Who will win the DEX market? A Tokenlon panel: With Dragonfly Capital, 0x, 1inch, Curve and Kyber

We recorded this panel for our Tokenlon 5.0 event. With: Haseeb Qureshi (Dragonfly Capital), Amir Bandeali (0x), Sergej (1inch), Michael (Curve), Shane (Kyber), Lucas Huang (Tokenlon).

Watch the panel on Youtube or read the full transcript below:

Intros

Hello everybody, I am Haseeb at Dragonfly Capital and today I’m bringing you an amazing DeFi panel with many of the technical leaders from some of the most important projects in DeFi today. So we’re going to go through and have quick introductions. Starting with Amir. Please introduce yourself.

Thanks for having me today. I am Co-CEO of 0x Labs. We build decentralized exchange technology, across a variety of different layers. The most important piece being 0x protocol, which is an aggregation and liquidity protocol more tailored towards professionals.

Next up we have Michael.

Hello, my name is Michael. I am the founder and CEO of Curve Finance. Curve is an automatic Market Maker for stable coins and single-priced assets. At least for now.

Next we have Sergeji.

Hi everyone my name is Sergeij. I’m the Co-founder of 1inch exchange. 1inch exchange is here to save time and money. So it’s an DEX aggregator. With additional protocols like the next generation AMM, Mooniswap and more stuff coming.

We then have Shane.

Hi everyone I’m Shane from Kyber Network. Kyber is a fully on-chain liquidity protocol that helps to connect many different, unique liquidity resources into a single on-chain endpoint for aggregators, DApps, wallets and even end-users to use for their liquidity needs. Very nice to be here.

And last but not least we have Lucas.

Hey everyone this is Lucas from Tokenlon. Tokenlon is a decentralized exchange, adopting a request for quotation architecture, and we are built on top of 0x protocol.

And special thank you to Tokenlon for organizing this and bringing this event together.

Who owns the customer relationship in DeFi?

So each of you are builders working at different layers of the DeFi stack. Some of you work on primary markets and others of you are aggregating over the primary markets.

One question that I have that I’m curious for your guys’ perspective on is: Who owns the customer relationship in DeFi? Is it going to be the protocols that own the customer? Is it going to be the aggregators that own the customer? Is it going to be the wallets that own the customer? Is it going to be exchanges that own the customer?

What do you guys think and why?

Let me start. From our point of view the application which is most user friendly. Right now we see of course Uniswap has 20k people swapping everyday. Because the user interface is really great.

Sooner or later, people will recognize more efficient ways to use. Aggregators for example.

We are one of the biggest ones who own like 1000–2000 people every day, swapping on 1inch.

Lots of people use Wallets. They don’t want to spend time to go somewhere else or investigate differently, find the best place to exchange for example. Or lend money. They want to use the most user-friendly application.

Michael or 0x, do you have thoughts? Go ahead, Amir why don’t you go first?

I very strongly agree that in the long run I think it’s going to be aggregators for the most part. There are just so many decentralized exchange protocols out there, in particular automated market makers these days.

And just from a user’s point of view, you don’t want to have to understand all these things: Figure out where the token you want exists and how the pricing mechanisms and all these work, you don’t want to trust all of these things individually with your funds. I think in the long run aggregators will most likely end up owning the user.

I think that’s kind of one sort of customer. The other would be the liquidity providers. I think that’s a bit more of a nuanced answer.

I think you can certainly create moats around liquidity, but we have also seen that it kind of can come and go very easily, depending on how your incentives are.

I guess I would give a little different perspective on who owns the customer. Of course if the customer is the one who exchanges, yes, DEX aggregators are the way those customers come in. But there could be many aggregators and the one aggregator which wins will be the one who integrates the best sources of liquidity or the most of them.

But there’s another side of it. DEX aggregator is just the demand side. But there’s also the supply side, which is basically users who upload the liquidity. For that what matters is two things: One is the safety of the protocol. So if the protocol is safer than my sources of liquidity which I have stored there is safe. And I would prefer that one.

But there’s also the efficiency. There are protocols of different efficiency for different assets.

For example, Curve is very efficient for stablecoins. And for Bitcoin and cryptos we have Uniswap, Mooniswap, Sushi, Bancor, DODO and whatever. For those ones, efficiency could be defined as how much yield do you get per given existing price fluctuations. And how deep the liquidity is, given those fluctuations.

