Ep #3 - Yulesa & Roberto from Messari - Financial reporting in DeFi

How to build a quarterly report with crypto data, what goes into that and how data is leveraged across the industry to find a basis for your decisions.


Boxer  0:01  

Everyone, welcome to another episode of the Weekly Wizard. I think today is Episode 3. And today, I'm joined by the wonderful data scientists from Messari. That is Roberto to my right and Yulesa just below me. 

And I think, just to start it off, if you could both quickly introduce yourselves and talk a bit about your background — like what you've done before joining the crypto space — that would be very cool.

Yulesa  0:31  

I take that as your list. So, I'm from Brazil, and I’ve lived here since I was born so never brought up. And prior to the privilege of joining Messari, I had a completely non-related tech job. I was an engineer for an oil company here in Brazil. 

Oh yeah, I've made a big shift in my career here. Joining OSI and starting to work with data like that. But for me, it’s been a pleasure because I love the blockchain space a lot. 

Boxer  1:19  

So that’s…

Yulesa  1:21  

How I came…I got a full-time job in the crypto so yeah.

Roberto  1:28  

Yeah, my name is Roberto. I come from a traditional finance background. I was there for four and a half years doing risk management and a little bit of fun development. Went down the rabbit hole around mid-2019 and started exploring crypto in genera as you know…as like an equity. 

You know, I can treat it as like trade equities or bonds or options. And I quickly realized that the theory was a little bit more comprehensive than just an asset, so I started learning about the EVM. 

Others started learning about DeFi, and they quickly realized that I need to leave my cushy job and join Ms. Ra. So, I joined this army full-time in March of 2021 as a researcher focused on the five, and then around, maybe like seven months later, I started heading the data science team. Then I grabbed this wonderful manual to start building some wonderful Dune dashboards for us.

Yulesa  2:26  

I'm Jeremy…sorry, I think it was a toddler last year. So, that was my first job in the crypto space. And I was really lucky to find MySite because Rebecca can talk a little bit more because…was such a coincidence of once I was starting to doing Dune dashboards for myself, and I was looking for someone to explore the on-chain data. 

I didn’t even realize that they had the idea…what it would become at the time, but like yeah…

Boxer  3:12  

Yeah, it's just like…we need somebody who's good at this thing. Let's grab some. Okay. Yeah. Yeah, it's interesting. It’s just the collection of people that end up in the crypto space is always interesting. 

So, what does misery actually do? Why is there a data scientist team and mystery? What's bizarre is Mission, and what are they doing in crypto space for those people who are not familiar?

Roberto  3:43  

Sure. Yeah. So, we're a data aggregator and marketing company with a goal to organize crypto information at a global scale that we can present it to everyone, so they can navigate this new digital economy. 

And the extension of why we have a data science team now is very natural. We need to find ways to extract the data from different data sources particularly blockchains because it's very simple for us to aggregate market and market data. 

You know, when talking about prices and volume, things like that, there's this other huge component in the space that is blockchain and on-chain information, so the data science team was born to tackle that area of let's have a team that can do the on-chain data and also aggregate in such a way that is contextualized or users can ingest it and then make decisions based on that.

Boxer  4:37  

Okay, so before like…I don't know…I think you've been venturing on nonsense data for…but before that, it was pretty much all just market data. 

So, you were having order flow and price and maybe token distributions. That's already kind of unchained but lockup periods and stuff like that. 

More from financial market perspective and not so much the online market perspective…is that kind of correct? Okay, exactly.

Roberto  5:06  

Yeah, that's exactly right. Our focus…I mean, it's bad because our data vendors were mostly on the market site, you know, coin metrics, coin market cap, Tyco. And then now we're venturing more into…let's just get all the data from all. Yes.

Yulesa  5:26  

And our analysts had a really hard time understanding the protocols, workflows, and KPIs like how the tokens move inside the protocol. And now, with the data team, we can provide much more data for them to understand how each protocol works so… 

Okay, the data team is sorry, like it's becoming more important as a resource internally each day, you know.

Boxer  5:59  

Yeah, are you hiring for new Dune visits at the moment? Maybe?

Roberto  6:03  

Yes. Yes, we are. We are. We are hiring at Dune doing research right now. Because, I mean, we can dive more into that. And how our working relationship with you, but I guess we are hiring right now. 100%.

Boxer  6:16  

Okay, we'll make sure to include that link in the description.

Yulesa  6:23  

I've been doing all their hard work on Dune from the Saudi…so yeah, I need help. 

Boxer  6:29  

Your initial backups. Yeah. Are you the only one who is doing the Dune wizard suffered misery at the moment Euro?

Yulesa  6:39  

Right now, I am. But there are another data scientists in misery that also can do it. It’s currently over in other parts of the Euro MSI. So, right now, all they are doing is mine.

Boxer  6:58  

Okay, um, yeah. So that's me. Sorry, covered? Just maybe the last question. What's the business model? Or how does Messari earn money or something like that? Because I think we haven't covered this aspect.

