The 3 Us of alternative on-chain data

Transcript

[Music] good okay so hello everybody great to be in Paris this year so today let's talk a little bit about

from the on-chain data perspective what we can derive in terms of how all those protocols and products are being used how the tokens unlock and at the end of the day what the unit economics of those projects are so first obviously a quick disclaimer but also I'd like to acknowledge the the work of the team that wouldn't and that was supposed to be a little French flag Emoji um but that we'll leave it for now so about us so I'm Sasha director of BD and strategy at the tie we are in an Institutional service provider where we provide information services so think data we collect together tons of very different data sets on social media to

news to Market data to on-chain data and as we'll see a lot of usage data so in this presentation I'd like to focus really on the usage data that we have on all those D5 protocols as well as a very unique data set that I'll spend most of the time actually on which is our token analog data set so those are the what I call the three use of alternative on-chain data unit economics how do protocols in a sense derive value usage how are people using those products and unlocks finally which is a very important piece of the economics of

those project which is you receive tokens from this project how do they vest and unlock over time and can we track this does that have an impact on the market so let's start with the unit economics so most of the D5 space is very early stage projects so as such you can think about them in very much in VC terms as it's a startup that startups needs to invest in order to capture clients and customers that's the cost of customer acquisition and from that they will over the lifetime of their interactions with those customers they will get some value out of it which is called the lifetime value and when and

that's a very open question by the way on where you set the boundaries uh and without getting into Securities laws and and what you define as Enterprise that can also be thought about in the context of building public goods because at the end of the day you're going to invest in getting users and you're going to get some kind of value capture associated with that value creation that you provide to your users so you you need to think about what that lifetime value to cost of custom requisition ratio is and obviously you needless to say that if it's one already you're losing money because you have all

your expenses and and and cogs Etc so it really needs to be above one ideally three and five between three and five so that you can have something that's sustainable but how can you derive an idea of what your unit economics are from the on-chain data well we'll get a bit later into the usage part but that's a bit different right what is the cost of customer acquisition so one very neat example is let's say you have a layer one blockchain that does a series of nft drops to engage people more and get new users on their blockchain what we can do is we can go and go through the all the events and look what the

average impact on the number of users on that blockchain is and if you look here the zero is the day of the airdrop and you can see pre and post what the average impact on the daily users on the blockchain is and so if you want to know you know you know how many millions of dollars you spend to do that nft job and do that collaboration with that F1 team and now you can look at on average how much did the drop increase the usage the number of use of users that you add every day you multiply this by the number of days it lasts for but and you and you can have an idea of how much you paid for each user that you got on chain the other thing of the the other

component of is the lifetime value for this you need to understand how much value you cruise from having that customer on board so here we're looking at Uni swap V2 and V3 aggregate fees on ethereum with a seven day moving average and you know this is pretty easy to do but still it's surprising how desperate those calculations are across the crypto space and so here what you can see is right after the FTX collapse a lot of people rush to unit Swap and so you see a sharp increase in fees um and you know another hint at looking at at this more from a valuation perspective is What multiples are those project trading at and for this you can

look at the market cap divided by the amount of fees that are being brought in by the by those customers and so if you look at here is when if you look back in 2021 you have very high multiples and if you get closer to nowadays you're much closer to the 10x SAS multiple Revenue so let's jump into usage because that that part is actually pretty interesting and you'll see that there is a lot more things that you can see into the online data than one often could think so things that you want to answer is well you have users use your own product you need to know how they're using it when are they using it what are they using it for right who are they using it with what other products they might be using

right why do they like your product more and not just from the perspective of well you know my product is differentiated from X Y and Z products because I built it this way but really how does that translate in the way people actually use your product and what's really neat and nice with on-chain data is well now you can have a lot of insights into your competitors products as well so this is where you'll see here a little thing on this chart it's basically a retention analysis on the uniswap users so you start here and so for all the users that start on this month here you can see at the end of month one

how many users were still active month two month three month four Etc so what you see is a year later only four or five percent of the user were still active and you can do this and by the way on a lot of nft projects and drops and things like that what you see is actually even a couple months later that number is actually zero percent so how are you retaining your users that's a key question when you build a product another interesting thing is you can really go down deeply into how users use their products so this is the median duration for which people you can drive from this the median direction for which people provide liquidity on unit swap so basically you're an LP provider on on

UDA swap how long do you provide that liquidity for before you withdraw it and that's really important you know if you're unit swap to really understand that very clearly but also let's say you want to build a product that's tangential to uniswap or you're a fund or how do you understand really how other people behave and maybe rethink the way the ways you use this products another thing you might want to look at is how wealthy are your users right you know how much you're getting from your users but you know what's the average distribution maybe one day we'll see projects and protocol that would actually have some kind of fees that depend on some kind of measure of user wealth

because if you want to have public goods or you know things that that are useful to people sometimes it makes sense to have people who can't afford it to pay a little bit more and people who cannot afford it to be subsidized to some extent what's where it gets really interesting is because you have all the cross connects between this what smart contractors addresses use you can look at the overlap of users between different projects and protocols so here what we're looking at is the overlap in and bear in mind these are log log plot so if you look this in a linear scale sushi would be completely out of the chart

um and what that this allows you to do is to basically see who are your comparables right what are projects that you have most similarity with and obviously this makes a lot of sense I have quick swap Ave crypto.com very trading focused projects so if we zoom in on sushi now if you're a unit swap how does your user retention right having sticky users is very important how does that compare to Sushi well long story short uni soap is doing better and what are your users actually using your product turns out it's actually pretty similar and you can see a big start of the activity around 9am

