Joshua Frank | Building the Data Layer for Crypto | The Curious Learners Ep. 31

My guest today is Joshua Frank. He is the co-founder and CEO of The Tie, an information services provider for digital assets. Josh and I met some time ago at a crypto conference and I was truly amazed with The Tie product demo back then.

On-chain data is usually accessible, however it is never easy to parse it given the different data structures across chains. In addition, converting data to signals and market intelligence for investors' use requires a lot more processing and a deep understanding of the market.

Josh and I discussed a lot on this episode starting from the extent of Blockchain data to building a product-focused business:

What does Blockchain data include? What does matter?

It is huge. It includes on-chain data, sentiment, news, pricing, NFTs, company funding rounds, SEC filings, regulatory press releases...

How do you convert raw Blockchain data to signal?

Most important thing in finding signal is having top quality and granular raw data such as news from primary sources (e.g. SEC filings, project updates, medium posts, etc.).  In addition, point-in-time data is critical as posts get deleted, links are changed, etc.

How to think about alpha decay in crypto? (alpha decay defined as the gradual loss of power of a certain strategy or information edge over time)

There is no such thing as alpha decay in crypto yet. Those who can find creative ways of exploring new sources of high-quality data will continue to find alpha. Things like cohort analysis for users of projects or movement of developers across projects or chains are some examples. 

What is the difference between traditional finance and crypto in terms of access to and interpretation of data?

Lack of tested fundamentals such as valuation metrics is one of them. And the biggest difference is that crypto has more data because every single transaction is broadcasted to blockchain. It allows us to understand Blockchain projects at a deeper level with something like user-level data, which is not possible to access in traditional markets.

Users vs. wallet addresses?

If, hypothetically in a particular blockchain, more than 90% of the wallets have less than a dollar, are they really users? Don't think so.

How do you build sentiment analysis? Of all the ingredients that go in such as price feed, Twitter posts, TikTok videos, etc., how do you even weight these different things to come up with a sentiment score?

There is of course a proprietary sentiment model. It is very difficult to use a traditional sentiment model and adapt it to crypto. It wouldn't work. There is a lot of intricacies specific to crypto. And it is important to look at crypto assets relative to one another in addition to thinking about in terms of sub-sectors such as DeFi, GameFi, NFTs, etc.

Going for institutional vs. retail platforms?

The Tie is focused on workflow for institutional users. The wide range and depth of features on the platform are best suited for sophisticated institutional users. In addition, those two user groups are very different from each other and it is not possible to serve both of them well.

What is the plan for the short term? And the long-term vision for The Tie?

Short-term: continue to penetrate into the workflow of The Tie's clients. To be the platform that a client uses from the time they wake up to the time they go to sleep.

Long-term:  We want people to think about The Tie when they think about data in crypto, similar to how they think about S&P, Bloomberg, etc. for data in traditional markets.

Curious about?

Understanding all the different businesses and all the different ways. Also doesn't hurt to watch the New