Roberto  7:16  

Yeah, we have different subscription models. We have the free version, which you can gain access to a lot of wealth of data sets. In our website, we have a prescription, which is geared toward retail investors about $30 a month. 

And then we have the enterprise tier, which is more for your hedge fund or your exchange, which is a little bit higher. So, that's how we work our business model right now. 

Yulesa  7:48  

So, we also have a lot of free content that people enjoy like the quarterly report that we are going to talk very much.

Boxer  8:00  

Yeah, I've been very much enjoying the quarterly report. But I think before diving into that, especially you Roberto, coming from a traditional finance background, I think an interesting question to explore is basically how does data availability in DeFi and the crypto space in general compare to the traditional finance world? 

What are stark differences that you could make out directly?

Roberto  8:29  

Yeah, I mean, there's a huge difference between the two purely because financial markets have centuries of datasets, and there are tools everywhere. It's extracting data very quickly. 

Put succinctly, if you're working in asset management or portfolio management, traditional finance, you have a Bloomberg terminal. A Bloomberg terminal has basically all the data that you can ask for any asset that you can imagine. You can get the data instantaneously. That's still one of the biggest things that we're lacking in crypto. 

For example, if I want to get TPL data, I have to clean up my lungs. If I want to get on-chain data, I’ve got a new guy. So, if I want to get fundamental data, I have to go to...so there's no unified area or tool that lets me pull all that information through my analysis. It's still incredibly fragmented. 

Now, I think that's a function of an agency of our industry. I think eventually we're gonna get there. But I think that's the biggest difference that I see. The data availability component, from a research point of view, is a lot easier in trust bias than it is in crypto. 

So, anyone in crypto that has the ability to extract data from either on-chain or through an API has a totally foreign language. Those researchers and those investment funds are going to have a huge advantage purely because it can get the data that a lot of people are not able to get.

Boxer  9:55  

That's actually not quite the answer I expected. You saying the day to access…accessibility is higher. From my perspective, it's like the blockchain data is free. It's publicly accessible for everyone. At least, uncertain platforms. 

And everyone can certainly go in there and make their own analysis, even if it's the super hardcore way of like I spin up my own nodes, a sinker, and then just go against my own node. But as you say, so many people can do that.

Roberto  10:30  

If I do the comparison to a Bloomberg terminal, I can just call a function inside Excel and then get terabytes. So, it is a lot easier for me to extract data from truck five because the database is the structure and the tooling to extract that data. 

And for me to do analysis is a lot easier. When you compare that to our industry, you need to…like no SQL, right for you. 

A lot of researchers, I mean, what we see even eternally…our researchers don't have that skill set in it. Whereas in a traditional world, it's going to be like a download, you know. 

I want these fields to get downloaded…click this link so that components of infrastructure…to make the data easy to get…it is a difference. 

Now, the caveat here is that you say blockchains are open ledger, so anyone that has a skill can get that data if they want to. So, it's more open in that sense. It is more than schooling to extract the data. That is very different between tripod and ad there. 

Yulesa  11:39  

Here in the crypto space, we can go into the project and see how the money flows internally which is the traditional finance. I think this is a little bit hard because the company needs to be transparent and allow you to have this data in here. 

In crypto you don't have to. You don't need permission from him. Unlike here, you can go into AVI or compound and explore all the pools that they have and all the other liquidity they have. You know, this is different from the traditional.

Boxer  12:17  

Yeah, that's also an issue. Yeah. And you can't hide shit in diva. You're used to…I've thought about writing his articles and using on-chain data in the past. But I've stayed away from it for now. 

But there's a lot because I think there's a big problem of analysts. They are mostly incentivized to publish good news because number go up, number is great. And nobody really likes shorting, and it’s looked down upon. 

So it's like publishing negative news. It’s probably not a good way to grow your network, but somebody should do that now. Because that's part of it. If your product doesn't return the numbers it should, then it's not…you can't hide it. It’s somewhere in your quarterly report. 

It’s all live on-chain. You can see people leaving. If there's somebody who's leaving Compound now with 60 million, I'd be immediately able to swap that. And I'd be like, Compound, you're in trouble.

Yulesa  13:22  

You can look at all the data. You can see what the stakeholders are doing. Everything is public.

Boxer  13:31  

Yeah, it's a blessing and a curse as we trained on some recent examples. Yeah, what do you think about the liveness of data. Is this quarterly? Quarterly? Usually companies report in quarterly installments. I guess that is like…yeah, you get what I'm saying.

Roberto  13:59  

I mean, I get it's just data availability from the records. You have a lot of these companies tracked by this big, international investor relations groups within the company. 

And they just report everything on a quarterly basis. So, this is why open blockchains and open data is amazing because you can just track everything on a daily basis and have it in a dashboard for everyone to see. 

I know we're gonna dive into quarterly in a bit. So what a lot of the quarterlies are doing is like actually following that. We're not really going to stick to a quarterly frequency. But we're going to have some sort of frequency to write the reports. But the idea is to have essentially this data available for everyone to see it.

Yulesa  14:45  

That way, with the dashboards that we create and don't, you can have an evergreen data on everything that has happened. So, you can refer to the last quarter report, like look how the product evolved from there. 

