Is Skunk Works for Banks and Brokerages Affordable?

Skunk Works for Banks and Brokerages is not only viable, it’s advisable!

As part of my executive responsibilities for a bank’s capital markets and retail brokerage, I was able to build a Skunk Works team for digital products and services. It was comprised of registered reps, traders, business analysts, digital marketers, and technicians.

Skunk Works, for those unfamiliar with it, was the brainchild of Lockheed Martin engineer Kelly Johnson. The name was inspired by a mysterious locale from the comic strip Li’L Abner.  Conceived in 1943, Johnson used this process to build America’s first jet fighter (the P-80 Shooting Star) in less than 150 days, in response to the emergence of German jets in Europe.

What allowed Johnson to operate the Skunk Works so effectively and efficiently was his unconventional organizational approach. He broke the rules, challenging the current bureaucratic system that stifled innovation and hindered progress. His philosophy is spelled out in his “14 rules and practices.”

At this point I should emphasize that the only rules we broke under our Skunk Works for bank and brokerage services initiative were the rules of waterfall development and limited channel marketing. As a financial institution there is little flexibility with regulations—as it should be considering the financial crash of 2008.

While it lasted, my Skunk Works budget was a little north of $2 million. It was targeted toward innovative products, services, technology, and digital marketing—and over 50% of it used in purchasing infrastructure. Not a lot of money if compared to larger bank’s new IT initiative budgets, but quite substantial for a regional bank’s investment division.

Idle Investment

While made-up of a representative cross-functional team for the division, none of the team worked full-time on the Skunk Works projects. All team members remained responsible for delivering the day-to-day tasks associated with their respective positions. This meant a lot of wasted value from the capital spent to build the infrastructure while it sat idle at least 60% of the time.

Still, from this group we produced the first retail fixed income integration to a retail online brokerage, a securities auction platform, a wholesale foreign exchange platform for institutional clients, and the first financial video channel on YouTube. We dissolved the group as part of the bank’s consolidation program, which began in 2014—just as AWS cloud services were starting to mature.

Skunk works for Banks and Brokerages has actually hit the big time, as seen in a 2018 Fortune article titled “How JPMorgan Chase Learned to Love the Blockchain.”

It discussed the bank’s creation of their own Skunk Works in 2014 called New Products Development. Don’t be dismayed by the boring title (maybe intentionally named to fool corporate spies at the beginning) the title doesn’t match the work—or the people in it!

Consider a job listing in February 2015 that appeared on the JPMorgan Chase career website.  “You care about disruption,” the description read, in part. “You have an opinion on Bitcoin and other cryptocurrencies, and you are probably ambivalent about the prospect of working for a large financial institution.” Not your standard job description for a bank’s technology department.

No information is available regarding the budget for JPMorgan’s Skunk Works, but with $11 billion committed to technology in 2019, it is likely to be considerably more than mine was.

Cloud technology allows for zero capital investment in Skunk Works for banks and brokerages.

What I wish I had during the height of our efforts was what is available with AWS cloud adoption today. Take two of the areas that JPMorgan Chase is exploring with their own Skunk Works—block chain and Artificial Intelligence.

Testing these technologies on legacy systems would be cost prohibitive for most banks and brokerages. The capital requirement in servers and infrastructure would be tremendous and would be impossible to justify with the standard bank measurement of Net Present Value.

Cloud Designed for Proof of Concept

However, one of the many benefits with cloud is that you only pay for what you use—no upfront costs required! And in the case of blockchain and AI/Machine Learning, there is a significant skillset required just to build the basic platform that cloud companies provide when you subscribe to their service—a huge savings!

Additionally, AWS not only provides block chain and AI as part of their many services, they also provide free training and a limited availability to their Solutions Architects when an enterprise begins working with them.

Now, using AWS (Azure and Google provide similar services, but with less maturity) you can form a team consisting of general ledger accountants, business analysts, some technology specialists, and begin testing the value of blockchain to your financial institution. No upfront capital costs and no cost when it sits idle.

Pre-cloud you were limited to wishful thinking upon reading that JPMorgan’s “initial results from machine learning fraud applications are expected to drive approximately $150 million of annual benefits and countless efficiencies.” Or that “machine learning is helping to deliver a better customer experience while also prioritizing safety at the point of sale, where fraud losses have been reduced significantly, with automated decisions on transactions made in milliseconds.”

With cloud, you simply recruit some risk and compliance personnel to join with some fraud business analysts and info security experts and begin testing on AWS’ AI provided services.

I’m an ardent supporter of competing with FinTech innovation through cloud services, but there are also many areas of standard bank processes that can significantly benefit with cloud adoption. Creating better general ledgers and fraud detection are certainly critical to every financial institution and are just two examples of core services that have the potential to be exponentially advanced with the help of cloud capabilities.