Zalando Payments: Enhancing the customer experience with faster resolution of purchases across 20 different payment options
Explore Zalando Payments’ journey with SAP
Payment services provider Zalando Payments GmbH mitigated risk from using external tools to calculate its service providers’ payment allocations by developing a solution on SAP Business Technology Platform. Automatically modeling, running, and sharing the forecast using machine learning allows Zalando to deliver more-accurate and timely forecasts.
| Industry | Region | Company Size |
| Banking | Berlin, Germany | 250 employees |
automated process using SAP solutions.
faster and more efficient in forecasting management.
compliance with required financial services regulations.
Lead SAP BI and Analytics, Zalando Payments GmbH
Automating highly accurate payment allocation forecasts
Zalando Payments GmbH is the payment services provider for online retailer Zalando SE, playing a crucial part in how customers perceive their shopping experience at Zalando. Offering convenience and flexibility at the Zalando checkout, the company processes payments for more than 250 million orders per year for more than 50 million active customers.
Receiving incoming payments from customers across the globe, Zalando Payments needs to ensure compliance with national and international regulations. As a German financial services provider, the company is licensed for electronic money transactions and is regulated by Germany’s Federal Financial Supervisory Authority (known as BaFin), Deutsche Bundesbank, and European Union institutions.
The company’s forecasting process started in SAP S/4HANA, with data exported to external tools for calculation of payment allocations. This presented various risks, including the potential for media disruptions when passing on data, data disconnection from bank information master data, and noncompliance. The use of external tools also limited the company’s ability to change forecast calculations.
Looking to improve the quality of its forecast model for payment allocations, Zalando Payments wanted to heighten the model’s flexibility and adaptability against a number of factors. These included unforeseen fluctuations due to special sales, entry into new markets, new payment methods, and variations in the mix of new and old customers. It also wanted to support its service providers to shore up workforce resourcing allocations to meet contractual obligations for zero daily backlog.
Lead SAP BI and Analytics, Zalando Payments GmbH
Reducing overhead costs through better payment allocation forecasting
For a more-accurate and robust payment allocation forecast that delivers the flexibility to support the company’s growth, entry to new markets, and different payment methods, the company wanted the solution to reside entirely within its SAP software landscape. Seizing the opportunity for better forecasting through more-effective use of technology resources, Zalando Payments set to work building an innovative forecasting model on SAP Business Technology Platform (SAP BTP). The company implemented a business data fabric architecture enabled by the SAP Datasphere solution, allowing it to apply robust data governance and access real-time data. Additionally, the SAP Data Intelligence solution was used for Python scripting, SAP HANA Cloud was used for machine learning capabilities, and the SAP Analytics Cloud solution was used for more-accurate planning.
As shown in the figure below, in the new forecasting process for payment allocations, data from SAP S/4HANA and the SAP BW/4HANA solution is federated in SAP Datasphere on SAP BTP. The Python machine learning client for SAP HANA is used to determine the forecast, which is automatically modeled in SAP Data Intelligence on a weekly basis. The result is written back to SAP Datasphere, where reports are provided to service providers through SAP Analytics Cloud.
Lead SAP BI and Analytics, Zalando Payments GmbH
Achieving a zero daily backlog goal for unresolved and open payments
Thanks to the new model for payment allocation forecasts, Zalando Payments now has greater reach to improve forecast accuracy and reduce risk, helping lower overhead costs while complying with regulations for financial services companies. It has also helped improve relationships with service providers, as they can now more accurately allocate resources to deliver zero daily backlog on open payments.
What’s more, through working closely with SAP and managing the entire project in-house, knowledge gained from the project has been captured internally. A deeper understanding of the technologies has given rise to greater confidence within the organization to adopt machine learning and artificial intelligence tools embedded in SAP solutions.
Since going live with the new payment allocation forecasts, more than 50 million Zalando customers and their banks have benefited from faster resolution of purchases across more than 20 different payment options.
Ongoing use of machine learning capabilities in the machine learning libraries provided by SAP can help continually sharpen forecast accuracy, as well as open a pathway toward taking advantage of SAP AI Services in the future.
Lead Customer Experience, Zalando Payments GmbH
Inspiring new applications that go far beyond machine learning
As Zalando Payments continues its work on new solutions using machine learning and artificial intelligence, a key focus is designing a tool for risk assessment using Monte Carlo simulation techniques. The goal is to run enough simulations to produce different outcomes that mimic real-life results.
Elsewhere, the company is developing a solution to calculate the credit risk for its dealer partners and is creating use cases for the use of artificial intelligence in data management.
Robby Finke, SAP business intelligence and analytics lead at Zalando Payments GmbH, sums this up, saying, “Greater direct knowledge of how to enrich the team’s essential processes with SAP solutions is leading to new ideas and proposals for strengthening the business.”