Building Products in Web3 Data Space

We have an exciting keynote from the representative from CoinGecko on the topic of building products in Web3 data space. I say this is also an important session for builders and product managers that you cannot miss. As building an effective product, it's not easy at all. So without further ado, let's give a warm welcome to Kim, Product Manager at CoinGecko. Welcome, Kim.

Hello, hi everyone. So today I'm talking about our experience building product in the Web3 data space. So... Yep. So, let's go next one.

So, a little bit about myself. So, I started building products at CoinGecko. So, as a product manager for CoinGecko for the past two and a half years. And, yep, that's basically me. And also part of the contributor for EVE.

KL, which is one of the community that supports this event as well. So, shout out to the team and yeah just want to ask you know any coin users here anyone that visited coin.com before or uses our mobile app cool quite a few users here i would love to talk to you if you have any feedback about coin gecko um definitely like to talk to you guys about that but yeah today um for those i guess some that doesn't raise their hand, just want to share a little bit about CoinGecko. So, CoinGecko is one of the largest independent data aggregators, and we've operated since 2014. And yep, this year is our 10-year anniversary.

We've been building for the space for 10 years. And we are basically a platform that just aggregates token data, such as price, market cap, or even like trading volumes across all tokens, all exchanges, or even different chains as well. And it's founded by TM Lee and Bob Yong, which is both our co-founders. They are not here today because we also have a conference going on today at ICONSIAM. So I'm here representing the team to share about our experience building product here.

But yeah, we track thousands and millions of tokens data in the single platform. So we have like 15,000 coins tracked on CoinGecko, 1,000 exchanges, 4.7k NFTs, 190 chains, 4.9 million pools tracked on CoinGecko as well. So we index all this data and show the final price of tokens and all like market cap and other information to our users.

Okay, so today mainly the main thing is to share is basically building products in the Web3 data space. So we've been building for quite some time. Just want to share some of the learnings that we have when we are building throughout the building in the Web3 data space. So I think one thing we always kind of highlight is that internally, even internally or within the team, is basically the Web3 space is always changing. So changes is always something that is a core part of the Web3 business, where there are new data points that come in, there are new changes that come in, there are new primitives that come in, new narratives that come in, and we always need to adapt to that.

So, I just want to highlight one other example of the landscape changes that happens on the WebG space, and how we navigated towards it as a product builder. So, first, in 2010 to 2015, this is the era where Ethereum haven't existed yet. There are only Bitcoins, Litecoin, or sometimes you call this native network coin, right? Where this is a new chain and there's only one coin, which is a native coin on the chain itself, like Bitcoin, Litecoin, etc. There are no programmable crypto yet, and this was the era for 2010 to 2015.

People believed that this is a digital goal. People believed that every chain, only chain can only can hold, only chain can have one coin and the one coin should be the currency of the world, of the network itself, right? But, you know, like, the landscape changed in like 2016 and onwards, right? Where until now, it's still the same thing, which is Ethereum introduced like the world computer concept where they want everything to happen on Ethereum, right? It should be a world computer concept where they want everything to happen on Ethereum, right?

It should be a world computer. Everyone can customize how they want things to work, introduce smart contract, and also introduce Solidity and all these different things as well. And after the world computer concept was introduced, we're seeing people start to deploy ERC-20 tokens, right? There are now concepts where you can have more than one coin on the chain. You can say, today I want to create a new coin.

I just deploy to the chain. I don't need to start a new network from scratch. So that's a new primitive, right? Last time, if you want to create a new token, you need to start a new chain, right? Now, you can do it easily, just deploying a contract.

And that emerges as well, right? Because we have ERC-20 tokens, then where to trade these tokens, right? And they are a centralized way of doing things, but because of a smart contract and a programmatic crypto, then people can start to code up DEXs, where you can trade tokens on-chain, permissionlessly. Then, 2019, until now, still the same thing, which is we're seeing more chains, right? Because it's kind of proven that the model that Ethereum is evangelizing, like a world computer actually works.

