobald Software’s Table CDC component for Change Data Capture to make updating large tables even more efficient.
“We see a growing need in the departments to access current data more frequently. Where daily data used to be sufficient, colleagues now need fresh SAP data every 15 minutes,” explains Dr. Janssen. “By being able to perform delta loads instead of full loads with Table CDC, we can update our SAP data stocks more frequently in the future, even in near real-time if necessary.”
Bliss concluded: “Instead of reloading a table with 30 million rows each time, we now only update about 1,000 rows that have changed. Instead of an hour, the loading process now takes only a few minutes.”
