 | |||||||||||
| |||||||||||
| | |||||||||||
 Happy Monday, Fintech Takers! In 1966, Warren Buffet bought 5% of Disney for $4 million. A big part of his investment thesis rested on the enduring value of great IP. As he recalled decades later:
This idea has always made conceptual sense to me, but it never became tangibly real until recently, when my daughter discovered the movie Frozen. Wow, let me tell you something: That movie resonates with her on a frequency that I didn’t even know existed. It’s basically the absolute perfect version of product-market fit. And, in keeping with Walt Disney’s business plan, it has led to the purchase of a great deal of Frozen-themed merchandise, which will likely include a Halloween costume later this year. Incredibly, Buffet sold that 5% stake in 1967 for $5 million, proving that even the GOAT misses occasionally. - Alex Was this email forwarded to you? Sponsored by Persona KYC is a solved problem. KYB somehow still isn't.  A garden pavilion in a forested landscape by Jules Lachaise and Eugene Pierre Gourdet. 3 FINTECH NEWS STORIES#1: A Bigger Walled GardenWhat happened?The Clearing House (TCH), a payments network owned by the largest U.S. banks, is planning to launch a tokenized deposit network:
So what?I’m going to go ahead and disagree with the Wall Street Journal’s framing, right off the bat. This is not “one of the most significant efforts yet to open up the crypto world to the banking industry.” First, as Ron Shevlin points out, the details on the project don’t give the impression that the big banks are all that serious. It doesn’t have a name (“the bridge,” “the chain,”). They haven’t selected a vendor, even though it’s expected to launch in the first half of 2027. And, as the Wall Street Journal reports:
This brings me to my second point: Bank-owned consortiums exist to solve problems for banks, not their customers. Think of B2B money movement — which is what TCH is aiming at with this initiative — as a dial. On one end is control: the money stays inside the bank, on rails the bank owns, where the bank can see it, hold it, and lend against it. On the other end is reach: the money can go anywhere, to anyone, instantly. The catch is that the dial only turns one way at a time. You cannot add reach without subtracting control, because for money to actually leave one bank for another, the deposit has to leave too, and preventing deposit flight is the primary objective that the big banks are trying to achieve when it comes to anything crypto or stablecoin-related. If reach isn’t a priority for customers, the big banks already offer plenty of compelling options. For domestic payments, TCH already offers an instant, 24x7 payments network (RTP) that large corporations can use to move large amounts of money (TCH recently raised the transaction limit to $10M, which led to a significant increase in payment activity). If programmability is important (and it does seem like corporate treasury teams are intrigued by the potential of programmable money flows), the big banks have been building their own private tokenization services to allow customers to move money in different ways within their own walls. If reach is a priority — if a corporate treasury team is struggling to get money to the long-tail of suppliers and overseas counterparties, in places like São Paulo or Jakarta, quickly, predictably, and at a low cost — we already know what the best answer is: a stablecoin. Open, interoperable, and accessible to anyone with a wallet. This is obvious. So much so, in fact, that the big banks explored the idea of jointly issuing a stablecoin, through TCH or their other consortium EWS, last year. That initiative went nowhere, and instead the banks landed on the idea of this tokenized deposit network. And that decision tells you what problem the big banks are really trying to solve. A tokenized deposit network operated by The Clearing House doesn’t fully address multinational corporations’ money movement needs. The counterparties that these corporations need to pay don’t all bank with JPMorgan Chase, Citi, Bank of America, and Wells Fargo. The money needs to move outside those banks’ walls, which is the one thing those banks are determined to prevent. Tokenized deposits are a way to maximize interoperability while ensuring zero deposit flight. That might be appealing to bank customers in a world without stablecoins, but, unfortunately, that’s not the world banks are competing in. #2: There are levels of survival we are prepared to accept.What happened?Experian is expanding its integration with ChatGPT:
