Wealth tech is crowded. I hear it every week. Every new logo with a slick deck and a Series B. That’s all fine. That’s not what this post is about.
This post is about the thing nobody on those decks will say out loud: every product in our category is a rendering on top of data somebody else assembled by hand.
We’ve spent the last decade inside the operations of family offices, RIAs, and ultra-high-net-worth households. Not the marketing deck version. The Tuesday-at-2pm version. And what we’ve seen is consistent enough to be a thesis.
The category looks healthy from the outside
By revenue, wealth tech is having a moment. Addepar, by its own disclosures, sits on roughly $9 trillion in platform assets across more than 1,400 firms, and raised a $230M Series G in 2025 at a $3.25B valuation (Wealth Management). SEI’s Archway platform holds $630B in family-office assets (SEI). Plaid says one in three U.S. adults has connected an account through its rails (Plaid). Envestnet, Orion, Black Diamond — all real businesses with real customers.
That’s the demand side. The demand is obviously there.
So why does every operator we talk to still describe their data situation the same way?
It’s mostly working. The custodian feeds come in clean. The alts are a mess. We have an analyst whose entire job is reconciling statements. We’re looking at three different platforms because none of them does the whole picture.
Three platforms, or one with 3+ APIs. To see one family’s money. It has to be easier than this.
Every product is a rendering
Here is the part that took us a long time to be able to say cleanly.
Every wealth-tech product you can name — every dashboard, every portfolio reporting suite, every client portal, every performance-attribution tool — is a presentation layer. A rendering. A view.
Bucket one — is the visualization layer. That’s the chart, the PDF, the iPad app, the client portal.
Bucket two — is the analytics layer. That’s IRRs, attribution, scenario modeling, benchmark comparison.
Bucket three — the one nobody sells as a product — is the structured database underneath. The asset-by-asset, entity-by-entity, tax-lot-by-tax-lot truth of what a family actually owns. Where the K-1 income flows to. Which trust holds which LLC. What the basis is in the Aspen house. What the unfunded capital commitment looks like across eleven private vehicles.
Bucket three doesn’t exist as a product. It exists as labor. And we’ll get into who’s doing that labor in the next post.
The McKinsey “SaaSpocalypse” moment is a symptom
McKinsey recently described what’s happening in wealth-tech equities as a “SaaSpocalypse” — investors finally figuring out that a lot of what’s been priced as durable software is actually a thin layer on top of someone else’s data work (Wealth Management). When the AI agents arrive — and they are arriving — the rendering layer gets commoditized fast. The data layer doesn’t.
I got off a call just this past week with an intelligent investor. Not to invest in us — to get his perspectives on every single one of my blind spots so we could block them. We spent nearly an hour where I shared screen explaining and re-explaining our org chart. The human one, yes. But then separately the AI Agent org chart. Right — we have an entire AI Agent org chart. Sam Junior reports up to Chief of Staff. Peter works in sales. Et cetera.
We are thorough and relentless about human-in-the-loop, and we have a lot of ethical opinions here. But trust me — it’s not arriving. It’s been here for a while for those who know what’s going on, and it’s speeding up.
Rendering is easy. The data is the new currency.
That’s not a knock on the rendering vendors. Many of them are excellent at what they do. It’s a category observation. If your moat is your chart, your moat is two quarters away from being free.
What this series is going to do
Over the next nine posts I’m going to walk through, in order:
(1) Where the structured data actually comes from today — and who’s pulling K-1s at midnight to make it happen. (2) Why dashboards lose clients to other dashboards on a depressingly regular cadence. (3) Why the public-markets aggregators and the alts aggregators have never fully bridged. (4) The actual economics of the family-office back office and where the AI math works — and where it doesn’t. (5) Why “White Glove × AI” is the only combination that holds up under real client data. (6) What it looks like when the data is structured and you can render it anywhere — your CPA, your estate attorney, Claude, your kids. (7) Why this matters more for inheritance than for performance reporting. (8) What the financial operating system of the next five years looks like — payments, taxes, estate execution, all on the same rails. (9) What we’re asking of the industry to get there.
The reframe
Wealth tech isn’t crowded at the layer that matters. It’s crowded at the rendering layer. The data layer underneath is mostly empty, mostly manual, mostly held together by very expensive humans.
That’s the real opportunity. That’s also why most of the category has historically avoided it — because data work, real data work on weird assets, is the kind of work that doesn’t scale by hiring three more engineers and adding a chart.
It scales by combining people who actually understand a partnership agreement with software that actually understands a partnership agreement. We’ll get to that in post six.
For now, here’s the framing I want you to hold: the structured database of your financial life is the product. Everything else is a rendering.