The yield matters for liquidity providers. And the depths of the market for the same liquidity matters for the demand side, for the aggregators.

Let’s say, if you want to compete with existing AMMs. There are really many of them, right. Let’s say if you take Uniswap. It’s actually a good example. Too bad we don’t have Uniswap on the panel. But it’s a good example because there are forks. There is a pretty much 1-to-1 fork of the algorithm — Sushiswap. You can actually see how much the brand and the UI give. What advantage does it give to the technology, and the trading volume to liquidity ratio is about 2 and a half times larger for Uniswap than Sushiswap.

This is the brand premium.

This means that if somebody wants to compete with Uniswap, they probably need to be at least three times more efficient.

So that’s kind of the perspective I’ve learned recently and I’m happy to share now.

Yea, I think, hearing everybody’s opinions on this I mostly agree.

I think when it comes to adoption in the DeFi space, especially when it comes to liquidity, it’s a two-way street. Definitely you can take into account, if you look at the customer as developers, like you guys, you would think of the protocol as very very crucial in terms of the products you can build for your customers. Whereas you look at customers as the end users and traders, sure I think at the end of day in the long run I also agree that users and traders are looking for a platform which allows them to get the best, most competitive prices.

From our experience at Kyber, when we first launched our own interface to our protocol back in early 2018, right after the ICO, we realized that we were servicing mainly the traders, right, the end users that come to our platform to trade.

But through calls through discussions with more developers and in the Ethereum ecosystem, through conferences, we realized that there’s huge potential to increase the whole taker network.

And that will also benefit the liquidity providers especially professional market-makers. They want to come into the DeFi space. They’re always looking at how big the taker network is.

So, you can’t have one without the other.

And who owns the customer really depends on how you define the customer: If you talk about end users (traders) then aggregators you know, very good UI platforms; if you talking about the custom of being developers building new products, then it’s the protocols.

The old tale of supply and demand — which is hard to find?

So that leads me to another question. Actually Lucas I would love for you to take this one. I’m curious whether you have ideas.

Lots of you are touching on the difference between supply and demand side. And noting that there are very different dynamics between the supply, meaning very often the liquidity or the market makers, and the demand, which is the end users who are trading and the customers. Lucas, you do growth at Tokenlon. How do you think about bringing on the different sides of the market and which of the two sides of the market is harder to get on board?

I’d say it’s a great question and maybe for this question I can kind of relate to your last question. Because my perspective is kind of different. Because I believe that whichever product has a higher switching cost for the user will actually own the customer.

So for example for the normal users who are not technical, actually they value the user experience a lot. I believe the user experience is kind of underrated in the space.

This is what we learned from basically integrating with the imToken wallet. So for example, for Tokenlon we ran through multiple versions. We changed our underlying architecture and even the protocol. But the majority of our users don’t really notice, they don’t really care, as long as we provide a price in a reasonable range.

So I think there is some value that users give to the user experience.

So I would say for the mainstream users, for 80% of users, the front-end like a wallet owns a lot of value here.

But for the technical users, for the people who know smart contracts it will be different. They seek numbers, they always want the best returns. The protocol or aggregator will be the most important for them.

Answering your second question, I believe the user side, the demand side will be the hardest one for the products today. Because there are many developers and especially after this DeFi summer we see many new products coming out.

And I think the most struggling part for them will be to attract the users. Because maybe they can have some good VC backing them which helps them to kick off the liquidity. But that doesn’t guarantee for users to come. So they will have to introduce yield farming. But it’s hard to make yield farming sustainable. That’s what I believe in general.

I’d love to hear from the aggregators on this panel. Lucas, you touched on the fact that many of these liquidity platforms are doing liquidity mining. Many of these new startups that are in DeFi do liquidy mining. But they really struggle to bring on the demand-side. The supply-side seems easy enough to incentivize to come on board. The demand side is hard.

What are you guys seeing when you see these upcoming new projects? What’s working, what’s not working? Are there any patterns that you’re noticing? Sergeij. Amir.

I think that they have proven that it is barely doable to build large amounts of liquidity. I think if you go back a year in time there was still this chicken-and-egg problem where you’re looking for liquidity first or demand side first. Had to try to build at the same time. Now with liquidity mining it’s doable to build large amounts of liquidity although it’s debatable whether that’s the most efficient way.