So today, you don't need to wait a new quarter to see how things evolved, you know?

Boxer  15:11  

Yeah, do you guys want to dive into compound or usurpers? I'll let you decide.

Roberto  15:19  

US is trying this. What do you want to talk about first?

Yulesa  15:23  

I think it can start on Compound because it was the first one that we've made. And it's the only one that so far has to report already. So, let me share my screen here.

Boxer  15:36  

Did you actually like…usually in a quarterly report, you get a 90-slide PDF? What everything? Which puts that thing into color? Did you guys do something like that as well? 

Or do you just…here's a bit of data and go with it like this usually like explanations in those quarterly reports?

Roberto  15:58  

Correct? Yeah. So, I think we're positioned nicely as a company because we have a massive research ranch in our team. So, it's great for us data scientists to cool that data. But there's still an element of contextualization that data into like actual insights. 

So, our researchers spend a little time understanding more how you compute revenue for this lending platform and escape your revenue for this text. So, the quarterlies essentially give you a contextualized view of the raw data that we're pulling from on-chain. 

This is where we’re very nicely positioned to be the company that is producing this quarterly reports, because we can infuse the raw data with this insights and percent that we use for them to make a rational decision.

Boxer  16:49  

Yep, we need more.

Yulesa  16:52  

I think I need permission to share my screen…oh.

Boxer  16:59  

How the hell would I do that? This…before? Okay, click on the button. Nothing is happening. So permission or not? Let me check for a second.

Roberto  17:16  

Technical issues. 

Yulesa  17:19  

No, I found the problem is that it opened in another window here. 

Roberto  17:25  

There you go. There we go.

Yulesa  17:33  

I think…what's the mission? Oh, it was the first port that we had. It's the compound one. So, this was the first dashboard that we don't have Messari to grab data. Or know how to start it like…

Roberto  18:01  

Let me give you an edge. I'll give you real quick. Yeah, I'll give you an edge so what we're trying to do with this quarterly release is provide a consistent view across subsets of that market. So, all lending platforms are going to have a specific layout. All decks are going to have a lab. 

And if you notice here, we will be showing you that we start providing pretty much a table that contains all the information that we use to create every single chart that we showed you. The one thing that we noticed, when we started doing, this is with the dashboards in general. They have different layouts. 

So, whenever you go into one specific dashboard, you have to basically learn what you're looking at, across every single dashboard. So for us, it was very important for the user to come in and have a very similar experience. 

They can very quickly start ingesting the data and understanding the data that we're presenting without having to understand the organization or the dashboard.

Boxer  19:01  

I see. So yeah, that's a really cool approach. And I'm actually really excited to hear that because I see the same problem in the NFT market. Something where it's like we have a million different NFT dashboards, and they all slightly report different data, and you're like how can I make… 

Yeah, there's a new data point but what does it mean? If you standardize it, then it's like really? Then you can actually work with the data. So yeah, really cool.

Roberto  19:31  

Sorry…you like…now you can take it away, sir.

Boxer  19:34 

Yeah, can you maybe just inside Boxer, right?

Roberto  19:37  

I'm listening to you, but not Boxer okay? You're listening to him, right? Yes.

Boxer  19:46  

You can't hear me anymore. Hello, hello.

Roberto  19:51  

You might be…it might be a stream. The stream is what might be causing the lag because I noticed that once you go live we are getting this signal. Oh geez. Look, you stopped streaming?

Boxer  20:11  

Does he not see? You're either of us? No, it was. Well, he's still moving.

Yulesa  20:19  

Hey, when I clicked to share my screen, I clicked so many times, and it opened many windows only? Yeah, I need you guys to recap a little bit.

Boxer  20:38  

But you can hear both of us?

Yulesa  20:42  

No, it's fine. I've closed all the windows.

Roberto  20:49  

We're just doing a quick overview of the of the compound dashboard. But I don't know, if you want to, you want to talk about…just go through it briefly. Maybe like two or three minutes. Talk about how you came up and the process that you take to when someone comes in, and we have to issue a quarterly…what is your process from research to a dashboard.

Yulesa  21:14  

So yeah, when you first initiate, I try to see all the contracts that the protocol has on donor on-chain. And I started to filter which data is relevant for us and which are not. Most of the time, I'm very concerned about the token flows because most of these DeFi reports that we've done, are about DeFi project. 

So, the oath of tokens decided the biggest concern we have. I started by doing it in pieces. For example, compound here, I started to look only to the deposits that they have. 

I make a query to get all the deposits or the readings or the borrows, and I capture the query to grab this data into blocks of code. And at the end of the query, when I already have all the data that I want, I just stick them together in one big table, which has all the data that we construct from this protocol. 

So, that's how I make this text here. When you see it, it's a big query, but everything is divided by blocks that you can analyze. That's how I do it. Ah, the compound one. 

I started to look at the main performance that they have. For example, deposit redeems borrower repay liquidations and adds them all together to see how much their outstanding deposits are, how much their outstanding loan is, the interest they're paying for the postures, and the utilization there they are having. 

