WealthTech's Bottleneck Isn't the Technology

Every WealthTech pitch deck I've read makes the same implicit argument: better algorithms, better personalisation, better AI-driven portfolio construction will unlock a bigger share of a market that's currently underserved by slow, expensive incumbents. I think that argument is wrong about where the actual bottleneck sits, and the eight hours I spent inside Bloomberg's actual professional workflow made the reason obvious in a way no fintech pitch ever has.

Bloomberg Market Concepts isn't a course about clever algorithms. It's a walkthrough of the live Terminal (the same interface equities, commodities, and fixed income professionals actually use), and what struck me wasn't the sophistication of any single analytic. It was how much of the Terminal's value has nothing to do with prediction at all. It's a shared reference layer. Everyone in the industry already trusts the same numbers, watches the same economic indicators update in the same place, and defers to the same platform as the default. That's not an algorithmic moat. That's a trust-and-distribution moat, built over decades, and it's a completely different thing to compete against than “our model is smarter.”

A software business would lose this fight on features

If wealth management were purely a software problem, a well-funded challenger with better UX and a sharper model could out-execute Bloomberg and the private banks within a few product cycles. That's what most WealthTech founders seem to believe, implicitly, given how much of the pitch is about the model. But wealth management isn't purely a software problem: it's a relationship business wrapped around a software layer, and the relationship is the part doing the actual commercial work. Clients with meaningful assets aren't choosing an adviser or a platform because its portfolio optimisation is 4% more efficient. They're choosing based on trust built over years, institutional credibility, and the fact that the downside of being wrong with someone else's money is personal and reputational in a way no amount of clever modelling offsets.

You don't out-algorithm a trust relationship. You have to out-earn it, and that takes distribution and time, not a better model.

I got a similar signal from the commercial side, not the technical side, at a JP Morgan networking session on Asia market strategy. What came through wasn't a room full of people debating whose model was sharpest. It was a room where the operative currency was relationship depth and distribution reach into markets where trust with local institutions has to be built, not deployed via API. The firms winning aren't winning because their quant desk is smarter than everyone else's quant desk: most of them are drawing from a similar talent pool and similar public research. They're winning because they've spent decades building the relationships that make clients willing to hand over decision-making authority on their wealth in the first place.

This is where I think most WealthTech strategy gets the sequencing backwards. The founders build the smartest possible product first and assume distribution and trust will follow product quality: that if the tool is good enough, adoption is a marketing problem to solve later. The firms that actually win in this category tend to do the opposite: they secure the trust relationship or the distribution channel first, often through partnership with an incumbent institution rather than head-on competition with one, and layer the technology in underneath a relationship that already has credibility. The technology is necessary. It's rarely what's actually rate-limiting growth.

You can see the correction happening in real time in how the sharper challenger apps have started behaving. The ones that launched purely on product (cleaner UX, fractional shares, a slicker onboarding flow than the incumbent app) have mostly ended up doing the thing their original pitch implied they'd disrupt: partnering with, or getting acquired by, the institutions that already hold the trust relationship, because organic distribution against decades of institutional credibility turned out to be far more expensive to build than the product itself ever was. That's not a failure of those companies. It's evidence for exactly the point I'm making: the product got them attention, but the trust-and-distribution layer is what determined whether that attention converted into a durable business, and building that layer from scratch took most of them longer and cost them more equity than building the software ever did.

None of this is an argument against building better WealthTech products: better models and better UX are still worth building, and they compound once you have distribution. It's an argument against the specific claim that the bottleneck is technological sophistication, when the evidence, both from watching how the industry's actual reference tools earned their position and from watching how institutional capital actually gets allocated commercially, points somewhere else. The bottleneck is trust at scale, and trust at scale is a distribution problem with a long time horizon, not a modelling problem with a fast one. Any WealthTech strategy that doesn't start by answering the trust-and-distribution question first is optimising the part of the business that was never actually the constraint.