But you know we are seeing pretty large pools of liquidity I would say that are not being effectively used like the turnover is very low.

And I think a part of that is probably just due to the underlying algorithms of those AMMs. If they can build very deep liquidity, but not very tight spreads then they’re just not going to ever be utilized by the aggregators in practice.

Like you can say that I have 40 different decentralized exchanges integrated but 99% of time you’re probably going to 3 of these different exchanges. Unless you have a really big order in which case it makes a lot more sense.

I think generally people are underestimating liquidity they need in order to actually be utilized by the demand side.

From our experience, back when we started in January, we had this algorithm but of course we didn’t have anyone to exchange it. So we came to 1inch. And I asked Anton and Sergeij to basically connect Curve to 1inch so that we have those trades incoming.

And already with just $50,000 of liquidity was enough to start the thing. And basically what made it possible is aggregators like 1inch aggregator in this case. So aggregators do bring the demand and it’s actually very important for new projects to both have very efficient algorithms with tight spreads and to interact with aggregators.

Because that’s what helps to start these things.

Yeah. Agree. When we see any new liquidity source popping up, we are happy like on Christmas. We try to integrate Aave liquidator to liquidate positions on Aave. And others are coming.

Just like Michael said, we bring the volume to those projects and protocols. The market itself finds the best and the more efficient protocol on the market. It choses automatically because I like your first example, we routed only fair volumes.

So for example we have our own AMM, Mooniswap. Theoretically we could make more money if we would have the same volume as Uniswap. But we don’t have to get this brand premium as Michael said as well. So we route only our fair volume. If there’s an opportunity to get better pieces of the liquidity and to a better rate, then we do that. And if it’s not then we don’t do that.

I see a lot of innovations in this space. Of course there are some copies like Sushi. But still one step forward with experiments with governance experiments and so on.

And we have much more liquid than the year before, much more teams working on products. And this is great.

Is token trading going to be a winner-take-all market?

I’d like to ask a question now 0x, Kyber and Tokenlon. Because the 3 of you have primary trading venues for potentially every asset in crypto.

The question is token trading going to be a winner-take-all market? Among the three of you. The aggregators, Curve, they don’t really care who wins. And Curve is like: Ok, I’m only doing particular poairs; Whoever wins I don’t really care.

But you guys are direct competitors to each other. Do you believe there is going to be one winner? If not how are you going to differentiate in the long run? 5 years from now are you still going to be in different companies or will you all be at the same company?

First of all, I just want to say, we are not competitors with Tokenlon. You know.

That’s true.

But you have to ask, who the customer is here. And I think Tokenlon is one of our developer customers. We also have end user relationships with our own user matched aggregator Matcha.

I think the answer is definitely that it is not a winner-takes-all market. And I think if you just like look at the exchange business throughout time, at the end of the day, there are at least a handful of different exchanges

And I think that they’re just too many different products out there in a good sense. There are just so many. And they will end up lending themselves to different types of liquidity models inside. It’s very hard to create a generalized solution that works well for all these things. So at the end of the day, it’s highly unlikely if it’s just that single solution that is able to eat the entire exchange market.

Yea, I completely agree as well. Just to add on to what Amir was saying. For Kyber Network, essentially we are a liquidity protocol but we do have our own in-house UI, which is Kyber Swap. But that is just one of over a hundred takers in our Network.

And 1inch is a major volume driver of Kyber as well. Infact, it has been number one in. If you read our monthly ecosystem reports. It brings in a lot of volume to Kyber.

And similar to what Michael was saying earlier, it really helps if new protocols find that taker network.

When it comes to token trading. Whether it;’s a winner-take-all thing. If you look at just the end-user, the traders, it might seem that they would just migrate to whichever platform offers the best price.

And you would think that DeFi users are always rational, in the sense of making money. But if you look at some platforms in the space where there’s a kind of value that they provide. Uniswap is very simple. it may not be the most Capital efficient when it comes to the liquidity provision. even for token trading. advanced users might want to go to 1inch or Matcha or Tokenlon, where they can get the prices that are more competitive. But there are some of my friends who stick with Uniswap because of the simplicity or just because they are so familiar with it, due to its branding. Its taker rate in the DeFi ecosystem. Michael talks about it as well.

