"A Linkedin for Crypto"

We introduce a structure on top of Twitter’s stream of information. In this sense, you can think of this as a Linkedin for Crypto.

Social media is as useful as the accounts you follow and the strength of your network. Each platform forms digital communities similar to their real-world counterparts. For example, Linkedin is like a city. It has clearly marked streets, so you can get around neighbourhoods you have never been to. I.e. explicitly defined roles and relationships between people and institutions. But like a real-world city, LinkedIn is full of posers, scammers, cheezy self-promoters and other kinds of bullshit artists who thrive in groups which lose track of individual’s reputation. People do not know each other, so they judge and are judged based on first impressions and institutions they associate with. It matters more how wealthy you look and where you work than how you conduct yourself.

Twitter, on the other hand, is more like a village. People tend to cluster into small “follower communities“ around certain topics or interest. Like villages, Twitter-follower communities require that you demonstrate your worth to receive a following from the “tribe.” These communities are small so people remember how you behaved yesterday. It pays off to contribute to the collective in such an environment.

Cities are great for business. Villages are not. The reason is that reputation does not scale well. Today crypto community clusters are like villages. They need to become thriving economies with plenty of businesses operating within them if they want to meet their ambitions. They want to be Rome 100 AD, but for now they are small settlements like Rome was in 700 BC. We think that it's not an accident Twitter has become their platform of choice.

For these communities to scale, they need structure. This typically meant abandoning a village for a city. Or Twitter for Linkedin. But what if this structure is derived from village-like stream of information? Perhaps we can have the best of the two worlds.


Our technology relies on a couple of basic assumptions about groups. First, we believe that every group forms a hierarchy. Sometimes these hierarchies are formally defined, like in the case of companies or governments. However, most groups have a fluid, emergent hierarchy, like a circle chatting in a bar or a cluster of accounts on Twitter. For this reason we design algorithms that track how attention flows within these groups on social media and beyond.

We believe that these emergent hierarchies can be mapped by tracking attention flows within the group. Simply, the more attention one receives from other members of the group, the higher his/her position in the hierarchy. However, what matters is not only how many people and how much attention pay to that individual, but also who these people are. Especially in large groups, it matters far more who pays attention. People who themselves receive lots of attention (and occupy high positions in the hierarchy) give one far more by paying attention to them. In order to capture this dynamics, we designed a 2nd order metric.

Finally, it is important to consider that one’s identity exists across multiple contexts. Identity is a series of ideas that people have about us. In different contexts we will have different identities, but they are typically combined into one. That’s how our ‘real’ identity works and that’s how identity works on the internet. For this reason it is essential that we allow for connecting one’s identities across multiple different platforms. In order to do that, we create incentives for owners of those identities to voluntarily link them with each other in our system.

Our initial goal was to build a prototype using Twitter data to develop our algorithm.
Test the prototype
In order to find out whether the results were accurate or now we decided to expose them to a large number of people who are members of the group we mapped. The results were promising.
Index Crypto Twitter
Twitter is the first platform that we analyzed. We started with simple idea: influential people will be followed not only by masses, but also other influential people. We assumed that the act of following is an Indicator of Interest.

Index Podcasts
We index podcasts using their RSS feeds. Every podcast creates an RSS feed that contains the metadata for each episode (title, description, link to the file etc.). What the feed typically does not contain is a clear indication of who appeared on the episode (hosts, guests). We need this information in order to connect each episode to the people that appeared on it.

Index Affiliations
We will index relationships between various identities. Not only a person has an identity, but also companies, brands and products have identities. They are often related to each other. We will map relationships such as:
  • John is a team member of Acme
  • Alice invested in Acme
  • Acme owns the brand Mousetrap
  • Etc.
Index Events
An event is as influential as the people who speak and attend. By matching speakers, attendees and organizers to a given event we can create an influence score for events. Matching events, speakers and organizers is relatively easy - we can replicate the approach we use for Podcasts. I.e. the organizers will be able to manually tag speakers (who will have an option to dispute the tagging).

Index Reddit
Reddit is a different platform from all the above. It is not about influence of individuals, but collectives - subreddits. There are subreddits that are immensely influential and others that wield no influence. In this sense, subreddits are like Twitter accounts. For this reason, our primary objective with indexing Reddit is to quantify influence of subreddits. The individual users come second.

Index GitHub
Index blog posts, news & whitepapers

Want to know when we release new features?

Leave your email address and we will let you know.