De Bijenkorf is one of the most established brands on the Dutch high street and enjoys a long heritage in luxury retail, but much of the company’s recent success has come from its thriving e-commerce business.
“In 2009, it was decided to place greater emphasis on our online operations, which resulted in a significant uplift in sales from that side of the business. As part of our premium experience strategy de Bijenkorf has a focus on omnichannel retailing”. says Maarten Schuit, team lead of data engineering at de Bijenkorf. “There is so much data coming in, managing it quickly was the challenge.
A couple of years ago the company was partly relying on Excel spreadsheets for much of its data management. It happened that multiple versions of the same spreadsheets existed across the business. Different e-commerce teams also had conflicting definitions, how to measure sales or of what a (web) visit was, which resulted in internal arguments and uncertainty. The wish was also to integrate, data from things like Google Analytics and ongoing marketing campaigns into the existing Qlikview platform they were using.
“This meant there was ample room for improvement on how accurately we could see how the online business was doing,” says Maarten. “We knew a better approach to data was needed.”
After consulting with e-commerce specialists, de Bijenkorf decided to consolidate all its online business data into a central data warehouse and implement Tableau for visualization and analytics needs. This approach created a single source of truth for the first time, eliminating duplicate data sets and inconsistencies between different teams.
“Implementing Tableau quickly provided everyone in e-commerce with the fast access to consistent data sets that they needed,” says Maarten. “Every team was working from the same information.”
Establishing a single source of truth also had a significant impact on a number of other business processes, most notably the time spent in internal meeting times.
“Previously, meetings would take longer than necessary because of unclarity between team members over the data sets they were using and who’s was the ‘right’ one,” says Maarten. “Now everyone is working from the same set, which means meetings are a lot shorter!”