So there are definitely different platforms for different use cases and different users.

At the end of the day I agree with Amir that in the long run that could be people that prefer certain platforms from others no matter whether rationally if you think about just the pricing they should be going to one but they could be other platforms that offer other benefits.

I also agreed with Shane and Amir. It’s really hard to have a single winner that takes all the market. But at the protocol level it’s possible that a few winners take the market. You mentioned Uniswap. Because Uniswap has very good liquidity and support. Many tokens there. Probably everybody like Matcha and Kyber, are already integrating with Uniswap.

And actually for our newest version, we are also integrating with Uniswap. So it’s possible for some asset classes, asset types, for example the new tokens. Everybody will just integrate uniswap for liquidity of these new tokens.

But for the frontend or exchange venue it will be a lot harder to have a few winners. Because there can be many reasons. For example, users can have different preferences in user experience. For example users in asia are mobile first. And also there can be regulation reasons. That some exchange venue will have to stay in some jurisdictional area.

Michael let me ask you a more pointed version of the question. Do you believe that stablecoin trading or mean reverting pairs, do you think that will be a winner-take-all market or do you think there’s room for a second besides Curve?

I don’t know. The stablecoin or mean-returning trading is just a small fraction of the market.

And if we stop innovating, we will lose market share. So we should continue.

I think it’s possible in any market it could be either way. It could be one winner takes all or it could be multiple.

If someone will try to innovate on the algorithms to make them more and more efficient, they may have an opportunity to take the market.

But at the same time if you look at DEX aggregators, I don’t think it’s going to be one winner-takes-all. Because at the end of the day the aggregators will be just about execution, adding new things.

Either way to stay competitive one should definitely move forward and not stay, stick with what works.

I would like to add it’s like in the series ‘Halt and Catch Fire’. Did you see that? It’s exactly the same. Halt and Catch Fire. You have to continue to innovate. Proof your algorithms, invent new things, solve problems.

It’s like in the beginning of Browser, Netscape and so one. Similar to the time right now.

Who knows, maybe it will be some AI which invents algorithms and launches AMMs.

Michael you were speaking earlier about Sushi Swap. I’d like to ask this question to you Michael and Amir at 0x.

What the fork? How to handle being forked as a protocol.

Both of you have been subject to fairly high-profile forks in the past. And DeFi summer has taught us a lot about forking. And the viability of it, where it goes wrong, where forking doesn’t necessarily work.

What have you learned? First of all I’d love to hear the story of what it was like when you experience a fork? What was it like to go through that with the protocol that you build? And what did you learn? And how does it make you think differently about taking your protocol in the future if you have another situation like that?

Yeah I guess the key thing when somebody forks the protocol they will always be a little bit behind. So if you keep innovating, they will not be able to keep up.

Secondly what really matters is the team. If the team can provide really secure things, which are safe, then you will still ahead of the fork all the time.

And it’s not just the code. Because the fork just forks the code. But it’s also the execution. Like the decision they make. Are they safe? And the community, not necessarily always can judge whether decisions are safe or not. Sometimes they can not. Really a big contribution comes from the (core) team.

And people can see if something is safe or not safe.

Another thing is incentives. Right now the total value locked (TVL) in Sushi swap is fairly close to TVL in Uniswap. Still less, but not too far. And I think this is because SushiSwap has the liquidity mining program and Uniswap right now doesn’t.

And of course Uniswap has this brand and many people know it, and that’s what helps a lot.

But SushiSwap is probably 75% of TVL of Uniswap at the time. Because Sushiswap has liquidity mining and Usniwap doesn’t.

That’s important. If you have some incentive, it’s better to not turn them off when you get forked. Otherwise, you could be in danger. It could work, but could also not.

As for example, we had Swerve forking Curve. Right now they reduced their inflation in hopes their token price goes up. Instead what happens, Swerve TVL is just 3% of Curve TVL.

That’s why incentives are important, and in this highly competitive world, it’s better to have them. If you have a fork with incentives and the original without incentives. That’s a dangerous situation.

I just wanted to comment on Michaels comment on the community. I think that’s very important as well. If you just look at Ethereum. There are tons of ETH-killers out there that technically claim that their design is superior. But the Ethereum developer community is not easily swayed by grants or incentives on other platforms. It has to reach a critical mass versus Ethereum to present itself as a moat. Even if its a technically superior platform

And I guess for a liquidity source. The community is not only the end user it’s also the developers who integrate with your protocol. I’m sure you experienced that as well.