Roberto  23:14  

So, the purpose of having the table at the top is…it was very…it was a decision that we said to me because we want to make this data available for someone to download it very quickly and very easily. 

They can recreate a lot of the charts that we're doing in the bottom. So, if you notice, our dashboards will always have our header, some explanation of the dashboard, and then a big chunk of data table at the top that essentially feeds into all the charts that we create at the bottom. 

So, if anyone needs to download data, mostly or researchers or anyone else in the industry, they can do it very quickly and then recreate those charts that we're creating ourselves.

Boxer  23:55  

That's it. That's a very interesting design choice. That's kind of influenced by how you're working with other teams and the space in more general terms. 

That's an interesting insight. So, maybe we can…I forgot to throw this in at the beginning, but maybe we can take a step back. What is Compound actually?

Yulesa  24:24  

Yeah, so Compound is a lending market that people can deposit their tokens. It's essentially the main function of a bank. You can deposit some of your assets, and this bank, which is Compound, can lend these tokens to other people that will borrow these assets in the US for the US. 

They pay interest for the bank, which is routed to the depositors and the bank take is a share, so essentially, what a bank does is the lending side of it. No. Oh, that's what found this.

Boxer  25:03  

Okay, so how many deposits do they have today?

Yulesa  25:13  

Look here at the outstanding deposit from last year. 

Roberto  25:15  

You have not seen our dashboard, or you would know the answer. Sure.

Boxer  25:19  

I have seen it, but I think if we walk through this like this, it might be very approachable to people. If we do it this way, it’s okay.

Yulesa  25:38  

Right now, people are already depositing 10 billion dollars of value into…

Boxer  25:49  

Okay, do we have a chart for that over time?

Yulesa  25:54  

Yes. All the deposits by tokens…okay, even see, our charters are being represented. The most of the value that…

Boxer  26:09  

Okay, I checked this the other day, and I think USDC and DI were the most popular ones. Do you have an idea of why it's so popular to…what do you think is people's move here? 

Are they depositing stables and then taking out volatile acids in the hopes of shorting the market? Or do they? Do they do something similar to that? Or what's your thinking here? What is the reason that people are using Compound with tables especially? I can see why I would put my Ether there and then I have my…I can take out tables again. It might be against my Ether and do something with those, but yeah, the other way around. I was actually surprised to find that. It was that many tables.

Yulesa  27:06  

Yeah, these tables are the one that pays better. So, people like to receive in yards on it. I think this drives most of the deposits that compound has, and I think they are the most utilized tokens on compound because this allows people to leverage on their position. 

So, they take both coins and invested in other assets, which they can even deposit again if they want to take another bigger loan on debt and increase their exposure to one asset. 

I think, when the market is in a bull market, a stable is going to be the most ROI assets that we have.

Boxer  28:04  

Yeah, yeah. If we look at the chart on the right it’s aggregated loans, right? So, that's the stables take notes on Compound?

Yulesa  28:15  

Yeah, we certainly Compound…also had the make a partnership with Coinbase, which allowed people from Coinbase to deposit tokens into power, which increased their reserves in USDC. I think.

Boxer  28:36  

Yeah. Can you show us this on the chart? I wouldn't be able to tell immediately where…

Yulesa  28:47  

I can see exactly what it is, but they announced the date. So yeah, I know that happened.

Boxer  28:53  

Okay. So, what are people earning on these on these stablecoin deposits at the moment? Do we have a historical, I guess?

Yulesa  29:07  

We have an Archaea on the interest they're receiving as a depositor — like the supply side of the bank. We need to go into individual assets to see it. So because we don't have an aggregate for the whole compound ecosystem…

Boxer  29:45  

So yeah, I guess I'm like, what are the key performance indicators or like if I'm an investor in compound, and I don't want to look at hey, how's compound actually doing? Like which were good shots? Would you recommend me to look at it?

Yulesa  30:06  

I think, right now, the most important metrics they have is how many deposits they have and what is the total value that they are holding there for you to be able to borrow and how much people are borrowing. 

So, how many loans have people taken and also the rate between the two. Like how much utilization is Compound receiving, which we can see here. 

The blue line is everything. The value of the assets that component has…that people have…deposit on Compound…the blue is how much of these assets people have and took as a borrow. 

And we can have a read between the two, which is the utilization rate and tells us how people are utilizing the assets that are available on Compound.

Boxer  31:06  

Action. So, actually, 40% of the assets that are deposited on compound are loaned out right now. Yes, yes. I don't know how that is in traditional finance, but on first, it doesn't sound like that. 

Historically, it has been much higher, though. So, in recent times, there seems to have been…I gotta know…

Yulesa  31:32  

Yeah, we had some events in the recent past that people got liquidated. So, it's the biggest fear of everybody that is lending money on Compound that we can see here. Like, in April, when the markets were taken a little bit. We had a bunch of liquidations on this…you're in right now. And the price of ETH. 

And Bitcoin itself also. We had a bunch of liquidation. I think this affects people’s mindset that you can borrow as everything that they are putting as collateral. They are like…the component of utilization to coming down is sort of flexion of the region market Musa we've had. 

