Regulatory reporting has long been defined by highly specialized jurisdictional knowledge, templates, spreadsheets, and a significant part of the compliance budget. Regulators publish new requirements, firms interpret them independently, technology teams build extraction and transformation layers, and operations teams reconcile outputs before pushing formatted datasets to supervisory authorities.
RegTech Insight sat down with regulatory reporting innovator REGnosys’ CEO Leo Labeis, to discuss their role in re-architecting regulatory reporting from firm-specific interpretation of regulatory rules to shared, machine-executable standards and potentially, over time, transition from the current template-driven “push” model towards a more granular, event-oriented architecture where regulators pull the data needed from a unified pool, a concept Labeis describes as an industry destination – “I think of it as a target state we should aim for,” he says.The Building Blocks: CDM and DRR
Working backwards from the inefficiencies of large-scale regulatory change programmes such as MiFID II, Labeis describes the foundational problem: firms were duplicating the same interpretative work in parallel. “Whenever there is a big reporting change like that, the way the industry is going to approach it is very inefficient,” he recalls, reflecting on his time leading MiFID II transition work in the industry. “Pretty much all of my peers across the industry were working on exactly the same thing.”
In its early years, REGnosys worked closely with the International Swaps and Derivatives Association (ISDA) to support the development of the Common Domain Model (CDM), providing much of the underlying technology used to express the model’s data structures and lifecycle logic. A key component of that work was the creation of a domain-specific language (DSL) Rosetta, subsequently renamed RUNE DSL (RUle Natural Expression) and contributed to the open-source governance framework under the FINOS (Fintech Open Source Foundation).
RUNE DSL is designed to express financial data models, operational logic, and regulatory reporting rules in a standardised, machine-executable form. Functionally, it allows industry participants to define the structure of business data, map internal systems to standardised models such as the CDM, and encode the calculations and transformations needed to generate regulatory reports. Because the logic is written in a technology-agnostic, human-readable language and can automatically generate executable code, RUNE separates regulatory and business logic from proprietary system implementations, enabling firms to share, test and implement reporting rules consistently across different platforms and institutions. “The building blocks that you use to assemble an interest rate derivative or cross-currency swap, you should be able to use exactly the same building blocks to handle a securities lending transaction,” explains Labeis. This design principle has allowed the framework to extend well beyond derivatives.Adoption: Supply Meets Demand
Today, the CDM ecosystem spans multiple asset classes. Industry groups including the International Capital Market Association (ICMA) and the International Securities Lending Association (ISLA) have extended the model into repo and securities lending markets.
According to Labeis, much of the required infrastructure already existed when those extensions began. “When the other trade associations… looked to extend the model to their market, effectively 80 or 90% of it was already built.”
Adoption dynamics can be viewed through a supply-and-demand lens. On the supply side, the infrastructure has matured. The modelling language, tooling and open-source components are established. On the demand side, the regulatory environment continues to generate momentum.
“If there is one constant in the regulatory reporting space, it’s change,” Labeis says. The cumulative impact of continuous regime updates is gradually shifting the economics for firms. “Every straw that you add effectively breaks more camels’ backs,” he adds, describing the growing pressure on institutions to rethink their reporting architecture.
In markets dominated by a small number of global broker-dealers, adoption by even a handful of firms can have disproportionate influence.
SFTR as a Stress Test
The Securities Financing Transactions Regulation (SFTR) provides a useful case study. Dual-sided reporting requirements and complex reconciliation processes have made SFTR one of the more operationally demanding regimes in recent years.
While SFTR looks like a natural next candidate for the CDM/DRR approach, it is important to distinguish between the underlying model and the reporting rulebook. On the CDM side, the distance appears relatively short: the model has already been extended into repo and securities lending, so much of the lifecycle and collateral logic needed for securities financing is already present. “The gap analysis is there,” says Labeis. “We just need to tweak the CDM here and there.”
A DRR implementation will require more work given SFTR’s specific reporting logic, including more detailed collateral and reuse reporting. However, SFTR has been designed to be consistent with EMIR (where feasible) which allows for a meaningful amount of EMIR architecture, mappings and reporting logic patterns already implemented to be reused for DRR.
In that sense, SFTR represents an opportunity: a chance to demonstrate how shared digital models could simplify future reporting implementations.
Lowering the Barrier to Entry
One of the most persistent concerns among firms evaluating CDM adoption is integration complexity. Many capital markets institutions still operate legacy estates built over decades, often combining proprietary schemas with older programming environments.
Labeis acknowledges the challenge directly. “Every firm that looks at this asks the same question – how much effort is it going to be for me to move to that?”
REGnosys has invested heavily in translation tooling to address this barrier. One milestone was the release of open-source mappings between CDM and Financial products Markup Language (FpML), enabling firms to convert existing trade messages into the shared domain model.
“If you’re using stock FpML,” Labeis explains, “all of the translations from FpML to CDM are already expressed directly in the model.”
The company’s 2026 roadmap pushes the concept further. The goal is to allow firms to experiment quickly and prove value rapidly.
“Our target is that within a day… you should be able to consume one of your internal messages and translate it into CDM,” Labeis says. The objective is to demonstrate working results “within the space of a few hours.”
In an industry increasingly shaped by rapid development cycles and AI-assisted tooling, lowering the barrier to experimentation could prove decisive.
Making Regulatory Change a Non-Event
The most compelling argument for digital reporting infrastructure is economic. Regulatory reporting remains one of the largest operational cost centres in compliance budgets, often rivalled only by financial crime programmes.
For firms already live on CDM and DRR-based implementations, recent regulatory regime updates have reportedly required minimal operational intervention. Labeis points to the derivatives reporting rewrites introduced in 2024, when jurisdictions such as Australian Securities and Investments Commission (ASIC) and the Monetary Authority of Singapore (MAS) rolled out updated reporting rules aligned with global data harmonisation standards. For firms operating on traditional reporting architectures, these changes typically required new mapping logic, validation updates, and testing cycles across multiple reporting pipelines.
By contrast, Labeis says some firms already using CDM and DRR absorbed the updates with little operational disruption. “We actually didn’t hear from our clients about it,” he says. “It was like flicking a switch – move on with your life.”
The underlying data supplied by the firm remains unchanged, while regulatory interpretation and reporting transformations are captured in the DRR logic layer. “You keep providing the same data – nothing changes,” Labeis explains. When a rule changes, the digital representation of that rule is updated centrally, allowing the reporting output to adapt without requiring firms to redesign their internal data pipelines.
Concluding Thoughts
For many firms, today’s supervisory reporting model remains push-based. Firms assemble and submit periodic reports according to regulator-defined templates.
A more granular, event-driven architecture – where regulators could access standardised lifecycle data directly – remains aspirational. Yet CDM and DRR provide the technical foundations required to support such a model.
Even without a full transition to a pull-based supervisory framework, the impact of shared digital rulebooks is already delivering value. For firms facing continuous rule change – from MiFID rewrites to SFTR reviews – the question is shifting from whether digital models can work in production, to how quickly they can be embedded.
The industry may still be moving gradually, but the destination is increasingly well defined: a reporting ecosystem where shared digital standards absorb the complexity of regulatory change.
Subscribe to our newsletter