Yeah I was going to say something similar. You can fork the technology but you can’t fork the users, you can’t fork the developers, you can’t fork the community. When 0x got forked, was a very different time and day, the end of 2018 or so.

And we didn’t have a user-facing product. And we didn’t have a lot of relationships with market makers and stuff like that at the time. So it actually was a larger hit but you know I think at this point in time it would just be much much more difficult to fork. Honestly these days when a protocol gets forked it’s almost a good thing for those protocols, because now have another group of people who’s working on very similar technology and they are innovating. And for the most part this stuff is all open source and learns from each other. Look at SushiSwap started as a 1-to-1 fork and now they’re going in their own direction. And I think that’s overall a good thing for the space. And Uniswap still has all of their users. I don’t think forks are actually an issue these days.

How did going through the fork experience change your guys perspective? Did it make you say ‘we need to have closer relationships to market makers’ or ‘we need to be more actively active in the community’?

It got us thinking a lot about decentralizing all aspects. There shouldn’t be really concentrated pockets of liquidity or users.

Ideally if any single person left the system, there is some network effect and the users continuing to use the fork getting roughly equal value.

We definitely started focusing a lot more on liquidity at that point. I think that’s when we first started thinking about 0x mesh as well, which is a peer-to-peer network for sharing 0x orders. It creates this place where liquidity can aggregate and liquidity becomes more a public good for everyone building on the protocol. And it doesn’t matter as much of it as a single person leaves.

What explains Uniswaps runaway success?

Make sense. In all the discussion we had so far we’ve kind of talked around the elephant in the room but the one major protocol that’s obviously missing from this panel is Uniswap.

I think we all know that Uniswap has been the runaway winner in DeFi over the last cycle.

What I’d love for to get your perspective on is why? What explains Uniswaps runaway success?

And what is your protocol in particular going to do to try to catch up to Uniswap? I’d love for Lucas to answer this first.

I think for Uniswap. They are able to offer a permissionless approach for listing assets and allow everybody to trade on uniswap. That’s really a key to their growth. For this DeFi summer, there are many new tokens showing up. And people naturally just go there to find a new token to trade. For projects, they will just go there to bootstrap new liquidity because there’s basically no requirement for them. There’s no listing fee or any KYC procedure.

For us, for Tokenlon we view that Uniswap is doing a good job. We don’t really see that we have to catch up with Uniswap. We focus on expanding and bringing users into the space.

So what we are doing right now is actually integrating with Uniswap. Sourcing liquidity so we can offer more assets for our users to trade, and we can grow together.

I don’t remember who mentioned that: If we zoom out a bit, we are actually on the same team.

Anybody else like to answer that question?

I see Uniswap is doing magic. 20,000 wallets using it every day. This is really magic. We from our side don’t understand how they do that.

But they do it somehow. The user doesn’t get the best price on the market but still uses Uniswap. This kind of brand premium. They do a really great job from my point of view.

I want to touch on two things. One is AMMs in general and two is Unsiwap. For AMMs, I definitely underestimated a lot in the past. A big part of their success is that they are very easy to use. Even if the user doesn’t necessarily understand the risks and rewards involved in using them. They’re just dead simple.

And they also enable users to do something they could not do on a centralized exchange. Which is providing liquidity. An average user isn’t going to be able to set up a market making bot and provide liquidity, but they can very easily put liquidity on Uniswap.

I think that’s a big part of it. And I think there’s probably always some demand for providing liquidity in that way.

Returns probably go down over time but I think long run AMMs are probably here to stay.

The question is what percent of the overall market are they going to make up.

For Uniswap in particular, they kind of memed their way into success almost. It’s very clearly not the best AMM protocol on a technical level. There is no thing particularly special about their pricing algorithm that allows them to outcompete everybody else.

There are plenty of others that do consistently outcompete Uniswap pricing. But it just became this shelling point for the ecosystem. And all liquidity mining programs and stuff just were put on Uniswap.

There was a lot of free marketing for them, ultimately a big part of their success.

I think that the opportunity is growing to the point where it’s very very large and more sophisticated liquidity providers are looking at the space now.

