Banks aren’t as stupid as enterprise AI and fintech entrepreneurs think

Announcements like Selina Finance’s $53 million raise and another $64.7 million raise the next day for a different banking startup spark enterprise artificial intelligence and fintech evangelists to rejoin the debate over how banks are stupid and need help or competition.

The complaint is banks are seemingly too slow to adopt fintech’s bright ideas. They don’t seem to grasp where the industry is headed. Some technologists, tired of marketing their wares to banks, have instead decided to go ahead and launch their own challenger banks.

But old-school financiers aren’t dumb. Most know the “buy versus build” choice in fintech is a false choice. The right question is almost never whether to buy software or build it internally. Instead, banks have often worked to walk the difficult but smarter path right down the middle — and that’s accelerating.

Two reasons why banks are smarter

That’s not to say banks haven’t made horrendous mistakes. Critics complain about banks spending billions trying to be software companies, creating huge IT businesses with huge redundancies in cost and longevity challenges, and investing into ineffectual innovation and “intrapreneurial” endeavors. But overall, banks know their business way better than the entrepreneurial markets that seek to influence them.

First, banks have something most technologists don’t have enough of: Banks have domain expertise. Technologists tend to discount the exchange value of domain knowledge. And that’s a mistake. So much abstract technology, without critical discussion, deep product management alignment and crisp, clear and business-usefulness, makes too much technology abstract from the material value it seeks to create.

Second, banks are not reluctant to buy because they don’t value enterprise artificial intelligence and other fintech. They’re reluctant because they value it too much. They know enterprise AI gives a competitive edge, so why should they get it from the same platform everyone else is attached to, drawing from the same data lake?

Competitiveness, differentiation, alpha, risk transparency and operational productivity will be defined by how highly productive, high-performance cognitive tools are deployed at scale in the incredibly near future. The combination of NLP, ML, AI and cloud will accelerate competitive ideation in order of magnitude. The question is, how do you own the key elements of competitiveness? It’s a tough question for many enterprises to answer.

If they get it right, banks can obtain the true value of their domain expertise and develop a differentiated edge where they don’t just float along with every other bank on someone’s platform. They can define the future of their industry and keep the value. AI is a force multiplier for business knowledge and creativity. If you don’t know your business well, you’re wasting your money. Same goes for the entrepreneur. If you can’t make your portfolio absolutely business relevant, you end up being a consulting business pretending to be a product innovator.

Who’s afraid of who?

So are banks at best cautious, and at worst afraid? They don’t want to invest in the next big thing only to have it flop. They can’t distinguish what’s real from hype in the fintech space. And that’s understandable. After all, they have spent a fortune on AI. Or have they?

It seems they have spent a fortune on stuff called AI — internal projects with not a snowball’s chance in hell to scale to the volume and concurrency demands of the firm. Or they have become enmeshed in huge consulting projects staggering toward some lofty objective that everyone knows deep down is not possible.

This perceived trepidation may or may not be good for banking, but it certainly has helped foster the new industry of the challenger bank.

Challenger banks are widely accepted to have come around because traditional banks are too stuck in the past to adopt their new ideas. Investors too easily agree. In recent weeks, American challenger banks Chime unveiled a credit card, U.S.-based Point launched and German challenger bank Vivid launched with the help of Solarisbank, a fintech company.

What’s going on behind the curtain

Traditional banks are spending resources on hiring data scientists too — sometimes in numbers that dwarf the challenger bankers. Legacy bankers want to listen to their data scientists on questions and challenges rather than pay more for an external fintech vendor to answer or solve them.

This arguably is the smart play. Traditional bankers are asking themselves why should they pay for fintech services that they can’t 100% own, or how can they buy the right bits, and retain the parts that amount to a competitive edge? They don’t want that competitive edge floating around in a data lake somewhere.

From banks’ perspective, it’s better to “fintech” internally or else there’s no competitive advantage; the business case is always compelling. The problem is a bank is not designed to stimulate creativity in design. JPMC’s COIN project is a rare and fantastically successful project. Though, this is an example of a super alignment between creative fintech and the bank being able to articulate a clear, crisp business problem — a Product Requirements Document for want of a better term. Most internal development is playing games with open source, with the shine of the alchemy wearing off as budgets are looked at hard in respect to return on investment.

A lot of people are going to talk about setting new standards in the coming years as banks onboard these services and buy new companies. Ultimately, fintech firms and banks are going to join together and make the new standard as new options in banking proliferate.

Don’t incur too much technical debt

So, there’s a danger to spending too much time learning how to do it yourself and missing the boat as everyone else moves ahead.

Engineers will tell you that untutored management can fail to steer a consistent course. The result is an accumulation of technical debt as development-level requirements keep zigzagging. Laying too much pressure on your data scientists and engineers can also lead to technical debt piling up faster. A bug or an inefficiency is left in place. New features are built as workarounds.

This is one reason why in-house-built software has a reputation for not scaling. The same problem shows up in consultant-developed software. Old problems in the system hide underneath new ones and the cracks begin to show in the new applications built on top of low-quality code.

So how to fix this? What’s the right model?

It’s a bit of a dull answer, but success comes from humility. It needs an understanding that big problems are solved with creative teams, each understanding what they bring, each being respected as equals and managed in a completely clear articulation on what needs to be solved and what success looks like.

Throw in some Stalinist project management and your probability of success goes up an order of magnitude. So, the successes of the future will see banks having fewer but way more trusted fintech partners that jointly value the intellectual property they are creating. They’ll have to respect that neither can succeed without the other. It’s a tough code to crack. But without it, banks are in trouble, and so are the entrepreneurs that seek to work with them.

Responses are currently closed, but you can trackback from your own site.

Comments are closed.