So what?You know that scene in The Matrix Reloaded where the Architect is explaining to Neo that his failure to comply will result in the extinction of the entire human race? Neo says, “you won’t let it happen. You can’t. You need human beings to survive.” And the architect coldly replies, “there are levels of survival we are prepared to accept.” That’s the first thing I thought about when I read this Experian press release. Today, Experian makes a lot of money from its direct-to-consumer business, in which it helps consumers monitor their credit and identity data, and shop for, select, and apply for loans and insurance products. Experian’s B2C business has more than 200 million free members globally and it contributes between 25% and 30% of the company’s overall revenue (split between premium subscriptions and referral fees). It was also Experian’s fastest growing business line in the first half of fiscal year 2026 and an area that Wall Street is watching very closely. What Experian and its competitors on the B2C side (Credit Karma, NerdWallet, LendingTree, etc.) worry about is tomorrow. According to a LendingTree survey, nearly half of AI chatbot users say AI has influenced a financial decision, and 30% specifically around opening or closing an account or loan. When OpenAI announced its new personal finance capabilities, it stated that more than 200 million people globally come to ChatGPT every month for budgeting, questions about their investments, and other financial planning queries. How long will it be until AI chatbots are the primary user interface through which consumers monitor their finances and shop for new financial products? Probably not that long! And in that future, lead aggregators and affiliates like Experian, Credit Karma, NerdWallet, and LendingTree will have to find a way to survive. Based on the actions of those companies so far, there appears to be two potential paths forward:
The concern with both of these approaches is that they depend, entirely, on the decisions that the AI chatbot chooses to make. When a user asks ChatGPT for a good rewards credit card, how should OpenAI decide what answer to give? Should it find the most authoritative source on the internet and trust the answers that source provides? If so, NerdWallet and other GEO-focused affiliates will be in a strong position. Should it partner with providers that can answer the user’s question and agentically assist them in opening the account? If so, Experian and Intuit, with their embedded marketplaces, have the advantage. Intuit, in particular, seems to be in a favored position here. The company was name checked in OpenAI’s personal finance announcement, in the section talking about agentic capabilities. Or should it do what’s actually in the user’s best interests? Ignore all the answers supplied by ad-driven affiliates and marketplaces and, instead, conduct actual product research and make an unbiased recommendation to the user, informed by the bank account data that they have linked into ChatGPT Pro? OpenAI could decide to do any of these. Or a mix of them (OpenAI could, for example, build its own competitive ad network for lenders and insurers and cut out the affiliates entirely). The point is that there’s no way for Experian, Intuit, NerdWallet, or LendingTree to control that decision. Giving OpenAI $100M might earn you some goodwill, but it doesn’t guarantee anything long-term. And given that uncertainty, I think Experian is the company that is best-positioned to make this bet. Even if its B2C business was to entirely evaporate thanks to the disruption caused by AI chatbots, it would still have its core credit bureau business — the credit data, attribute libraries, scoring models, and decisioning tools — to fall back on. I’m sure it would prefer that not happen, but in the absence of humans, there are levels of survival they would be prepared to accept. #3: A Ramp to Future GrowthWhat happened?Ramp raised more money:
So what?This latest fundraise puts the total amount of money that Ramp has raised since it was founded in 2019 at north of $3 billion. During that time, its valuation has climbed in a (mostly) upward direction, from $1.6 billion in 2021 to $8.1 billion in 2022 and then down to $5.8 billion in 2023. In 2025, it raised four different rounds, jumping from a valuation of $13 billion in March to $32 billion in November. And now, after this latest round, it is valued at $44 billion, and I’ll be honest with you: I don’t really know what to make of it. I tweeted about my confusion last week, when this news broke, and the tweet kinda blew up, so I thought it was worth fleshing out a bit more in the newsletter. As I stated on Twitter, Ramp is a great company. There’s basically no one who disagrees with this statement. The product is best-in-class. The people at the company are ridiculously smart and relentless in the speed and quality of their execution. They seem to have a genuinely strong handle on AI and the vibes surrounding the company are immaculate. The company’s total payment volume rose 170% year-over-year in March, which was its highest growth rate in