Whereas a year ago like the obvious side was just not there yet. It probably wasn’t even worth it for them to build the infrastructure to provide liquidity.

The success of Uniswap is going to attract liquidity providers, more professional liquidity providers that are using off-chains systems, RFQ systems, that are going to be able to do pretty consistently outcompete on pricing with much much much less capital.

I think aggregators are going to end up driving a larger percentage of their volumes through their systems.

Yeah I think that was a very good perspective. I think with Uniswap. The algorithm, although very simple, it’s actually fundamentally pretty good.

Like it’s fundamentally pretty good in terms of the balance of providing market depths and earning some interest on what you provide.

And they were pretty early. So they’ve acquired quite a good user base. So they got this whole premium. And once they got the brand premium, if some AMM wants to compete with them, they should outcompete on the fundamental level by quite a bit.

As I said if we compare Sushi and Uniswap, with the same algorithm, the difference between their volume to TVL ratios is big.

If someone wants to compete with Uniswap, they would need to have algorithms that are three times as efficient. Either on APY, which you are getting for the same with liquidity, or on the market depths or actually both.

A good example would be DODO which I think is not managing to complete and I will explain the reasoning in the moment.

They use external price feeds as a price oracle and build liquidity around that with tighter spreads.

That works great as a liquidity source, because an aggregator would drive volume there because with smaller deposits you have a higher chance to go through the same.

But at the same time, because the used price of oracle on-chain and on Ethereum, it will be always like 10 seconds behind. Just because of the blocktime. They are probably doomed to lose money because of the frontrunning problem. Of course they can increase spreads and do other tricks to avoid that, but then it becomes not so efficient.

They don’t seem to be out competing on that fundamental algorithm problem. Although on the surface it could be seen as they are providing deeper liquidity for the same deposits, so be more efficient. But if I put money there and look at how the money grows, it actually doesn’t. And that’s because of the frontrunning problem.

So they didn’t solve the algorithm-should-be-better problem.

If an algorithm which is better than that fundamentally appears, one wouldn’t be able to compete with Uniswap on efficiency of the AMM.

A good example is Balancer, because Balancer seems to be going a little bit an orthogonal way. The algorithm is just as efficient as Uniswap if you have a 50/50 ratio. And you can provide multiple assets. That’s great. But I think the way they are going is to basically say look: you can make your own pools with any fees, any rations and use those tools to manage your portfolio.

So you just get your portfolios rebalanced and earn a little premium on fees. That sounds like a proposition because maybe you don’t have to have 50% ETH 50% ETC. You may want to have a portfolio of multiple tokens and let them earn money for you.

Although it’s less efficient than Uniswap, it’s useful for those who want to store their portfolio. ANd that’s their way to compete which is not about the fundamental algorithm. It’s more addressing the actual market and that’s a good one.

I fully agree. If you build a new AMM, you have to be 2–3 times more efficient than Uniswap right now. With the same liquidity. To get the same liquidity is quite hard.

We experimented in the beginning of this year with oracles, similar to what Bancor released. And we found out that it’s not the way.

There’s a lot of room to innovate in the AMM space.

I think Uniswap’s success boils down into two points. One I think a lot of the rest of them also covered this has its simplicity for both the end users as well as for the liquidity providers. And secondly, the permissionless liquidity contribution: The fact that anybody who wants to do liquidity mining can actually just reward their anonymous LPs depositing liquidity on Uniswap with even more tokens.

And that, Amir mentioned, helps the whole market for Uniswap. In the long run, as you can tell from Michael’s explanations, for different resources like Balancer, Curve, Uniswap. In the long run there are going to be a lot of different types of AMMs, different types of liquidity sources altogether.

Bancor is trying to innovate when it’s coming to reduce impermanent loss. There’s CoFiX coming up, there’s DODO. Personally I am LP for Curve and Mooniswap along with Kyber. In the long run, there are going to be so many different liquidity sources and all offer something different. So in my opinion and Kyber’s our approach, when it comes to working with all this protocols, is to connect them. At least are the on-chain ones that can work with Kyber’s protocol. To pool all this liquidity from those different sources and make it easier for aggregators and DApps to access this liquidity. And we have been doing some research with members of the Define Alliance and professional market-makers. Professional market makers cannot work with AMMs at all. Because they do not offer any control over the pricing. The capital efficiency is not very high. Typically order books are the most efficient for market making. But when you bring them fully on chain there are going to be a lot of issues and challenges. So what we realize talking to professional market-makers is that for them to come and onboard to DeFi there needs to be other and more advanced systems, which cater to their unique needs. They need to be able to build their own bots, algorithms and customize a trading strategy, adjust the risk exposure.

