Our Investment in Delphia

Today I’m excited to announce that Multicoin Capital has led a $60M Series A in Delphia alongside Ribbit Capital, FTX Ventures, Valor Equity Partners, FJ Labs, Lattice Ventures, Cumberland, Road Capital, and M13.

Delphia is the first instantiation of a new form of capital formation that we have been thinking about for some time that we call “DataDAOs.” DataDAOs aim to harness user-owned data to 1) achieve some greater benefit, and 2) share those benefits with the user who contributed the data.

DataDAOs

One of Multicoin’s core theses is that blockchains provide a novel mechanism for capital formation. We recently highlighted this in our essay on Proof of Physical Work, in which we walked through some examples in which crypto enables coordination for geographically distributed physical infrastructure at unprecedented scale and speed. We believe that large numbers of small data sets can be aggregated to produce emergent value that exceeds the sum of its parts.

Before starting Multicoin, I co-founded a company called ePatientFinder that allowed physicians and clinics to analyze their electronic medical record databases to discover clinical trials or other advanced treatments for their patients. While one individual's health record data is not very useful beyond the scope of the care of that immediate patient, collecting a large data set of many patients’ health records was tremendously helpful to unlock important insights relevant to the entire population, versus just the individual.

However, as I acutely learned, it is extremely difficult to collect a large number of unique data sets because there is a fundamental alignment problem between the contributors and the aggregators. Data contributors are rarely rewarded for their contributions (beyond their own care) because they don’t have a right to the value created by the aggregate data, nor are they entitled to govern how their data is used, which leads to a massive leap of faith in the aggregator—which, unfortunately, has been violated many times over. DataDAOs solve this misalignment by giving data contributors direct economic upside in the aggregate data and the ability to govern it.

Delphia

Delphia is a premier algo-advisor that helps people invest smarter together. Underneath their mobile investment platform, they are building a DataDAO. In addition to unlocking greater financial inclusion with a minimum $25 investment, users (i.e., fund investors) are also invited to contribute their data alongside their money to improve the trading algorithms that manage their money. With each new investor that joins, the algorithm gets smarter.

Our thesis is that users will willingly contribute their commerce, device, search, and social media data in exchange for tokens, which serve as both their interest in the data’s aggregate value and their governing power. This data can then be used by hedge funds and other investors to forecast the key performance indicators of public and private companies, which, in turn, can generate investment returns (alpha) attributable to the data. Here are some simple examples of forecasting and attribution that Delphia facilitates:

1. A large spike in search queries for a specific product can predict that the revenue of the company that makes that product will increase
2. Significant movements of employees between directly competitive companies can predict a dispersion in corporate performance
3. A material change in foot or car traffic at a business can be predictive of revenues for both that company and its competitors
4. Influencers promoting a product or service on social media can predict an uptick in sales

Following its attribution algorithm, Delphia shares revenue generated by the data with users contributing to the dataset. This creates powerful alignment that has never existed before.

There has been a long running joke in tech that Google (and more recently, Amazon and Shopify) could operate the world’s most profitable hedge funds. We think Delphia can actually fulfill that vision, and do so in a way that shares the upside with the data contributors.

Historically, retail investors pay a robo-advisor a percentage of AUM (e.g., 0.25% / year) to manage their money using relatively undifferentiated strategies. Delphia disrupts this model and allows retail investors to invest in differentiated strategies with an active return layered on top—for zero fees. “Pay to play” is replaced with “contribute data to earn” (as described in the next paragraph). All of the contributed data is then used to improve the managed portfolios in which the user has invested.

Delphia’s app (live in the App Store) gives users the ability to 1) create an investment account and 2) opt-in to contribute their own data — which will soon include Amazon purchase history, Google search history, LinkedIn data, Twitter data, credit card data, GPS data, etc. Soon, Delphia will also be launching a chat-like interface in the app for asking users questions about themselves and the world around them. This data is then transformed by Delphia into features for consumption in machine learning models.

By collecting data that is 1) not commercially available and that 2) uses a unified identifier across data sources, Delphia can build a dramatically richer data set for predicting asset prices and consumer trends than what exists today.

To be clear, Delphia does not give any consumers of the data access to personally identifiable information. Rather, the system anonymizes the data set and will only allow access to the data via a secure training environment.

Returns to Scale

Delphia is the classic example of a network-effect based business model: as more users contribute data, the returns for funds using the data should theoretically improve. This creates a flywheel, in which more users are incentivized to provide data because hedge funds and other businesses (who benefit from crowdsourced data at scale) are willing to pay more for access to the aggregate data.

While one individual, or even 1,000 individuals, contributing data would not generate incremental returns for data purchasers, hundreds of thousands or millions of users contributing data (and reaping the returns for providing that data) provides a material edge.

Curation Games

In addition to providing users the opportunity to contribute data and receive tokens, Delphia and data purchasers can use subTokens to incentivize subDAOs/communities around a specific data connection (like Twitter or Robinhood data), or even around a creator and their audience. Investors can then value the subTokens of the various communities based on how valuable they view that specific data set to be.

As an example, Alice can speculate that Twitter sentiment data will become increasingly important for capital markets forecasting over the next few years, while shorting Facebook data (e.g. she thinks that Gen Z users primarily signal their preferences on Twitter, and they are slowly becoming a larger part of the economy over time). The curation games will be settled based on data set purchasers’ attribution analysis in identifying and signaling which data sets were most valuable for them.

These curation games provide a powerful mechanism for the Delphia DataDAO to gather more data which further strengthens Delphia’s edge.

The Rise of DataDAOs

Trusting a company with your data is a scary proposition for some. However, trusting your data with a DataDAO presents an entirely new trust proposition. When a company like Google or Facebook monetizes their users’ data, not only are the users not compensated, but the users also don’t have any say in how their data is used. These companies often have misaligned interests between their users and their stockholders/executives.

With a DataDAO, the data contributors can use their tokens to vote and govern how their data is used. Gone are the days of relying on a corporate entity like Google promising not to be evil, instead users can take control over their data into their own hands.

We believe that data is the most valuable commodity of the modern era. Users deserve to be compensated for the value they create with their data. Users deserve governance over how their data is used. DataDAOs will help make this vision a reality.

...

Disclaimer: This content is provided for informational purposes only, and should not be relied upon as legal, business, investment, or tax advice. Potential DataDAO participants and Delphia investors should review the written materials related to such DataDAO and Delphia fund, if any. Multicoin is not involved in the operation of (and does not act as an investment adviser to) any DataDAO or any Delphia fund. This blog post does not constitute a recommendation in connection with any DataDAO or Delphia fund.