Paris time then the overlap between European and U.S hours and big spike of activity a little bit after 9am New York City time um you can also look at the number of users you have over time and you see that you know the V3 release of unit swap coincided with a huge increase in activity so there was a supposedly a great launch you can even go further and do some uh categorizations of the wallets right and you want to do user segmentation so how are who or who is using your your product and when are they using it um and I'll skip a little bit to the unlock Section because there's quite a

few interesting things about it but you can also look at you know when are people depositing on unit swap versus withdrawing maybe as a way of having an idea of the what the market health is um so let's move on to I think the the most interesting part of this talk unlocks there's very little transparency on unlocks of how the tokens are unlocked their each project each chain have a different process for that they have a different um vesting ways of doing investing some of them you'll be they'll say that they took invest monthly but if you actually look at the Smart contract they invest

every 30 days so if you look way into the future well that starts really piling up there are so many other things sometimes also a project will say in the smart is in the white paper the unlock will start at that date and then in in the execution actually the smart contract was pushed out either a little bit too early or a little bit too late right so these are discrepancies that can be quite important so we went and literally put analysts full time going through every white paper or going through every piece of public information that they could find even when they were dapped reaching out to the teams to the foundations to try and understand

um what where the gaps were what was the most accurate representation of those unlocks um and then you have to do some approximations as well because in a lot of cases you can have definitions in terms of number of blocks so we do a daily approximation of when the unlock is going to occur and all this study is that I'm going to show you is based on that huge data sets or of 350 000 plus uh unlock events and these go back all the way to 2016 to um to very uh to to to the as far into the future maybe 21 200 or like whenever those tokens end

and they're unlocked but for this we'll focus on 21 June 2021 to June of this year um so what does it look in terms of price so you normalize the price around the price at the time of the unlock so one is going to be the price at the time of the unlock and you look at trajectories of that price uh around so it's really hard to see here so you'll you'll see we'll we'll maybe uh have uh the need to go into some statistics about this but this is what it looks like now if you look at the volumes you need to plot this in log because you have huge spikes they're very volatile and you have huge packs of activity for some of those projects well actually how do those volume uh how are they distributed and what you see here is

actually it looks like a parallel so if I'm do the exact same chart of the distribution of the of the volumes in terms of daily traded volume this is very you know a very common way of measuring the act the the liquidity of an asset you see that they're actually power laws right so when you see in the lockplot it's it really behaves like a straight line and so for this study you know the another big challenge has been getting good accurate prices because especially for tokens that are very thinly traded having a robust price is actually challenging so here for example we'll just cut it off at 100 times daily volume which is already huge

so this is where you get into the different types and sizes of unlocks so what happens when you have tiny little unlocks small unlocks that happen maybe sometimes every day well you get something like this on average so the price decreases ahead of the unlock and after the unlock rebounds but not after a month not quite exactly to the original price so you'd think when you do a lot more when you have a much bigger unlock that would impact the price a lot more and it does but what's actually surprising is you get that sort of very complex structure and it might not be

statistically significant but this looks like something like this a narrative around this could be you get ahead of the unlock the price that decreases then people see that the price didn't that you know there wasn't so much cell pressure or something and so the price rebounds but then uh token holders actually really start getting their tokens and maybe start putting them on the market and so it goes down and and then they're they're done selling and goes back up again that could be one potential narrative um and so if you look by different size so overall this is the the green line this is what happens on average for small tokens right you see sort of an over correction around the unlock and

then it recovers and for the large tokens you um you get this sort of complex shape and in terms of volumes obviously you know small tokens you get a small impact on the trading volume and large tokens larger impacts medians behave completely differently so we don't have to get time to get too much into this for another time but medians behave completely differently then another very interesting fact is depending on the type of unlock that you have you're going to have a different behavior and so here we classify them by allocation investment emissions liquidity mining we have a few other types but here the allocation type is

going to be Foundation unlocks developers advisors communities and do you see that those actually impact the Price Less and they also impact the volumes less whereas all the rewards staking Community rewards those impact the price a lot more and also the volumes so it's interesting right you you might think this as foundations advisors are more long-term actors long-term oriented whereas all the small rewards they people just dump them because they're more um short-term oriented so that's it um basically just to give you an idea of what can be done with on-chain data to

really profoundly understand how users use these protocols obviously for Builders to make better products how value is not only created but also captured and what it looks like and you know should you do an F1 sponsorship or not because how much at the end of the day are you getting in terms of uh new activity and lifetime value and that brings unpresented ways for people to learn about not only their product but their competitors product in comparable products and on the token unlocks more to come and this is a very preliminary study but you you can see that there is different impact on the market based on the type of unlock and most importantly of course

and it's quite intuitive the size so I leave it there um and one other mentioned on unlocks there's a huge need for very for a lot more transparency and disclosures on how those vesting unlocks happen when where and having a muscular way of getting that information thank you very much and I'll open for questions [Applause] [Music]

[Music] thank you thank you