In the broad ecosystem of modern banking, leasing is often treated as a specialized product line with its own rules and operational needs. It sits somewhere between a standard business loan and a rental agreement, governed by its own complex logic of depreciation, economic ownership, and tax benefits. Because of this complexity, leasing divisions frequently evolve with their own processes and sometimes their own technology stacks.
For years, the leasing division of one of Europe’s largest banks operated somewhat separately from the rest of the institution’s core digital platforms. While other parts of the bank migrated to private clouds and standardized API marketplaces, the leasing department continued running key parts of the landscape on bare-metal virtual machines and self-hosted components. The setup worked, but it required ongoing attention and made it harder to plug leasing into the same customer journeys as other products.
We sat down with Shantonav Sen, a Principal Solution Architect at Ximedes to discuss the 18-month effort to close that gap. We talked about the technical realities of asset finance, the challenge of aligning Dutch and Belgian requirements within a single code base, and what it takes to move an established platform toward a more integrated, cloud-aligned banking environment.
Here is our conversation.
Question: I want to start with this metaphor of the "island." In large organizations, we often hear about silos, but you’ve described this project as something more extreme: almost a completely separate entity from a tech perspective. What did that landscape actually look like when you took the project on?
Shantonav Sen: It was a technological island. Usually, when we talk about silos, we mean teams that don't communicate. Here, we meant infrastructure that wasn't connected in the same way as the rest of the bank’s platform ecosystem.
The client has a very mature, standardized technology landscape: private cloud foundations, shared services, the customer portal, and the mobile app. But the leasing department reflected its own history. We are talking about self-provisioned bare-metal virtual machines and self-hosted software like Keycloak for identity management, which can work well, but in this case, the versions in use were far behind current levels.
Question: Fifteen versions? That raises concerns.
Shantonav Sen: It creates a significant maintenance effort. You spend a lot of time patching and keeping systems current instead of focusing on new capabilities. But the bigger issue was the limited integration. Because it wasn't aligned with the bank's broader platform, it didn't automatically benefit from the same security and compliance patterns.
From a business perspective, the impact was also clear. A customer logging into their business portal couldn't easily see leasing next to loans. Leasing capabilities were less discoverable in digital journeys, and for many use cases, the process still relied on human interaction instead of self-service.
Question: For the uninitiated, why is leasing so distinct from a standard business loan? Why couldn't they just copy-paste the loan architecture?
Shantonav Sen: Because a loan is purely about money: you borrow €50,000 and repay it with interest. Leasing is about assets. It’s asset finance. When you lease a truck, a machine, or a solar panel array, the asset itself is part of the risk model and the data model.
In a financial lease, the customer often gets economic ownership immediately (so the asset sits on their balance sheet and they can access certain tax benefits), while legal ownership remains with the bank until the contract is completed. This structure can help customers preserve cash and can enable different pricing dynamics, especially for certain asset categories.
A standard loan system doesn't need to know whether you bought a Volvo or a Tesla. A lease system does. It needs the make, model, supplier, depreciation logic, and residual value. That’s why the architecture and data requirements are fundamentally different.
The Tale of Two Countries
Question: You mentioned that part of this transformation was harmonizing the business across the Netherlands and Belgium. How distinct were these two markets?
Shantonav Sen: They were strong in different areas, which made alignment valuable.
In the Netherlands, the leasing division performed very well in larger-ticket contracts and had significant coverage, particularly in high-value deals. At the same time, with the existing approach, it was harder to serve the SME segment in a scalable, self-service way, like a local bakery needing a van or a startup leasing IT equipment.
Belgium was almost the reverse. The organization was strong in small-ticket, high-volume business, but had less focus on the largest deals.
Question: So the goal was to build one platform that could handle both ends of the spectrum?
Shantonav Sen: Exactly. We wanted to expand coverage in both countries: bring more high-volume capability to the Netherlands and strengthen the ability to support larger, more complex deals in Belgium.
Technically, this is challenging. You want a unified customer journey, a single code base, and a consistent architecture, while still supporting local legal requirements. We call these "country-specific hooks."
Question: Can you give me a concrete example of a "hook"?
