Operations professionals are facing a new "who moved my cheese" moment. The last decade plus has been marked by asset management operations departments embracing a transition from a production line to a more data-centric capability, responsible for the delivery of complete, quality data such as positions, valuations, etc. This evolution has been positive for both employers and operations staff, but a new paradigm shift is emerging, driven by the promise of complex analytics, data science, and cloud-based data architectures.
Today, industry leaders are making sizable investments in cloud-enabled data and analytics platforms and are expanding the target audience for these platforms through citizen developer initiatives. Quants and research departments have traditionally been heavy consumers of these offerings but making Python and predictive analytics accessible to new user groups is front page news for many industries, including financial services.
Unfortunately, the adoption of advanced analytics and non-developer coding platforms remains relatively low in investment operations. Which begs the questions: why does operations remain on the outside looking in? And where are there opportunities for ops professionals to embrace these new capabilities?
Why data science is still a walled garden
Many operations departments are struggling to drive engagement on citizen developer initiatives and analytical techniques. Why? Two primary reasons stick out to me:
Focus on structured data: Operations functions have been, and will continue to be, focused on determining the right number and completing a pre-defined set of data requirements to complete a task. Accounting, trade settlement, and other functions aren't overly concerned with data set correlations, their job is to ensure the number, transaction, or regulatory report is correct and complete. Getting into a data sandbox to experiment or find data correlations can be viewed as a risk, not a benefit, to many operations professionals.
Daily operations use cases don't look compelling compared to others: The world is awash in data and our society is infatuated with predictive analytics. Investors are racing to see what can be done next with data but the highlighted use cases of innovation in our industry rarely even discuss operational data, although they are heavily dependent on it. Macro research trends, economic signals based on satellite imagery of cars parked at malls, or sponsor platform distribution trends sound exciting and get priority. Operational data? We already have a system or a partner for that.
Opportunities to empower operations with data
The reasons listed above, among others, create a disincentive for operations professionals to engage in analytic-driven activities and can be viewed as a distraction in the face of growing volumes of rote BAU support. In reality, operations departments are excellent, perpetual sources of motivated talent. Younger professionals are now heavily exposed to data science concepts in school so finding and promoting paths for them will be critical for operations departments to continue to attract good people. While momentum within investment operations groups has been slow, there are great opportunities out there for firms and staff alike:
Using trend data to identify and resolve common operational errors: The holy grail of modern data analytics is predictive analytics, but operations professionals would gain a lot by setting their sights a bit lower and focusing on building blocks including "descriptive analytics" ("what happened") and "diagnostic analytics" ("why did x happen?"). Most operational knowledge in our industry have been the result of handing tribal knowledge down to others but a digitized world is changing that. Operations managers have the opportunity to wield more powerful, analytics-driven tools to help staff perform at a higher level and support overall efficiency and output despite the practice being relatively new. As a comparison, we've seen analytics-driven services sprout up that help coach and "nudge" portfolio managers on things like security selection or timing of decisions. It's clear that those same opportunities exist to help optimize operation work and create additional efficiency.
Embracing new data governance roles: Today, firms struggle with defining data and creating/updating important meta data to support enterprise data catalogs. My suggestion is that ops pros take the leap to help understand these critical tools and lend their knowledge and subject matter expertise. Operations professionals have embraced data product owner roles, but data governance has remained a "four letter" word. The burden of participating in onerous data governance discussions without a clear connection to action is rough. In the very near future, data discovery is likely to be a foundational task for most employees and operations professionals know how to tap and leverage the data that is exposed to them from internal systems, service providers and counterparties.
Data management has been an evolving space since the advent of digital technologies in financial services, but it seems like we may now be on the cusp of a data renaissance. With the maturity of cloud technology and powerful analytics tools designed for broad business use, there has never been more opportunity for operations talent to sharpen their skills and deliver on the promise of data analytics.