People want it. People want more customizability. People don't want to spin new network for just a small use case. And a lot of more chains come out to tackle different problems. For example, some EVM chains or even SVM chains, non-EVM chains, some of them want to be faster, want to have lower latency.

Some want to be focusing on AI. Some want to be focusing on, let's say, DeFi, right? So, a lot more new chain emerges like BNB, even Solana, or even Polkadot and more. And because we are going to a, like a multi-chain era, we're seeing more and more, we're seeing bridges become more and more important, right? Because if you want to connect between two different chains, due to the oracle problem of crypto, there are no ways for two different chains to communicate.

That's why bridges solutions come in, where you have a separate network that can connect between two chains, you can swap assets, you can move your assets between two different chains. Yep, so that's for 2019 to now, and 2021, NFT become a hot thing, right? Last time, people only trade fungible tokens. Now, non-fungible tokens comes in, right? And people are speculating it, or even people just think that NFT is a very interesting concept where you can have something permanently yours and non-fungible that happens on chain and recorded on the chain itself.

Yep. And 2022, we see roll-up technology matured. And there are more roll-up spin-up through the past years. So, for example, I think a year before or two years before, we have like less than 100 chains on Coin.agore.

Now we have 200. And we do expect it to grow way more in the coming years. Because now to spin our roll-up atau bahkan L2 atau L3 adalah seperti satu klik. Anda bayar $3,000 sebulan, anda sebenarnya boleh mempunyai L2 yang baru. Dan dikira ada lebih banyak jenis rata yang ada di dunia, is also like these chains all have different purposes.

Some are like general purpose roll-ups, maybe like scroll or zk-sync where they are like we don't focus on like DeFi only, NFT only or something. We just do everything. And there are some apps basically roll out like AVO which is focusing on like trading or even like trading futures and all these different things. And all of them have different primitives, right? Some of them focus on tokens, some of them focus on NFTs, etc.

And that's what actually happened since 2022. And now, I guess this year, everyone knows about meme coins, you know, meme coins gain further adoption, everyone starts to trade meme coins. We're seeing launchpacks such as Palm.fund emerges, and even on Coin.weco, we observed that there are way more tokens created this year than the years before.

Like millions of tokens are created even close to in the past few months. So it's quite crazy as well. And that's basically it, right? Where we're seeing that the crypto landscape or we keep on changes as the market evolves, right? And so, I think the question we always ask ourselves is that, okay, so let's say these landscape changes mean to our product builder in the data space, right?

So, people starting from only Bitcoin, going to Ethereum, going to multi-chain, going to NFTs, going to roll-ups, right? And so, going to, let's say, these mean coin primitives, then what as product builder we can do? So, I just want to share an example of, like, some of the actions that we take to actually, you know, like, to keep up with the market and keep up with the narrative changes, primitive changes, so that we can serve our users well, right? Because, let's say, people are focusing on Layer 2's chains, right, IO2's tokens. But those tokens are not, those tokens data are not available on CoinGecko.

You just wouldn't get fulfilling because you can't find the information that you want and that's why we need to keep up to date with that and serve this data for our users. So we just want to share a little bit here which is like, for example, 2010, 2015, right? We only track, Koneco hanya mengekalkan aktiviti pada sekses kerana kebanyakan penggunaan trading ini hanya berlaku pada perubahan sentral. Tidak ada banyak perubahan sentral yang boleh ditemani. Mereka seperti shape shift dan semua perkara ini tetapi mereka bukan AMM dan DEX yang kita fikirkan seperti Uniswap dan mereka tidak sepenuhnya, mereka tidak sepenuhnya terbentuk juga dalam sesuatu maksud.

Jadi itu adalah apa yang berlaku pada decentralized as well in some sense. So, that's basically what happened in 2010, 2015. We only tracked central exchanges and that's kind of like a source of truth for token prices, right? Because you need kind of like a bit of spread to actually understand what is the real price discovery of a token. And that's 2010, 2015, right?