Shantonav Sen: Signatures are a good example. In Belgium, digital signing is well-established, so we built a journey where a customer can select an asset, get approved, and sign digitally in the portal.
In the Netherlands, for specific legal and legacy reasons, a "wet signature" is still required for certain documents. Even if the digital journey is strong, at the end, the customer may need to print, sign, scan, and upload.
If you build two separate apps, you double the ongoing maintenance. So we designed one pipeline: the core logic (credit checks, asset validation, pricing) is shared, and at the final signing stage, the flow branches based on country requirements. This keeps the core consistent while isolating local variation to the edges.
Architecture of the Mainland
Question: Let’s talk about the solution. You moved from bare metal to the bank’s private cloud. What does the new architecture actually look like?
Shantonav Sen: We moved to a cloud-native microservices architecture that integrates directly with the bank’s enterprise services. We call it the old leasing platform.
It starts with discovery. We built an orientation tool, a lease calculator on the open web. You don't need to be logged in, and you can explore scenarios like “Mercedes Sprinter, €40,000, over 48 months,” and get an indicative monthly price. For leasing, that kind of visibility is important because it helps customers understand the product before they enter a formal journey.
When the customer clicks "apply," we transition them smoothly into the secure environment: the customer portal. That’s where deeper integration happens. We check eligibility, consult the trade register (KVK) to confirm the business has been operating long enough, and we typically exclude very new businesses for risk reasons.
Question: And the backend? How are you handling the complex asset data?
Shantonav Sen: We integrate with the Central Asset Catalogue system, the bank’s enterprise solution for asset configuration. Instead of maintaining separate asset definitions ourselves, we consume the bank’s central asset definitions.
We also integrated with Baker, the bank’s workflow management system, which orchestrates the lifecycle. If a deal is "instant", for example, under €150,000, it can flow through automatically. If a deal triggers a risk rule (inactive credit facility, low prospect rating, and so on), Baker routes it to a human underwriter.
We implemented a "Prospect Rating Override" capability so the system doesn't only take binary accept/reject decisions. Standard cases can be handled automatically, while edge cases are flagged for expert review.
The Velocity of Money
Question: You mentioned "Self-Service Lease" for deals under €150,000. In the world of enterprise finance, speed is often promised but rarely delivered. How "instant" is instant?
Shantonav Sen: Decision-making can be immediate. If you have an active credit line and meet the criteria, approval can happen right away.
This matters for SMEs. Business owners often can't wait weeks to know whether they can move forward with a new delivery van or equipment purchase. We can use the customer’s existing relationship with the bank to extend a suitable level of trust to leasing, which reduces friction.
We also introduced a lease specific upper limit. It's a complex risk concept, but it allows us to treat a group of companies (a parent and its subsidiaries) as one risk entity. That can simplify approvals for more complex corporate structures that would otherwise require more manual processing.
The Future is Automated
Question: You’ve spent eighteen months building this foundation. The platform is connected. What happens next?
Shantonav Sen: Now that data is flowing, we can focus on further automation and customer experience improvements.
One area is AI for financial analysis. Today, a lot of risk assessment relies on structured data. We want to be able to take customer financial documents (PDFs, spreadsheets) and use AI to extract the parameters needed for risk scoring—while still using the bank’s enterprise models.
We’re also looking at assisted journeys for customers who drop out of the self-service journey. If someone gets stuck, we should be able to route them quickly to the right support.
And we’re expanding the asset catalog so we can support more custom lease deals: assets that may not be part of the standard catalog, such as specialized solar installations or niche industrial equipment.
Question: It sounds like the "island" is becoming a hub.
Shantonav Sen: That’s the goal. We moved from a legacy setup to a modern platform, reached Security Maturity Level 3, integrated with the bank’s API marketplace, and went live in two countries. With the foundation in place, we can focus on improving flow and delivering more value through automation.
If you’d like the concrete “what we built / how it was rolled out / what it enabled” view, Ximedes has a short case study on the Direct Lease Application and associated portals available on their website. It provides additional context on how the solution supported expanding into the SME segment, alongside implementation notes and outcomes.
