Jumbo turns customers into fans using automation and AI

“Turning a customer into a fan means you have to improve daily and ensure that you’re even more relevant tomorrow. The data and insights we get from sales and our loyalty program is essential to this.”

Wendell Kuling, Manager of Data Science and Analytics at Jumbo, is explaining the thinking behind the supermarket chain’s recent digital transformation. Although the organization is currently the second largest supermarket chain in the Netherlands, maintaining that momentum is difficult because retail is notoriously competitive.

“We know that in order to turn even more of our customers into fans, we need to offer them a more personalized and responsive experience,” says Kuling. “But to do this, we needed to move from analogue processes to automation, infusing AI into our current processes to get insights faster.” 

Having previously migrated to Microsoft Azure two years ago, the organization already had a solid foundation in cloud technology on which it could begin transforming its processes, scale out where needed, and keep the environment stable in a cost-effective way.

Unlocking the potential of data

Owned by a family company which celebrated its 100^th anniversary in October 2021, Jumbo is a Dutch chain of supermarkets headquartered in Veghel in the North Brabant province. The company has around 100,000 employees that work across its 700 stores located in both the Netherlands and Belgium. 

Roughly six million households shop at Jumbo every year, making it a huge competitor in the current retail market. “Loyalty is very important not only for maintaining the number of customers who visit Jumbo, but also the overall sustainability of our supermarket chain. But loyalty isn’t a given,” admits Kuling. “As a company, we have to work hard to earn that loyalty and consistently live up to our customers’ expectations.”

Jumbo Extra’s, their loyalty program had just started and was producing increasing amounts of data such as when customers shopped and the products they typically bought. It became clear that if they could find a way to harness this data, the insights would be invaluable. 

So, the data science and analytics team embarked on a journey of digital transformation, to move away from labour-intensive processes and spreadsheets, and explore AI and machine learning.

Automation drives personalization and optimization

Jumbo faced the challenge of creating a seamless, omnichannel experience. “Our customers expect their experience with Jumbo to be relevant to them,” explains Kuling. “They want products and offers that are personalized to their needs both online and in-store.”

The organization had already been using a suite of Microsoft technologies; two and a half years prior, it had migrated everything to Microsoft Azure and established a data lake. This formed the foundation for the company to experiment with infusing AI and machine learning into its traditional processes. 

Jumbo used Microsoft Databricks and Microsoft Azure Synapse to manage, explore and prepare data for analysis. This data was then made accessible and understandable using the self-service analytics tool, Microsoft Power BI. These insights are helping to better understand customer behaviour and predict the products and services they would want.

“We see three levels on which data and AI impact our business: directly in the loyalty program; offering a personalized service; and using customer insights in decision making processes,” says Kuling. “By automating operations, using machine learning to understand customer behaviour, and analysing data, we have further improved our customer-centric, optimized experience.” 

One example of this is Jumbo’s ‘Remind Me’ service, which provides customers with an online list of recommended products that they may have forgotten. "We use machine learning to optimize this service and find patterns in a customer’s omnichannel purchases over time, so that we can then offer product suggestions. This personalization is key to building loyalty and is enabled by our intelligence-driven process.”

But undergoing a digital transformation wasn’t without its challenges. Jumbo’s employees were highly skilled at managing its core systems, but when it came to using cloud technology and machine learning, there were a lot of new techniques and methods to learn. This meant that the organization had to upskill engineers while rolling out new technology. 

“In order to stay ahead, we need to build skills faster than the market. So, we started the Enterprise Skilling Initiative to deliver online training to our engineers to help them advance their skills and career,” says Kuling. “This initiative required a whole new mindset and a shift in company culture to move from the traditional processes of the past, to automated capabilities enabled by Microsoft Synapse and Databricks.”

Powering the future of the retail sector

Once Jumbo’s employees had a good understanding of the technology, its possibilities were limitless. The business has been transformed both in terms of scalability and effectiveness, by fuelling many processes with data about its customers.

“Many of our core processes that were manual have now been modernized using predictive models using customer data,” explains Kuling. “From algorithms that can optimize the space and layout of a store, data processing that can select optimal product assortments, and machine learning that can forecast how many of each product will be sold, automation has rippled into multiple areas of the business.”

For the supermarket chain, the contribution of data and analysis is clear. Using the Microsoft suite, Jumbo has been able to extract value quicker from the vast amounts of data available from its loyalty program. “Using advanced analytics, we can now determine the most relevant offers to show to our customers,” says Kuling. “We can also serve our customers even better by acting upon the insights from the data from our loyalty program. In specific instances, conversion has gone up by a factor of 3.”

The continued feedback loop that this technology has created, means that Jumbo can continue to improve at an even greater pace. “It’s almost scary to see how quickly data and AI capabilities are becoming the new normal,” admits Kuling. “I believe the adoption level will only increase, not just at Jumbo, but across the retail sector.“