The financial technology stack has evolved into an increasingly modular ecosystem. Fintech companies today often assemble their products using a patchwork of specialized services: one provider for payments, another for identity verification, another for banking-as-a-service, and so on. This modularization of fintech has clear benefits. It allows firms to pick best-of-breed solutions and build faster by “plugging in” ready-made services via APIs. In fact, large enterprises value this modularity for critical operations – if payments can’t be processed, fraud can’t be stopped, or accounts can’t be reconciled, those failures become existential risks. Using multiple providers gives flexibility to minimize such risks through redundancy (avoiding single points of failure), broader coverage, and cost competition. However, this unbundled approach also introduced a new challenge: fragmentation – of data, processes, and accountability – leading to operational complexity that can offset those gains.
Anyone who has integrated several fintech services knows the pain of stitching together disparate systems. Customer data ends up scattered across multiple platforms. Without deliberate integration efforts, critical data remains fragmented in silos rather than flowing freely between your onboarding service, payment processor, ledger, and compliance tools. One fintech infrastructure company describes the situation well: “every API speaks its own dialect, forcing developers to spend more time fixing integrations than innovating”. In other words, the more modular your stack, the more you may find yourself acting as the “orchestra conductor” just to make these pieces work in concert.
The result of this fragmentation is operational inefficiency. Firms resort to building internal glue code, manual reconciliation processes, and data pipelines to connect the dots. This is not just hypothetical. We’ve seen it first-hand in our startup journey at Rexi. The productivity we gained by leveraging third-party fintech APIs early on eventually faced friction – data mismatches between systems, duplicate entries, and growing overhead to ensure everything stayed in sync. Our experience is echoed across the industry. Regulators have observed that while fintech partnerships and outsourcing can improve flexibility, the rise of fintech also “raises complexity” and operational risk in banking because the sector is becoming more modular. In short, a modular fintech stack can turn into a double-edged sword: it accelerates product development, but without proper coordination it can devolve into a fragmented maze.
This maze isn’t just an IT inconvenience – it can lead to serious failures. A cautionary example arose recently in the fintech banking-as-a-service arena. Synapse Financial Technologies, a middleware platform connecting fintech apps to sponsor banks, collapsed in 2023. In its aftermath, partner banks like Evolve Bank & Trust discovered they lacked direct access to Synapse’s ledger and had no independent copy of transaction data. Without a shared source of truth, the banks struggled to reconcile accounts, and some customers’ funds were effectively “lost, stolen, or misplaced”. This ugly incident underscores the risks of data fragmentation: when each module in the chain maintains its own records without robust alignment, a break in one link can propagate failure across the whole system. It’s a stark reminder that operational fragmentation isn’t just inefficient – it’s dangerous.
Why is fintech integration so fraught? A core issue is the absence of common standards for data and interfaces. Traditional financial networks benefitted from decades of standardization. By contrast, the fintech API explosion happened quickly and organically. Each provider defines its own data models, file formats, and API endpoints. There is no equivalent of a “common language” that all these services speak. The Bank for International Settlements (BIS) has flagged this as a systemic challenge: ”the lack of harmonised standards and interoperability in key enabling technologies like APIs is a major hurdle, leading to inefficiencies and a fragmented digital ecosystem”. In practical terms, when there’s no commonly accepted API standard, every integration is a custom project – and that slows everything down and introduces errors.
An analogy can be drawn to the early days of computing or telecom before standards emerged – lots of innovation, but nothing “talked” to each other out of the box. In fintech, even initiatives that aim to open up systems have suffered from this problem. Europe’s open banking movement, for example, saw slow adoption in part because each bank initially built its own API with no common standard, forcing fintech developers to adapt to dozens of unique formats. A recent industry review noted that this proliferation of one-off APIs meant smaller banks incurred prohibitive costs to develop and maintain connections, hindering the broader ecosystem from taking off. Simply put, without standards, every new connection reinvents the wheel.
Data fragmentation is the natural byproduct of this environment. If one service labels a field “last_name” and another uses “surname,” a developer has to map between them. If one system reports a payment status as “settled” and another as “complete,” someone has to decide if those are the same. Multiply these small frictions across dozens of services and thousands of data fields, and you get a fragile patchwork of translations. It’s easy for things to slip through the cracks. As one technology provider described, when every API follows a different structure, integration failures pile up and digital finance “grinds to a halt”. The lack of common data definitions not only wastes developer time, but also poses a security and reliability risk if inconsistencies lead to mistakes in compliance or transaction processing.