In 2016, Pada tahun 2016-2020, ada lebih banyak token, crypto programmatik datang, orang boleh mengembangkan kontrak pintar dan kita melihat lebih banyak token yang dikembangkan di dalam rancangan kerana ia sama mudah seperti mengembangkan kontrak pintar, kode soliditi dan ada standar yang sudah ada, anda tidak perlu membuat perubahan, anda hanya perlu mengembangkan. and we see tokens deployed on chain and we at Coin.ag also start to track these tokens on chain as well because users want to know what is the price of these tokens, right? People don't just care about coins. The primitive move to tokens as well and other things also like last time all the trades, most of the trades actually happen on central exchanges but since, you know, DEXs comes in and getting more market share, people really appreciate the non-custodial approach of like trading tokens, right?

Where you don't need to put your funds on a central exchange to actually believe it to store your crypto. Now you can just connect a wallet, you just swap in on your behalf. And we see trade started to happen on DEXs, right? And we also want to capture the data Because every trade actually signifies price discovery of tokens or even market cap or trading volume of tokens, which means something to the users, right? If you are a trader or even you are just a researcher, all of those data matter for you.

So we started to track that as well. Then in 2019, as mentioned, we're seeing more chains come in, right? And some are even non-EVM chains. For example, Solana and like some others as well. Then we're seeing, also seeing tokens start to deploy on multi different chains.

For example, a coin can last time only can be on Ethereum. Now it can be on Ethereum and some L2s or even some other L1s, right? For example, USDT is available, I think more than 10 chains, right? Then how you track that? So we also to track all those tokens because it's important for our users to know the supply of the token, right?

Because it's distributed across so many different chains. How can you know what is the real final supply of the token itself? So we start to track that and we think that's important and we make the shift to actually try to track all these different chains that are available in the world. And other than just know know like the tokens deployed on multi different chains, people also want to move coins to different chains as well. That's why bridge token emerges, right?

For example, all your like wormhole, like layer zero, all this stuff actually comes in and we're also seeing that and we start tracking that as well. Then, okay, 2021 NFT become a hot thing. It was there for a very long time, just like people don't really find the value of it. And after that, 2021, it become a hot thing. So NFT is also another challenge because it's a different token standard.

Rather than a fungible token, it's a non-fungible token. So how to track it is a slightly different thing because every single token have slightly different value. They are not fungible at all. So like, there's one challenge that actually faced and one reality. And users want to know, for this NFT that they buy, how much does it worth?

How much trading value they are? Then we start to track that as well. And we see NFT starting to trade on NFT marketplace, not the traditional exchanges that we're aware of, like AMM pools or even like a bid-ask spread kind of juga perubahan di sana dan juga ada metri baru yang datang bukan hanya kali terakhir orang hanya berkata apa harga akhir dan sebagainya. Sekarang orang bercakap tentang harga lantai NFT. Jadi ia sangat berbeza dan juga orang mengatakan volume, jualan atau bahkan aset total yang dapat diberikan oleh NFT.

Jadi perkara berubah lagi dan again. And 2022, you see more roll-up comes in, lower entry barrier for starting L2 and L3. You can really start, you can really just deploy your L2 and L3 today if you want, using service like even Cartesi or even like some other raw as a service platform, right? So maybe in a week or two weeks, every one to two weeks, there will be a new L2 comes in, then how can we track that? How can we track all these different assets and how can we link between these different assets as well?

Some tokens are on five different chains and how can we say, okay, this token on this chain A is actually the same deployer as the token from chain B. So these are the things that comes in that are complicated as well. And we also see tokens deploy on Kita juga melihat token dikeluarkan di pelbagai jenis syarikat yang berbeza melalui kontrak berbentuk. Oleh kerana ROAP adalah konsep yang agak berbeza daripada dua L1 yang berbeza kerana mereka mempunyai kontrak berbentuk di L1 sendiri. Jadi bagaimana anda boleh melihat itu dan bagaimana anda boleh menggabungkan antara mereka seperti bagaimana anda boleh mencari korelasi antara aset juga.

that comes in, you have exploding amount of new tokens, right? And how can you make sure that you are able to discover all these tokens in time and in a very, very short time because maybe in every second, there are one new token, five new tokens, how can you keep up with that? And also, like, the primitives also changes, right? Last time, you just deploy your token, you just trade on DEXs, right? So there are new concepts that come in as well where the way people trade tokens is not just through DEXs anymore.