So all these different advanced techniques can’t be done with just typical AMMs. There needs to be new innovative advanced resources liquid systems in the future. And that’s my opinion but definitely AMMs are here to stay. Just the simplicity of liquidity provision appeals to the majority of the space. Right now the volume is really very high, Uniswap does $400 million on the biggest day. Even more than Coinbase Pro if I’m not wrong.

And the total DeFi value locked is already $14 billion. Right now it’s just a fraction of DeFi’s potential.

So in the future with more liquidity sources, with more retail users coming to the space, it’s not a winner-take-all. (A reference to your previous question.)

What did you learn from the 2020 DeFi summer?

I do want to wrap up with one final question and I’d love it if you could keep this short.

What I’d love to ask each of you is: It has been a year, a year of a lot of dynamism, a lot has changed, a very challenging year, but it’s also been a very good year for crypto. And we’ve all learned a tremendous amount from what’s over the course of the entire year, but particularly in DeFi summer.

So I’d ask each one of you: What is one lesson that your company or protocol is taking away from what you saw and learned about in DeFi summer? Michael, you can start.

That’s a tricky question. You couldn’t boil it down to one thing learned.

If you had to, let’s say for a panel.

I guess what I’ve learned is that things can go significantly more decentralized than one had thought. First when DAOs appeared, they were kind of unutilized. This DeFi summer truly help DAOs to grow. That’s really the way to go into the future.

Also hard to boil it down into a single thing. I would say a lot of our approach has been validated this year. Around the end of last year is when we started building 0x API, which is our developer-facing aggregator product. We’ve seen aggregators really take off this year.

We’ve also just seen as the total volume in the space grows. That these off-chain RFQ systems are actually able to consistently outcompete these systems is growing as well. Just saw our approaches validated over time.

We recognized that security is a very important topic. So we have seen a lot of hacks. And we are very experienced in smart contracts. My co-founder is an auditor. On everything we wrote, we ordered like 11 audits for the last smart contract. To just ensure that we didn’t make any mistakes. Like Michael said it’s very important to keep everything safe and not lose the trust.

We have seen other products losing trust of their user just because of stupid hacks. Also complicated hacks. It can happen. To avoid that, we ensure to stay safe to do audits with all our money. We hope that next year we have more security companies offering better quality services. We are seeing really poor quality on security audits. This was last year for us. Keep safe! Don’t trust auditor’s who write ‘everything is safe’. It can’t be.

It’s a good lesson.

Shane what has Kyber learned over the last summer?

I would boil it down to innovation with collaboration. I guess that’s one way to put it.

A lot of the interesting use cases and value propositions that come out this year were not around last year.

At Kyber we came up with the Kyber pool framework only after we saw such innovation with the DeFi Alliance. 0x came up with the 0x API which is great for other developers to integrate. 1inch has come up with the V2. They have had private transactions recently.

Curve. When I met Michael back at ETHsg, when he was representing NuCypher. Curve wasn’t around right. Maybe you thought about the idea then, but still.

Just constant Innovation. The DeFi space moves so fast. We really need to be adaptable. That’s really something that Kyber learned during the DeFi summer and it’s what we want to bring forward to the new year.

Lucas I’ll give you the last word. What has Tokenlon learned from the DeFi summer?

I think the biggest lesson we learned from this DeFi summer is that tokens are a powerful growth hacking tool.

We saw all these forks happening. The yield farming. They are all distributing their tokens in some way, driving the incentives. That’s also a reason why we launcha token. We are using this as a growth tool.

Also I’d like to mention that we learned that it’s better for projects to have something in their pocket. Don’t give away all your incentives all at once. You need to be prepared if there’s another project trying to fork you, steal your ressource saway, it’s better to have something in your pocket as defense.

Great points.

Ok. That’s the end of the panel. Thank you so much everyone for participating. It has been a fantastic discussion. And thank you Tokenlon for hosting this and bringing this event together. I’m Haseeb from Dragonfly.

And that’s it. Thank you everyone

Bye :)