An Experience Talk: How Can Performance Disrupt the Finance Technologies?

The financial sector is being embraced by the so called digital transformation [1] and it seems that the discussions wouldn’t disappear easily in the following years. Though the finance companies are actually one of the forerunners of the industry, they are rather slow in adapting new IT technologies. Among other things, slow law regulators, mandatory global standards and customer portfolios are conservative and skeptical about state-of-the-art technologies and business models, thus impeding this transformation. Hence, a lot of finance institutions are still using their legacy systems and processes. In contrast, FinTech (“Finance Technologies”) startups disrupt the markets: They are new players in the neighborhood, who are trying to change how people and businesses do their finances by introducing alternative methods and services based on newer ideas and technologies (e.g. blockchains or know-your-customer). They are small, innovative, fast as well as agile and try to tackle risky obstacles of this transformation era by approaching early adopting customers with their solutions to disrupt finance industry unlike the bigger, older players [2,3].
We investigated several quality attributes of a potential disruptive technology in a recent project between FZI and a long-lasting, mid-size enterprise in Germany that is one of the major players and a provider of IT solutions for finance and retail industry. The main project goal was the thorough check of the bandwidth of possible realization scenarios of a new real-time omnichannel payment product directly between debtor and creditor with a response time just under several seconds. It requires responsible institutions to clear and settle transactions not in batches but individually and finalize them immediately, similar to messaging platforms available 24/7/365. This new technology is called “Instant Payments” [4,5,6] and is supposed to be the successor of “Target2” platform [7].
Based on this major performance aspect, our tasks in the project covered evaluation of the possible scenarios, definition of a generic payment process model, investigation of alternative architectures for valid scenarios as well as their simulations for verifying possible bottlenecks/limitations with help of component-based architecture modeling and simulations in Palladio toolchain [8]. We eventually focused on one question from a software engineering perspective: “Which approach within the predefined design space is the most suitable one to meet the performance and scalability requirements? (a) Adjusting/extending the existing payment transaction system, (b) developing the system completely from scratch, or (c) a symbiotic solution based on the existing payment process?”.
Corresponding to the constant information flow between the consortium partners, our efforts included gaining certainty in architectural design and forecast for non-functional requirements. The first clear outcome was that it (a) would be infeasible to exploit the existing payment system to meet the requirements. The process was slow and the corresponding data model was heavy containing information not necessarily needed for the new product. For management reasons, the definition of a completely new payment process and its development from scratch (b) wasn’t favored. Although, based on the results of our models and simulations, the requirements would have been met easily. Though the final decision is still on a halt, we were able show that a symbiotic approach (c), i.e. a lightweight variant of the existing payment data model within a specifically defined process model running on several market-ready infrastructure alternatives, would meet the defined requirements.
During our work, we eventually thought about the role of performance in FinTech and how such a payment method advertising mainly its performance and availability can be a disruptor. In the end, the service being introduced isn’t necessarily a completely new, innovative idea, but rather a better implementation of an already existing concept. But wouldn't it be enough to change something in the finance industry? Although, for a concrete answer further factors need to be considered, we think, that a digital real-time payment wouldn’t necessarily change how we pay, but would definitely change how banks and institutions handle our payments, which may in the end lead to other disrupting technologies, i.e. allowing them to use their resources in further areas and thus creating new opportunities. In other words, performance as a quality attribute can surely be used as criteria to build better products of existing concepts or improve design and development approaches, which eventually would disrupt the industry.