As you can imagine, we're already like…forward where we're tanking…we're literally diving down to really low…the appetite for either. 

Roberto  32:12  

I mean, if you're going to short, you can short. The appetite for borrowing deliberately…positions become a lot less based and just purely because you know, like going down, you're gonna get liquidated. 

So, I think that compared to liquidation, utilization coming down to social components of our market environment. And do you know, what I mentioned, too, is there's another metric that I was looking at for companies in general.

The amount of money that they take from that “for sure” factor from each one of the commitment of the assets that they're borrowing, which is how we're computing the revenue. 

So the audit protocol, say the maximum political side revenue is what we we’re also looking at. Certainly, it’s supply side but also the protocol side money that they're taking in from the lending and borrowing is happening.

Boxer  33:20  

All right, I see. Yeah, I think…

Yulesa  33:23  

I should explain this a little bit for each asset that you borrow far off of the interest that you're paying. It’s routed to the compound to increase their reserves, and they use a phone…something bad happened…or also to pay for protocol revenue. There is a doubt that can govern how this money is going to be utilized. 

And we can see here how much is going to the people there. How part of the interests are going to the people they are depositing which is the supply side revenue that we are making out of interest and how much is going into the protocol itself. 

So, we break this down this chart and then we can see which tokens are paying more revenue which are because not all of them have the same reserve factor so a bit different for the protocol revenue.

Boxer  34:29  

Okay, so if I was on the Compound strategy team, I would now look at this and I would see hey, there's a big gray bar. So, dye is really valuable to us because it gives us a lot of supply side revenue. I guess is it both types of revenue on the right chart? 

But oh, it actually says protocol revenue in the title. So, that's actually really good for Compound the DAO or Compound the project like that's actually money into their bank accounts or wallets?

Yulesa  35:08  

Yeah, but the revenue goes both to the depositors and to the protocol. So, it's just a fraction of the interest that goes to the protocol. 

So, here is a breakdown of the protocol itself for Compound. But the same bigger chunk of it is going to the depositors. Yeah.

Boxer  35:30  

Okay, yeah, yeah, we can see that distribution on the left, but on this righthand side chart, it's only protocol, revenue…is that a daily charge? So, on an average day, they earn like 60-70k just from their depositors. Is that a correct interpretation?

Yulesa  35:51  

It's a daily chart.

Boxer  35:52  

Okay, they're printing money. For sure. That's pretty interesting.

Yulesa  36:02  

But there is a cabinet, which is the graphs below. They are also paying people to deposit there, which is the incentives they are paying. They pay income tokens. So, they incentivize people to deposit and increase the interest. People are betting for deposit tokens. 

And also, they incentivize people to borrow money. So, they pay also the low and side of the equation, incentivizing people to take loan and decreasing the amount that they're paying. They receive income tokens, which we can see here. 

We say the incentives are like… this is our aggregate chart. So, we can see that it's our increasing comp, but here, we multiply the price of console, that's why the sharp graph is a little bit choppy, you know, because it's multiplying by the current price component.

Boxer  37:07  

Yeah, I see, do you have a chart somewhere that subtracts the total? Or the protocol revenue minus the incentives that are paid? Not yet. Okay, that would actually be really interesting. 

I think there was a big story about Spell like two weeks ago or something. That Spell was profitable for the first day in their entire life, basically, because the Spell emissions were less…or maybe a week or a month…I don't know the time…I need more. 

But basically, they had crossed this…they were basically making money for the first time. If you look at all this built-in stereo imaging and look at the revenue that's coming in, it's a hugely important thing to get to that point, right? Because you're talking emissions can't go on forever. 

And once you cross that threshold of being like, hey, we're actually earning money. Of course, we're still giving out not free money, but it's kind of marketing expense. 

You could call it monetary. Yeah, yeah. Customer acquisition cost. There's actually a better. So yeah, crossing that. Yeah, I'd be really interested in seeing that chart.

Yulesa  38:21  

Yeah, we don't have an evergreen chart. Don't infer this. But in the report that Masada makes, they put up a table, which like this data can. It's aggregating quarters, and we can see this in quarters, you know. 

So, here we have how much revenue they're making. Let me see here, that income. So, that's how much I mean. And we can see here how much no interest income. 

And I think we have a line here for how much we are paying as incentives. Oh, we have the adjusted net income, which is the subtraction of both of them.

Boxer  39:03  

Okay. So, they are not net positive at the moment. Because that is the brackets around the numbers actually means a negative number, right? Yeah. I've never understood that. Yeah, it looks better.

Roberto  39:21  

And this is, I mean, what you just showed us…kind of like the component of data…and also the report. And this is where our massive, very intelligent research team comes in and constructs this report to contextualize all the data that you're seeing on the dashboards. 

This is more what you would see in traditional markets. But then, the caveat here is that you always have…you're always going to have the Dune dashboard or the dashboard…ever been desperate to look at these things on a daily basis? Which is astronomically improvement when you compare it to the latest sheet.

