2019/08/19

All In One 2.0: When One Simply Isn’t Enough

by Lucido Group

Executive Summary

Market forces are driving derivatives trading firms to re-evaluate how they use their enterprise trading platform and what role it will have in the future, if any at all. The explosion of innovative fintech solutions, combined with significant improvements in integration means these systems can shift from one that tries to do everything not so well, to one that excels at the basics and integrates for the rest. In this article we explore some of the approaches derivatives trading firms are planning for or already doing to prepare for the future.


Enterprise History in 30 seconds

Enterprise systems grew quickly on the back of market demand. There were many challenges they addressed, most importantly those that were pressing to investors everywhere: market risk management and the use of multiple and disparate systems. ‘One system’ became the mantra – this eased vendor management, ensured risk models and reporting captured the full trade set and it even sounded simple. The truth was more complex and many clients soon found that they would be doing more customizations, vendor co-development and, for that matter, work outside the system than they would have liked. Nonetheless, ‘all in one’ made for a powerful siren song and so went the market.

The Star Loses Some Luster

Technology has a life cycle and enterprise as an innovator, now finds itself on the back side of the mountain. The ‘all-in-one’ model simply does not have the flexibility to handle tough verticals (think complex reporting or collateral management) within its code base. Compounding the challenge of getting much needed functionalities into their system of record, clients are finding that software company owners are maximizing the ‘sticky’ nature of their product to improve EBITDA, not further investment. Although many clients can find this surprising, it really should not be: as key functionalities shift outside enterprise, why invest in new features that are neither market leading nor driving new sales?

Simultaneously, fintech innovation is the hottest party in town. Blockchain settlements, Back Office AI, integrated payments, collateral management, machine learning driven regulatory reporting…the list goes on and, interestingly, it is not necessarily just solving problems that drove the innovations. The innovations are based on technology from outside the sector and create solutions to problems that clients may not have been aware they had. Quite an opportunity! For many enterprise users, these new tools are highly desired but they are blocked by tough choices on the base platform and a lack of a clear and flexible way to access their data.

This makes for what might be called a ‘knotty’ challenge. And there is yet another layer to add: vendor driven upgrades. Reducing their work overhead is part and parcel with managing fewer builds. Those customizations or unused reports that need to be cautiously and strategically removed are now going to be dragged through an upgrade. For many firms, the time, effort and energy needed to execute this project means that the strategic choices and more complex work will have to take a back seat.

Even for the ever-changing technology and markets sectors, this is a lot to have on one’s plate.

As clients considered the best strategic action to push, they were pulled. Executing a minimum viable upgrade, extending functionality for new instrument types and reports, addressing regulatory compliance issues and taking a closer look at those confusing software contracts that were hastily signed over the years. These challenges, in turn, alerted firms to the risks that could have significant negative impacts in the future. Clients are doing their best to ensure that the next integration need or regulatory compliance obligation is not the one that catches them on their heels: it is time for a plan to stay ahead.

Enterprise As A Launching Pad

We start with a mis-step. Understandably, a gut reaction by some clients is to simply replace the vendor. Vendors have, in the past, not only created some great products but employed exceptional sales and customer services staff as well. So that must mean it exists elsewhere, just not my vendor, right? Unfortunately, no. The few clients that switched endured a huge implementation, millions (or tens of millions) of dollars in costs, and substituted one set of incomplete customized code for a different one. The decision fails to address the long-term strategic decline of applications born in the 90s, a design that cannot possibly predict nor address challenges so far into the future and a code base out of step with the myriad technical innovations that have become market standard in the past 20 years.

How can they tap into those without breaking the bank?

Clients see now that adoption of modern application design, accelerated by production-grade open source software, within an agile framework greatly improves business outcomes. Getting there is entirely doable, and without a new implementation. These are some themes and best practices we have seen optimize the process:

  • Agile solutions: because change is accelerating, the system must be ready to adapt accordingly. Adding a new vertical or quickly addressing a market need for a business simply cannot continue to be a long churn and an established, agile structure is part and parcel in making that happen;
  • Automated testing: in order to manage the challenging task of testing an enterprise system, testing need to be built into the process of changes while they are being made, not as an afterthought. The automated tests must offer broad test coverage, and not be brittle to changes, such as UI updates;
  • Standardized components: vendors and service partners need to have solutions that utilize market standards, such as FpML for transaction messaging. Further, key domain expertise is critical in a service partner in order to offer long term viable solutions and tools that will extend the enterprise value proposition;
  • Automation and operational support: standardization of components also must include explicit planning for operational support. Operations can now be automated more than ever before, including software release cycles.

There are many features of the enterprise application that remain more than sufficient: static and reference data management, market data capture, trade repository and auditing and security, to name a few. Retention of the enterprise application also allows an organization to migrate to new solutions in a progressive, controlled manner, all while keeping a central code base that benefits from more than 20 years’ of vendor and client testing.

Further change is coming. It is hard not to imagine a drive to re-baseline maintenance costs given the changes in functional use – not when trade software is being posted to github! But no one is uploading an entire trading backbone. Yet. And the clients with whom we have spoken are not anticipating that level of wholesale change anyway. This is a vendor de-risking process that enables cost savings, greater stability and flexibility to tackle the biggest challenges for at least the next five years.

And for timing, most with whom we have spoken are kicking off the most urgent parts this year and budgeting the next phases for 2020. You should as well.

Latest Posts