Financial Data is so Darn Lucrative

The Initial Data Offering (IDO) community is a place for data enthusiasts to discover new datasets daily.

The mission is to build a community of data enthusiasts and curate high-quality, unique datasets for businesses, researchers, and organizations worldwide.

The Terminalist is out with his latest blog post and dives deeper into the data & information services world. Continuing to talk about Bloomberg, Factset, LSEG, and the other big players.

It’s a heavy post with 20+ pages which is great if you like to nerd out on the details of the data industry like me. For those that want the TLDR, here are some things that stood out to me:

“Like shipping, a pure distributor rarely owns its data, except when they control production or activation assets (like Bloomberg with fixed income data or LSEG through Refinitiv acquisitions). Without ownership rights - third-party data isn't exclusive and even proprietary reference data isn't unique - pricing power is limited to the value of the productive work.” Something I have said for a long time is that without proprietary or exclusive data, the pricing power and differentiation are limited.

“The threat of data substitutes limits it further. Reference data, being a public good, is available from all major distributors. As automation costs plummet and LLM capabilities advance, smaller players can shortcut the painstaking process of extracting unstructured data and organizing it into usable information. Sure, no large customer is going to change vendors overnight. Trust takes years to develop. But we are approaching a future where trusted reference data can be available not just from a few big vendors, but also from niche vendors coming together through data marketplaces.” This is the big one. The big unlock. Instead of 10s of thousands of people offshore crunching data, the belief is that the costs are going to plummet given LLM capabilities. Small players are going to be able to produce as good or possibly better datasets when it comes to coverage, accuracy, speed, and capabilities with 1000x lower costs. Trust will take time. Switching will take time. But if you build it they will come. Every financial player and I mean every player, needs redundancy. Everyone buys multiple feeds of data especially if it is core to your organization running well. From pricing and volume data to fundamentals data to corporate actions. The list of the must-haves is longer than you would expect. Competition has always been in this industry but it is changing rapidly.

“Another artful form of triple dipping - getting paid for the creation, maintenance, and licensing of the ratings. A highly margin accretive practice.” Triple Dipping- this might be my new favorite phrase. The triple dipper. A few examples are given in the index world, but also the ratings agencies. Taking data and getting paid for it 3x. It’s genius. It’s also why there is so much opportunity. I think of data where you charge clients to consume it on a UI/UX or terminal, you charge for the API feed, you charge to download it, and you charge extra if the people consuming it via a feed do anything with it outside of using it in a “black box” i.e. quant strategies. This is more than triple-dipping, but let’s figure out another sexy term another day.

Lastly, this table is great for anyone trying to understand why the data industry is so interesting. These returns are ridiculous. These businesses at scale print money. They take a LONG time to build, but if done right, they are well worth the gray hairs.