Boxer  39:57  

And there's no like…you could theoretically generate this via script. If you build the right tools, you could just generate this from any arbitrary date to any arbitrary date. 

And on command, there wouldn't need to be PVC or whatever. KPMG wouldn't need to come in and go through terabytes of data manually and actually do this, but it's just on-chain. And you can audit that anytime. 

So, yeah, that's really cool. Yeah, what's your guys’ thinking in putting this in this ancient form? Is this to like, basically, I guess like standardizing the data. 

But that's already what you're doing in the dashboard formers. But then, this is to also… I don't know…get the more traditional investors to understand these newer topics in a format they understand or…

Roberto  40:59  

There's a few elements here. One is the written components as a lot of context around the data that you've seen. So, you need to pick some frequency, it doesn't have to be quarterly. It doesn't have to be monthly. It  doesn't have to be annual. 

In our case, what we're doing is doing it quarterly. This is not to replicate what traditional finance has done. But to an extent, you need to have a certain kind of frequency and cadence in order to produce these kinds of reports. 

The other component is like, yeah, you're correct. So, if we're trying to be like, sorry, one of our goals is to just contextualize with this information organized in such a way that anyone can start navigating this new digital economy. 

And when you present it like this to investors, not only in traditional finance, but also in crypto, you can ingest that information a lot more simply wet. And like you mentioned, you nailed that. If  an investor from Trust looks at this, they're going to seem a lot more familiar and a lot easier to digest.

Boxer  42:02  

I see. Yeah, I actually…yeah, no, no, I think about it properly. It actually makes a ton of sense. And, yeah, maybe Dune should work on a feature, which is publishing quarterly reports on our platform. That would be interesting.

Yulesa  42:21  

Legal…leave it to us.

Roberto  42:27  

I don't have those out. Okay, come on, man.

Boxer  42:31  

Maybe we can work together? Sure. No, yeah, very interesting. I'm just another…

Yulesa  42:41  

Just one more thing that I would like to mention here on the Compound. One is that here we have seen the big picture of how Compound is yes, but we can also deep dive on Compound and look at each into each asset that they have. 

So, we can go here in the macro financial statement. We can see the same data but divided by token, you know, so I'll just show it a little bit.

Roberto  43:15  

To you. To clarify these things like the way we take, but the way we approach a quarterly is from a macro view. So, that's protocol — wide view and micro view. It’s either a collateral type when we're talking about a lending platform, or you're talking about a DEX like a specific. 

Because each you can look at the protocol called uniswap as a whole and like how much liquidity there is…and proceed...the other is…and also, you can narrow down the east side. 

And that also contains a ton of information. So, the way we structure our dashboards in this case is to have that macro view where you can see everything that's happening in the protocol. And also you can click on the micro view, take a collateral asset, and then look at all the statistics that are really specific assets.

Boxer  44:03  

Yeah, that makes a ton of sense. And especially, I think there are always moments in DeFi where, certain pools or certain purse on a lot of stress, or there are interesting things happening in them. 

And then it's really useful to have this…. we're just gonna look at this whatever pool or lending pool or liquidity pool, and just really like, hey, what actually happened on-chain in that specific thing. 

And it's really interesting to dig down into the deeper data. So yeah, I love how you guys pulled this up of like, Hey, we're gonna take a look at this whole thing. But then also, we're gonna go down to almost individual transaction level. 

I think the time unit you're using yours like day. But it would be like a task of five minutes to switch just to like…I'm going to look at individual minutes or transactions. So yeah, I really love it.

Yulesa  44:58  

That's why Dune is so powerful for us because the Compound one was the first one that I've made in society. I didn't know exactly how to do it. So, I did it daily. 

But the newer dashboards that we have created, you can select which timeframe you want. So if you can, if you want to look it by day or by hour and select a time interval that you're interested in, you can filter this data in this way. 

So, jumping a little bit to today, uniswap here, you can choose here. They wait for more and see in a better for my blood. Some data is better viewed on a daily basis. 

But other data is better to look at monthly basis, for example, like how many users are interacting with the protocol, or it's better seen on my ABS. Usually, it's better on our basis. Or you can choose here to look at the data the way you want.

Boxer  46:05  

Yeah, yeah, the composability of this is really, really useful. And big shout out to Michael Swiveling which was made this popular in Dune.

Yulesa  46:19  

I would like to make some shout-outs here because Michael is one that I like. I always look at his prairies, and I got this from him. He's such a good visit. So, there are the people that helped me on my way.

Boxer  46:39  

Yeah, but we are all standing on Michael’s shoulders. Yeah. Yeah, so I guess that covers Compound. And one thing between those two protocols. How did this actually come to play? 

Like did compound approach you? And they were like, hey, we'd like to have a detailed financial report. How do these deals come to fruition?

Roberto  47:07  

Yeah, basically. So, a politician comes to us, and they're asking if they will have to have a quarterly report written about them. And then part of the deal is, I guess, we'll read part of the report, but if the data is available, undoing rules are created. 

I think that's right for you to always keep track of it. So normally, that's the ceiling. That compound was the first one. And he like, was there a guinea pig with this coal service? 