Interestingly, traditional finance has long recognized the need for coordination and common standards – often enforced through regulation or industry consortia. Consider the global payment networks. Banks worldwide communicate over the SWIFT network using standardized message formats (the ISO 15022/ISO 20022 standards). These messaging standards act like a common language. In fact, a key goal of the ongoing migration to ISO 20022 is to replace the “fragmented standards” of legacy payment messages with a unified data model. Today, cross-border payments suffer from different local protocols and formats, which makes straight-through processing and automation difficult. Regulators and industry groups responded by pushing ISO 20022 so that all participants adhere to one standard, eliminating those inefficient format conversions. ISO 20022’s promise is greater interoperability and richer, consistent data, which should reduce errors and costs in global transactionscitibank.com.
We’ve seen successful examples of standardization dramatically improve financial plumbing. A frequently cited victory is the creation of the Single Euro Payments Area (SEPA) in Europe. Prior to SEPA, each European country had its own domestic payment scheme with unique rules. Over 30 different national systems resulted in a fragmented payments landscape that was costly and slow for cross-border transfers. Regulatory action via SEPA unified these into “one common payment scheme,” turning that fragmented landscape into a cohesive zone for euro payments. The result was that a German consumer can send money to a French business as easily as a domestic transfer, with common standards ensuring compatibility across banks. SEPA shows that when regulators mandate coordination, it can level the playing field and significantly improve customer outcomes.
Another example, from the industry side, is TriOptima in the derivatives market. In complex markets like OTC derivatives, each institution originally maintained its own records of trades, which often led to discrepancies and bloated portfolios of offsetting contracts. TriOptima introduced a centralized post-trade service (triReduce) where multiple banks could regularly reconcile and compress their trades together. This multilateral coordination allowed redundant trades to be torn up, drastically reducing outstanding notional exposures and operational risk. Today, TriOptima’s compression service is used by hundreds of firms and is recognized for “lowering costs and mitigating risk in OTC derivatives markets”. It’s a powerful example of how standardized processes and cooperative infrastructure can solve fragmentation—banks effectively agreed to a common method and platform for managing their data, to everyone’s benefit.
These cases underscore a key insight: financial systems achieve resilience and efficiency when there are shared standards or utilities acting as the backbone. The fintech sector, in its youthful boom of innovation, hasn’t fully caught up to this principle yet. In traditional banking, coordination is often imposed (through protocols like SWIFT or regulatory frameworks), whereas fintech has been more like the Wild West, with each company moving fast on its own terms. The consequences are now evident in the fragmentation we discussed. The question is, how will the fintech ecosystem evolve to address this? Will it learn to coordinate via common standards, or will it seek other mechanisms to orchestrate the chaos?
Broadly, two approaches are emerging as solutions to fintech’s fragmentation problem: build orchestration layers or push for standardization. They are not mutually exclusive, but they represent different philosophies on how to achieve a more unified, efficient fintech stack.
This is the market-driven path. If the issue is that there are too many disparate systems, the solution is to add a metasystem on top that aggregates and coordinates all the moving parts. In practice, this means new platforms that fintechs or banks can use as a central hub, through which all their other integrations are managed. For example, in the payments domain, the rise of “payments orchestration” providers shows the appeal. Companies with global customers often end up using half a dozen payment processors, gateways, and bank connections to get broad coverage. Instead of directly juggling all those integrations, they can now use a payments orchestration platform that provides a single unifying layer, abstracting away the individual connections. Yuno, a startup in Latin America, is one such orchestrator: it pitches that you “only need one global payments orchestration provider” instead of integrating each provider’s tech, simplifying operations dramatically. The demand for these solutions is growing as large companies realize they can outsource much of the integration burden – as Yuno’s CEO noted, many enterprises are now actively seeking orchestration platforms after learning what they are.
The orchestration approach isn’t limited to payments. We see it in data infrastructure and other areas too. Alloy, for instance, offers a “data orchestration” layer for banks and fintechs in risk and identity: connecting to many data sources (for KYC, fraud checks, etc.) and funneling the results into one workflow. They describe the value in almost musical terms: with orchestration acting as the conductor, all data sources can finally play in harmony, yielding a synchronized and consistent workflow. In practical terms, a good orchestration layer can handle the heavy lifting of translating and normalizing data between systems, managing API connections, and even smart routing (deciding which provider to use when). We even see incumbents embracing this concept; JPMorgan recently launched a “containerised data” service via its Fusion platform – essentially a common semantic layer that models and normalizes data from diverse providers to give investors consistent information. That’s orchestration by another name.