Exclusive: CoinGecko’s TM Lee Talks Evaluation, Malaysian Crypto Scene, and Interoperability between Blockchains
by Liam J Kelly
CoinGecko has emerged as a solid alternative to the information hungry crypto sphere. Naturally, many are already familiar with first mover CoinMarketCap, but TM Lee’s project offers a broader suite of diagnostics. The traditional offering of market capitalization, liquidity, and price are all on display, but in the following interview, Lee outlines why the inclusion of things like developer, community, and public interest benchmarks are also attractive.
The conversation wanders through Lee’s entrance to Bitcoin, the principles behind CoinGecko’s inception in 2014, current obstacles facing the platform, and what’s on the horizon. All of it reveals the difficulties in wrapping one’s head around such a fast moving target.
First things first TM, how did you first get involved in cryptocurrencies?
Back in 2013, I just happened to wanted to learn more about Bitcoin at that time and I just ended up trading and buying a few bitcoin. I was doing a lot of trading and using CoinMarketCap a lot and I was seeing a lot of pump and dumps, so I wanted to make a way for people to quantify coins beyond just looking at market capitalization. That is principally why we ended up creating different aspects of a coin and use these for evaluation on CoinGecko.
CoinMarketCap on June 4, 2018. Source: CoinMarketCap
Aside from market capitalization, we look at the developer activity as well as at the community activity as well. This comes from places like Reddit, Facebook, and Twitter. We also look at the public interest by examining the Alexa ranking and search engine results as well.
We did that in early 2014 on all the publicly traded cryptocurrencies at the time and we’ve been running it since then. Obviously things have changed a little bit in 2018. For instance, back in 2017 or late 2016, there were a lot of these ICO projects that came in. We’re still maintaining the same viewpoint, in the sense that we want to try to find all the quantitative data and metrics that we can use to evaluate cryptocurrencies. From there we just went out and built that tool as a side project and now we are scaling the team and going full time.
What were you doing before CoinGecko?
I studied computer science, graduated, and then did a bunch of programming and software engineering work for people both as employees and on a contract basis. Typically, the work I used to do before cryptocurrencies was for short-term rental websites. There was an AirBnB competitor working in Singapore and I was working with them for a bit.
Then there was a company based out of Indiana that was doing precision agriculture. Basically, trying to use technology to increase the yield of soybeans and other produce for farms in the U.S. I was always working on the software side of things prior to this.
You earned an excellent taste of different projects and experience before cryptocurrencies.
Yes, of course, but what got me really into cryptocurrencies was my interest in wanting to do a Software as a Service (SaaS) from Singapore, Malaysia. And when I got started, it was really difficult to charge users. There’s PayPal, there’s Stripe, but it’s really hard to get those services working for you in Southeast Asia because you don’t have a U.S. bank account, which was the requirement at that time.
Map of Malaysia, a growing hub for crypto projects. Source: CIA
This is what got me really into Bitcoin. When it came in, people could finally accept payments and people could pay you for software without needing to go through all this hassle. That was the starting point that got me into learning more about this stuff.
And how did you meet your business partner at CoinGecko, Bobby Ong?
I was working in Singapore in 2013 and Bobby happened to know my CTO and he came to visit. This was also at a time when I was already into crypto and I was always talking to my colleagues about it, even my CTO.
Bobby Ong, co-founder of CoinGecko. Source: Mint
Then, eventually, my CTO came and told me “Hey, there is a fellow Malaysian guy in the office, who’s also into crypto, why don’t you have a chat?” And that’s pretty much how we got to know each other. I told him “when I get back to Malaysia at some point let’s do something together.”
In the early days, we were just exploring the space together, just trying different things. We did some newsletters and things. We were thinking of setting up an exchange, but we didn’t end up going that route. We end up doing CoinGeck in 2014 instead.
At that time, CoinMarketCap was kind of the main actor in the space. When you launched CoinGecko, did you get any pushback or criticism?
CoinMarketCap was definitely the top dog back then, even up to today, actually. I think the main reason they have been so successful is because everybody could relate to it. It’s really easy to understand price and market cap. So, in that way, it was really difficult for us. When we put out all these stats like developer and community activity, people don’t really get it. They wanted to know from where we got all our numbers.
CoinGecko offers a handful of other evaluation tools compared to CoinMarketCap. Source: CoinGecko
Based on that, in our early version, we had to be very transparent with how we got our data. We actually put up every single raw number that we obtained from all these sources and we showed it all the way up until the frontpage. When people see that, they say, “ok, these numbers do actually come from GitHub.”
It took a while for people to understand that there are other metrics to look at rather than just market capitalization. And when the market kind of matured, then we could start to clean things up and only show the higher levels of numbers and if you want to, people can now dive deep into the site to see the rest of the raw data.
Initially, it was a little bit rough introducing all these different metrics, but over time the market picked up and people got a better idea of what it is they were looking at.
Earlier you mentioned the rise of new projects and ICOs in 2016, 2017, and into 2018 forced you to change some of your metrics. What did you mean by that?