You go from a launchpad, then call DEXs or even people trying out new stuff nowadays as well. So this is always the landscape changes that actually happen in the space and other than just landscape changes right, what user need also changes. Last time people only care about like price, trading volume, debt spread right. Then nowadays people care about number of buy sells, number of holders because like if you want to further advance your speculation, you want to know all the early leading indicator towards a lagging indicator such as price. And people care about what is the number of buys, number of sells, and number of holders that comes in.

Tokenomics, people care about it, like circling supply, total supply, float, vesting schedule. Last time, people just care about circling supply because it's the market cap, right? And people start to care about FDV. Now people care about market cap over FDV because they indicate how many tokens are still not circulating. And even vesting schedule, people want to know when is the next major unlock for these big tokens so that they can make investment decision as well.

And more like TVL that comes in, TVL is really a matrix that is like really meant for crypto in some sense. It's not really available in the traditional industry. And people, other than TVL, people also care about those traditional metrics, such as fee and revenue. They want to know how much money a protocol is making, right? And also inflows, outflows.

Let's say you have a token deployed today. How many inflows, how many outflows of a token? Categories, people want to know, given there are 15,000 tokens in CoinGecko, how can you make sense of all of them, right? you need to divide them into categories so they can people oh ai coins they are here there's some deeping coins here there are some main coins here etc so you can understand the tokens better so what we learned throughout the journey i just want to summarize the three different points um because we i guess the highlight here is what the landscape always changes and we need to adapt towards it so i think first thing is basically understand what your users demand from you and add that to landscape changes so understand what's the core value of your platform and what your Pertama, anda perlu memahami keperluan pelanggan anda dan menyesuaikan keadaan. Fahamlah nilai utama platform anda dan keperluan pelanggan anda.

Untuk CoinGecko, ia berkaitan dengan harga token dan data yang terkait. Saya percaya bahawa tidak ada hari yang pelanggan tidak mahu memandangkan CoinGecko. Mereka tidak mahu memandangkan harga token. to go. There wouldn't be a day, right?

And there wouldn't be a day I say, I don't want the price of a token, right? So we believe in that and we keep on striving to make it better. Invest into our core products, invest to our core infrastructure to make it happen and if there are new primitives that our users care about, we should strive for it. Next thing is data sources. Last time, I think I mentioned 2010, 2015, people mainly trade on Centride exchanges.

Now, a lot of things actually happens on-chain and on DEXA as well. So, from Centride sources now going to on-chain, right? Blockchain can do way more now, right? Unlocking way more different and creative use cases. So, how can we understand all the data on-chain and make sense for our users as well, right?

But one thing I also want to highlight is that Bukan setiap data berada dalam syarikat. Malah sekarang, contohnya, saya hanya memberikan contoh yang sangat sederhana. Katakanlah, token sebenarnya telah berpindah ke kontrak baru. Biasanya, ia tidak dikira-kira, katakanlah, kita berpindah ke syarikat baru ini dalam syarikat. Ia lebih seperti perkara yang tidak dikira-kira.

monitoring the space closely, you might miss that out. So we also want to understand how can we get this data sourced accurately. Last one is stay in the loop. We always believe that for us, it's basically, we constantly try out what are the new protocols, new chains or even assets in the world, and also talk to our users, understand how these are useful and important to our users, so we can continue to deliver value to our users, and make sure that it is contributing to the core experience, which is users care about the tokens, the prices, the data points associated to it on our platform. So that's about it.

I shared about the journey of building the Web3 data space and some things that we learned throughout the journey. So that's all. Thank you. And follow us on our socials. We are on Twitter and all social platforms that you are aware of.

So I'll be here for a while as well. If you'd like to chat, I'd definitely like to chat with you as well to share your experience. I want to listen to your experience with CoinGecko. And maybe if you are curious more about building the Web3 space, I'd like to chat more. Thank you.

Wow.