And now we're just seeing appetite from more and more and more particles wanting to get this kind of coverage? Purely because we have a really strong research team that can just aggregate all the data and contextualize it? Like one single report that encompasses everything that's happening. Yeah, interesting.

Yulesa  47:55  

I think at the beginning, when we first came out with this idea of the quarters, I think Compound was not aware that they would receive a dashboard with the deal. 

But yeah, I think this is a very…I think your dashboards are even better added for these protocols. I think they want to show this data as evergreen data, you know?

Boxer  48:25  

Yeah, there are merits to both, I would say there's a reason why traditional finance does this and quarterly reports. And then there's also a reason why it's probably better if we do this live. 

And if you just have this because you can't contextualize the data in every segment, you wouldn't need to have analysts online 24/7 for all the protocols. That's not really doable. 

So, if you just say once a quarter, hey, we will look back, and we'll like analyze this and see what has happened, but in the meantime, you can always just look at the data and actually see for yourself, then you maybe need to figure out for yourself what's actually happening in the quarterly. 

And maybe there are other kinds of low-tech solutions, but still, involving some tech where one could solve that. So, for example, a message board or something embedded inside of the Dune dashboard. Just be like, hey, this is the news in the last seven days and Compound, and this might be why the deposits have gone down or something. 

Yeah, there are lots of ways to basically improve upon the quality report with what the current methods like….I love that you guys are doing this, and I think it wouldn't be manageable either. Otherwise, that's right. So yeah, let's dive into uniswap. It's also just amazing what you've built here.

Yulesa  49:55  

Yeah, you thought it was a totally different dashboard from the Compound one because the amount of data on uniswap is massive, you know. They have almost on every block. There is at least one swap. You can swap transactions on a Tyrian. 

So, the dashboard for uniswap was more challenging than the compound was. And here is when I took leverage of the abstract tables that we have for the indexes. And most of this dashboard is based on the abstraction table, which I think is the future on…don't…I think as this protocol evolves the data is going to get ever more complex with time. 

We're going to depend on these abstraction tables that aggregate data from a bunch of contracts or different protocols, and we're going to rely on that more to get the data that we want to make the shots that we want. 

The uniswap is a good example of that. So, it has the same effect structure and the same presentation as the Compound one. So here, we look at swap volumes…like key metrics that are more interesting to us. 

So up, then different from the Compound one. The architecture is the same. We have here a table with all the data that we use. You use to make the charts. They break down here by version, and that's it.

Boxer  52:00  

Yeah, again,  if I'm an investor or potential investor and uniswap, what would you recommend for what kind of metrics I should probably look at? 

And I think with the Dexter traits table that we have because we have all this aggregate data, there's probably also comparisons you can make to other protocols and stuff. So, maybe if you can dive into those aspects of it, that would be cool.

Yulesa  52:28  

First of all, you're putting us in a very hard position because we're the data guys. Anyway, yeah, we can talk a little bit about Usopp. So, I think they have metrics that I think are the most important ones, which is how many users are using uniswap. 

So, it's this first chart here on top and also the volume that they are doing. I think those are the metrics that are the most important ones for you. And the cool thing about uniswap is that they launched a new version of the protocol. 

You can compare the uniswap, v2, and unistrut victory in this dashboard. So, here's divided by each version. It makes comparisons between the two very easy.

Roberto  53:31  

I think you nailed it. Now volume, the last one that I would add is that liquidity for a specific pair. I would also be interested in that kind of not going to metric. So, the depth of liquidity volume, this volume is directly related to the fees that the LPS has generated. 

In the case of uniswap, there's no political revenue. Everything goes directly to the supply side from 3%. You can compute the supplier's revenue directly from the volume. So, volume in the liquidity and also the usage of the platform will be my top three years as an investor in uniswap.

Yulesa  54:14  

Although, I add one more metric here that I like very much. It’s the number of markets that are being created on uniswap. Because as these protocols evolve, uniswap is just a place for you to exchange your tokens when you want to make a trade as a user. 

And it became an infrastructure piece of them of the money Lego that our protocols use. So, we can see how many protocols are developing in Ethereum by how many tools are being created because it's such a basic infrastructure. It's a fixture for these protocols that are this data Latino.

Boxer  55:04  

Yeah, I totally agree. I'm doing sub-B3, and it seems to be much more of a tool than it is. Dennis is a protocol in the end. So yeah, like every liquidity pool, every token pair that distro Australia traded will at some point probably have UNISAW, p3 pool somewhere.  

It just seems very likely, but we will have to see what curve we three…we told us. I don't know. But that's beyond this conversation. Yeah. So with uniswap, everything's going to the supply side, actually. So, there's a revenue charge. Okay.

Yulesa  55:45  

Yeah, so that’s the revenue breakdown by version here. You can see here on the table here that protocol side revenue is zero. So, if they ever make the change in the parameters, here’s how much they charge for the protocol. This could change in the future. This graph would reflect it.

Boxer  56:13  

So what immediately stands out to me is that it seems like the three depositors actually earn a lot more money than the two depositors, which is pretty interesting. So yeah, that's all the things you can learn by just studying these dashboards. 