There was a lot of ICO projects that launched two years ago and now there are a lot people who are trying to do ICO evaluations or ratings, but their methodology is a bit different than what we have for a few reasons. Number one, in a cryptocurrency project (i.e distinct from an “ICO project”), the team isn’t necessarily the most important aspect. This is because teams are often started by anonymous people, or by people who’s background you don’t know and then they prove themselves with a successful cryptocurrency project.
On the other hand, the team matters a lot more with these ICO projects. You need to know what their past experiences were and what else they have been working on. It goes both ways too as they need to put a lot of emphasis on the team.
— Yassin Hankir #savedroidico (@YassinHankir) April 18, 2018
Another thing to keep an eye on is the total supply of the coin. In the early days, for all these mineable cryptocurrencies, they needed to be mined over time in order for teams to offer whatever supply has been circulated. These days, the coins can be minted out of thin air. The evaluation on market capitalization will also be much more tricky in that sense. For instance, an ICO project can claim that they have X number of coins publicly circulated, but you really need to do your homework to figure out if that number is true.
Also, when ICOs came in, they started to adopt community services like Telegram, Discord, and Slack, which in the early days was much different. Most crypto projects back then, never saw themselves as building communities on all these channels. Reddit seems to be the one that has stuck around since then. Another example is also something like GitHub. These days most of the ICO projects, when they do their fundraising, they don’t necessarily have an actual code base on GitHub yet. Typically, it’s more of a token contract that they’re issuing.
This just in from the Tari group on Telegram…! pic.twitter.com/KYyOPoviEV
— Riccardo Spagni (@fluffypony) June 4, 2018
Whereas for cryptocurrency projects, they typically have have a code base out there for you to run a full node and you can use that as a gauge on how active the project is. These are just a few examples of how things are different and right now we’re evaluating ICOs in one way and cryptocurrencies in another way.
That being said, we’re looking to change these models and have a solid metric for both of them in the next few months.
Will that mean you will add some more metrics?
Maybe not. We haven’t really thought about that yet, but I think our four pillars are the major pillars that we can rely on. We probably just need to revisit the way we evaluate them. Currently, for ICOs, we are not doing any sort of evaluation. Of course, when they become publicly traded, then they will fall under the same model as we use for cryptocurrencies. At the moment, however, what we’re doing for ICOs is aggregating all the ratings and comments that people in the space are giving and putting it in one place.
Bitcoin, for instance, has a high percentage of developer input and good community support, but it only has a 43 percent percent public interest rating. My question is as follows: do you think there are other projects that maybe have a high developer rating, but a low community rating? And if so, do you think these coins get missed out when using CoinGecko’s metrics?
Our evaluation metric currently takes into account the total developer stats during the lifetime of the code repository, rather than that data over time.
For GitHub, for instance, there is this popularity measure which shows the number of stars and the number of issues posted. We’re currently taking the total number of this information rather than a time series of it.
Binance Coin on CoinGecko. Source: CoinGecko
If you look at dogecoin, for example, that coin has a pretty good developer score score because in the past it has seen a substantial amount of involvement from developers. Over time, however, it started to die down a little bit.
Dogecoin on CoinGecko. Source: CoinGecko
An improvement that we are bringing in soon would be to evaluate developer metrics based on time series.
We believe this will paint a better picture of how active a developer is at any given moment. This is in contrast to our current method, which just measures the amount of commits or how many stars a developer has obtained since the beginning of the project.
What are the obstacles and daily challenges of running CoinGecko? Is it a daily activity or do you simply work on it here and there whenever you get a new idea or need to update something?
Since things in the crypto scene have picked up, Bobby and I we have gone full time into CoinGecko. It’s definitely a daily routine. We have actually started to expand the team. There are about six or eight of us in the main office here in Kuala Lumpur, Malaysia, but we are also hiring contractors from all around the world. They’re working with us remotely both on a part-time and full-time basis. There’s definitely a lot of things that we are doing now and that we plan to do.
Right now we are trying to experiment with features. The way we do that is we hire a few market researchers who come in and look at our data and try to see if they can find some insights out of it or could they use this data and present it in a way that r new users coming to the crypto space can relate to. If you have seen our quarterly report, those are instances of what our market research team is doing. They were able to come up with something that even people who are not in that space be able to relate to.
ICO and Airdrop Data from CoinGecko’s Quarterly Report: “ICO interest remains strong, continuing the strong momentum from 2H 2017 despite crackdowns by governments in certain jurisdictions and concerns over tax and regulations.” Source: CoinGecko
We use all this market research as an opportunity to learn from what users want and like and then we can fit that back into the product itself and turn it into a feature. That is definitely what we have been working on every day. We’re trying to find ways to get this data out and to make it easier for people to make sense of.