three years, despite the business being roughly 20x the size. And Ramp’s annual revenue run rate is now more than $1.5 billion, roughly double what it was in March of last year. So, what do those numbers mean, in the context of the company’s new $44 billion valuation? Well, at $1.5 billion in revenue, that valuation represents a 29.3x multiple, which is, in the year of our lord 2026, quite good. It’s in the same ballpark as Stripe’s and higher than Revolut’s, which are the only two private fintech companies operating in Ramp’s stratosphere. And when you compare it to the market caps of public companies, the picture gets more confusing. The market caps of PayPal, Fifth Third Bank, and Block — all companies that generate substantially more revenue than Ramp — are all below $44 billion. Affirm is perhaps my favorite comp because the intangibles — the raw horsepower, quality of execution, AI fluency, and vibes — feel the most similar. Affirm is on track to generate more than $4 billion in revenue this year (a 33% increase year-over-year) and currently has a market cap of $22B. If that market cap were to hold and Affirm hits its projected revenue, the revenue multiple would be 5.5x, which is … [double checks the math] … less than one fifth of Ramp’s revenue multiple. What accounts for that discrepancy? Based on my Twitter replies, there are two factors. The first, obviously, is the difference between the private market and the public market. The public market is brutal. It operates on realism, not optimism (and public market bank investors operate on pessimism, as I wrote about here). By contrast, the private market is, to put it in the nicest possible terms, a bit untethered from reality. It measures value over longer time horizons. And, to the benefit of Ramp, it has become intensely competitive within a very narrow band of companies that fit investors’ appetite for a grand slam winner. The second factor — and it’s related to the unique dynamics of the private market right now — is growth. I was told over and over again that the reason Ramp is valued as highly as it is is because it’s growing incredibly fast; 170% year-over-year, the company's highest growth rate in three years. This growth argument is fascinating to me because, while I hear it all the time, I’m not always sure what it means. Why do private market investors, particularly those that get into companies at later stages (like this Series F for Ramp) value growth so highly? What does it signify? In the case of Ramp, the answer could be boring: Ramp offers a great product for business spend management and the market for that product is enormous. Or the answer could be cynical: Ramp is great at storytelling and $3 billion can buy you a lot of good vibes if you know what you’re doing (I wasn’t a huge fan of the company’s Super Bowl ads this year, but that’s just me). Or the answer could just be dumb luck: Ramp’s TAM expanded at exactly the right time and in the right direction thanks to the explosive growth in AI spending. It’s impossible to say from the outside. Sponsored by Oscilar Getting an AML agent to produce a plausible disposition isn't the hard part. Everything that comes after is: SOPs, analyst review, model risk validation, QC, examiner scrutiny. Agents handle the volume. Analysts handle the judgment. 2 READING RECOMMENDATIONS#1: SoFiUSD Is the Bait, Galileo Is the Hook (by Ron Shevlin, The Fintech Snark Tank) 📚Ron’s take on this topic is better than mine and I’m jealous of it. (I can’t tell you how many times I’ve said that over the years.) #2: Are Cards and Accounts the Same Thing? (by Matthew Goldman, CardsFTW) 📚I enjoyed Matthew’s discursion on the differences between cards and accounts. Fun BBQ conversation topic, if you’re looking for one this summer! *Bonus: Your Customers’ AI Agents Are Coming for Your Data. Are You Ready? (by me, with MX) 📚22% of consumers already use AI to manage their finances. You don't have to build anything for AI to change the relationship you have with your customers; it's already happened. Read my new article on what to do about it. 1 QUESTION FROM THE FINTECH TAKES NETWORKThere are a TON of interesting questions being asked in the Fintech Takes Network. I’ll share one question, sourced from the Network, each week. However, if you’d like to join the conversation, please apply to join the Fintech Takes Network. If I wanted to learn more about the history of the Federal Reserve and why it’s structured the way that it is and operates the way it does (because I’m a giant nerd), what would you recommend I read or listen to? Are there any people I should speak with?If you have any thoughts on this question, reply to this email or DM me in the Fintech Takes Network! Thanks for the read! Let me know what you thought by replying back to this email. — Alex | |||||||||||
|
{beacon}
Workweek Newsletter