But yeah, you were saying earlier, I'm putting you in a bad spot. And I really feel that as a data scientist, I'm always pulling all of this data out of the blockchain. And then you're sitting there like, so what does this actually tells me? You're like, I don't like smarter people than me or people with a different skill set. 

And you're more used to actually making decisions based on this data after I interpret it. So, I very much understand the limitations like questions I can ask you.

Yulesa  57:02  

Okay. One thing that my son is doing is when we put together a quarterly report. We sometimes make calls with the thing. 

So, people analyze this data in-depth, what things are evolving? What are the key metrics they are looking into? Like in a more analysis doc, you know?

Roberto  57:30  

Yep, that's why I think you need that element of research involved in this process because we can get you all the data that you need. Someone that ingests that and then just gives you the insight.

Boxer  57:42  

Yeah, yeah, I totally see. So, how many people work on a quarterly report? For example, on Zoom? So report, I guess it was you EULAR on the data side just building this uniswap dashboard? 

And then how many research analysts are there on the other side trying to coach you?

Roberto  58:01  

Well, you know, once we have the template, which is basically what kind of metrics are important for a DEXA uniswap. Then it’s just downloading the data and writing the report. 

In this case, Jerry Stone was the one that wrote the report for uniswap. The first Compound report was written by Ron Watkins, and then the following one is going to be written by someone else, but what I'm trying to say is that we already have a template for what our opponent should be. 

They just have to download the data and then read the report to give insights and contextualize what's happening with the…

Yulesa  58:42  

The Compound one was actually released two hours ago. It was written by Dustin.

Roberto  58:47  

Yeah, there you go. Okay, so yeah, we got Rocio. Nice.

Boxer  58:55  

You have announcements today. Yeah, so I think one last…sorry, go ahead.

Yulesa  59:10  

I just wanted to say that uniswap is even deadly from the data perspective of Dune SWAT was a challenge because, as the amount of data they have, they also are launched in different chains. 

So, this was something that was really nice for us because we're going to use Dune to explore another change. So, here we have Optimism one, and he's gone last year. Yeah. Last year. We're here.

Boxer  59:48  

Yeah, he's kind of coming back. But yeah, that's actually what I wanted to talk about as well. Let's wait for him for a second.

Yeah. Yeah, we heard something. Oh, I'm here, I'm here. Okay. Sorry. Sorry, afterward, you're watching. Yeah, I was just gonna say I saw in the UNICEF report, there was also arbitrary data in there. 

And I guess you guys have some other ways of pulling the data. But what I really look forward to is in the future when we will just be able to have like a drop-down of let's look at arbitrary data. 

Let's look at some data when Dune…like when we finally enable this cross-chain query function. And I think that will really be a game changer because just being able to see how my product behaves on different chains and immediately being able to compare the data in the same format. 

It's in the same way essentially — that will be really exciting. And I'm already looking forward to seeing what you guys can do with that new feature?

Yulesa  1:01:09  

Yeah, anxiously waiting for your guys' motion.

Boxer  1:01:15  

Yeah, blockchains multi-chain carrying, I guess, Solana data. Yeah, it's gonna be a good year for cooperation, I guess. Yeah, for sure, Dune only lives through its users. 

And whether that is individual wizards or whole organizations, in the end, that doesn't matter to us. It doesn't really matter to the market. It's just great that somebody is surfacing the data.

Roberto  1:01:48  

We'll do that…you just give us infrastructure? And then we'll handle that. And are you with them?

Boxer  1:01:56  

Yeah, I think I'll have a look in the chat and see if there are a few questions there. And then I will try to let you go.

Roberto  1:02:09  

Yeah, no problem. No problem. Great. Awful.

Boxer  1:02:11  

Nice to be here. No questions today. If there are quite a few listeners. So, if there are any people in the chat who have questions, please drop them. And otherwise, yeah, what did you do to go?

Roberto  1:02:34  

Drop the mic, and I'm leaving.

Boxer  1:02:38  

Time's up, man, I gotta jump to another meeting. Yeah, it was a great pleasure to talk to you guys. We are really happy. Dune, internally, is really happy about misery building. This is great stuff and also kind of a source of trust because that's what we sometimes see on our platform. Like who is this wizard? 

Can we trust him? Has this data been audited by someone else? And we just know that, if it's coming from the Messari analysts or from EULA, then somebody else probably checks over the data. 

We know that the analysts actually working with the data, so any discrepancies would actually they would be seen, so it's really nice to have a reputable…I don't know the institution, even though we don't like the institutions in the FBI. 

There's probably a way of like…not trusting you guys but trusting the code. Anyone can go into your dashboards and audit the code itself. 

But still, it just usually helps, and yeah, we really love your stuff. Keep on doing what you're doing. Thanks so much for coming on the podcast, and I will send you guys Max in the near future.

Roberto  1:03:57  

It's been a pleasure chatting with you. Yeah.

Boxer  1:03:59  

Nice. Thank you. Have a nice day. Bye, bye Yulesa. Yeah, bye-bye.