The other problems that we face is hiring because talent in the crypto space is hard to come by. You don’t normally find people who are particularly interested in the space, especially engineers in this part of the world. The decentralized nature of the whole industry, however, doesn’t really limit us on who we can hire which is why we have a small core team here in Kuala Lumpur, and we make sure that we take advantage of that fact, but we also recognize that there are people on the other side of the world who are interested in the space and can contribute too.
Are you guys often approached by projects for consultations?
We do get that kind of request, but we are not taking that at the moment as we are focusing on CoinGecko the product. We want to focus on product and we don’t plan to do any sort of consultation.
Are you guys in contact with the Malaysian government? Do they come to you when they’re curious as to how best to regulate the space?
At the moment, there’s no real regulation. They care a lot more about the exchange side of things and they hold a lot of meetings with wallet providers and people who run exchange services rather than CoinGecko. We actually see CoinGecko as serving a global market despite being based in Malaysia.
Legality of Crypto from CoinGecko’s Quarterly Report: “Cryptocurrency and blockchain gained attention from the major regulators in Q1 2018, with most governments remained positive over the impact of cryptocurrencies to the economy as shown from the results of G20 Summit on March 19-20, 2018.” Source: CoinGecko
How would you describe the space in Kuala Lumpur? How did you hire the people who are working in your core team?
Through friends of friends of friends and also we are quite involved in the crypto community in Kuala Lumpur. Although cryptocurrencies is not as big as say Singapore, China, or Korea, there is definitely a niche. Within this niche there are engineers, traders, and entrepreneurs. By being involved in these meetups you tend to find the people that can help you one way or another. There is also a blockchain center opened by the guys at NEM here in Kuala Lumpur. It mainly serves as an incubator and co-working space.
To look at Kuala Lumpur and Malaysia, in general, I think historically there was a lot of mining activity going on here. Of course, now it’s much more competitive, but there’s definitely a good amount of traders here as well. We are starting to see a lot of projects coming from this side of the world as well.
You see a lot ICO projects and right now the Malaysian government is trying to understand how cryptocurrencies work while trying to regulate exchanges. You do also see some people trying to start an exchange based here as well. I would say we are making pretty good progress because the government is not putting on any kind of blanket ban. The government is really trying to work together with the people in the industry.
Are there any projects in particular coming out of Malaysia that you’re excited about?
There’s actually one project that most people do not know they came from Malaysia, Etherscan. There’s a couple of ICO projects coming out for sure, but I don’t want to really name them outright. Most of them are mainly startups that have already raised venture funding and are now planning to do an ICO. It might be a good or bad thing, depending on how you look at it.
In which sectors are these ICOs emerging? Are they targeting finance or digital currencies specifically? Or is there another niche that they are pursuing?
There is definitely a wide range, but one ICO project based in Malaysia called HelloGold is interesting. They are basically tokenizing physical gold and did their ICO sometime in 2017. Prior to having their ICO, they actually received venture funding too.
They started off as trying to give the unbanked, or people who have no access to purchasing physical bullion, an opportunity to own some small fraction of gold through a mobile app. That’s where they started, then they adopted a blockchain, did an ICO, and are now trying to promote a stablecoin, or a gold-backed token, in the ecosystem.
I would say that Singapore has a much longer list of ICO projects that it wants to do, however.
Do you have any final sentiments for what you might think could happen in 2018 or 2019? What does the space really need in your opinion and what sorts of things are you looking out for in the future?
I’m not very good at predicting the market, but my gut says that we are in a bubble. So, my guess is that I expect a correction in the short term.
Outside of that, I also think that this space requires a lot of fundamentals. We would like to see more products, services, tools, and infrastructure in the ecosystem rather than just new blockchains that are pretty much doing something of the same. Basically, we need more tools to reinforce the ecosystem and make it more accessible to the masses. That’s one.
Number two, I would like to see is more interoperability between blockchains. There are some promising projects out there that is trying to solve this problem. The reason why I like this kind of project is because there’s no one chain that can rule them all. By the looks of it, there’s always going to be tradeoffs. There’s always going to be something that a chain is good at or not good at. If you have some interoperability, one chain can take advantage of the other chain’s advantage.
4/ There was a point in time when I believed that there would be “one chain to rule them all” with tech like sidechains enabling peripheral chains to peacefully, but secondarily, exist next to bitcoin.
But the realities of entrepreneurship and open source development soon set in
— Jeremy “Chaosnode” Gardner (@Disruptepreneur) May 22, 2018
The third point is decentralized exchanges. I think they have a very important role to play in the whole ecosystem for a few reasons.
First, it’s very dangerous to see a centralized exchange collapsing. So it’s nice when decentralized exchanges give an option to trade and safeguard their currencies.There are a lot of tokens being launched not to mention the non-fungible tokens they’re launching soon. All this needs a safe way to be traded between the users and a decentralized exchange will be a great tool to facilitate this kind of operation. A majority of the ecosystem will massively benefit from this kind of thing.
BTCManger would like to thank TM Lee for the time he has devoted to this interview and the insight he has provided. We wish him, Bobby Ong, and CoinGecko the best of luck in the future.