Hi there, and welcome to the canvas podcast where we bring business and data leaders together to talk about how to make data easier for everyone. Today, I am super excited to have one of our investors, and CEO of sendbird. John Kim on the show. So thanks so much for coming on. John, do you want to start by maybe telling us about yourself?
Yeah, well, thanks for having me here. So um, just to give you a high level introduction about myself. I'm founder CEO of sendbird cembra. We are the world's number one conversations platform for mobile applications. Basically, powering user to user communication, as well as a brand new use for communication within your mobile app. For instance, we have customers like DoorDash, who uses us for communication between the delivery person and the end customer. We have a lot of healthcare customers connecting doctors to patient, like Walgreens, and then also a lot of like dating apps. You know, one of the world's largest online communities, like Reddit uses us for their community chat as well. And bunch of games and live streaming and sports etc, kind of use us for that user engagement plus conversion for some of these commerce businesses. We power over quarter billion users on a monthly basis sending 7 billion plus messages, were seriously raised $220 million in funding valued slightly over a billion dollars. And then employ about 330 40 employees globally in seven different countries. The third is my second startup. I did my first company, way back in late 2007. Yeah, perfect time to start a company with the financial crisis, you know, waiting right around the quarter. Again, experience for the second time here December 1, couple of weeks of social gaming company grew to about 5 million users, which got acquired into info. And then kind of side by side fun fact is used to be a professional gamer curious and the undefeated champion in Unreal Tournament and played a lot of Quake, two Duke Nukem 3d, etc. So that's a little bit of me. Amazing.
Congrats on that congrats on all the growth. Maybe let's start with how you got started in technology. I can probably guess by the the gaming background. But, you know, I saw you studied AI as an undergrad. So maybe what drew you to technology and data specifically?
Well, I mean, technology goes way back, right. Even as elementary schools who actually my mom found a diary that I used to write, I thought I had this like fake memory of like growing as a very well rounded child, you know, playing the playground or doing all this fun things. But in my diary, every single page was about computers, right, using PC tools to edit new games, Hex editing, using GW basic, and like whatever those programming languages back then to create, like, you know, very basic elementary games. So pretty much all my life, I guess, as a child, I've been in love with technology and science, pretty much. I think my dream was actually to become a scientist one day, then having gone through or work at work, a couple of technology company, I kind of saw this potential to build tools that increased productivity. Because really, across humanity, and I kind of wanted to do that for the rest of my life. So that's kind of when I kind of pivoted from trying to become a scientist to becoming more of engineers, how to build tools that I can actually really see the world get feedback and iterate and, and really feel that meaning and purpose from it. And my background is went to into university as a web, electrical engineering student, but kind of pivoted to computer science. Because I want to do actually build a lot of stuff myself, myself as an engineer, programmer. And I just want to experience that kind of really threw me into this, like data and productivity, in general is I used to work for a company called NCSoft, which is like one of the largest gaming companies in the world. Previous to that I worked as an engineer and software engineer at a company, that company shut down, they raised $10 million, I was 100 100, employee n. And that was around the.com. Bubble days, when it started crash, they couldn't fundraise anymore. And then I was, you went from 100 employees to 10 employees. So last engineer, who inherited about 15 engineers work. So it was the next company, I'm like, Okay, I want to try something else. I actually went to the other sites like working for a business unit at NCSoft. And they quickly realized a lot of people were using doing manual work like using word processors and spreadsheets to crunch the data. Use kind of like macros in Excel spreadsheets to increase, you know, build kind of less hacky productivity tools. But because, you know, I had a background in engineering, I started building observing what people did actually build a internal operating tool and that was actually not part of my job description, but I kind of did it As a side gig in the evening, evening, after work, normal working days was over, I started like spinning up my own my own, like Linux servers, my local basically desktop. And they start programming this operating tools that I wanted to use myself. And I quickly realized how much of that can actually benefit the rest of team, my manager found that I was like, Hey, what is the tool that you're using? I'm like, Well, it's kind of a tool that I want to use for myself. It's like, well, if it's good, can our team use it because our team was like, 160 people large. I'm like, I guess we kind of did a rollout within our own team, and still sort of status, like the actual officially became the official tool within the company. So that really gave me a positive experience, like, hey, if I build something and get feedback and iterate on it, there's a lot of value and concrete. And a lot of that has to do with like, handling data, and running automated reporting, increasing productivity for the organization. So that's kind of how I guess we've got it as entrepreneurs churning out building software, and then turn it down into something that's useful. Amazing.
Yeah, I think we had some experience of seeing seeing problems internally at Flexport. And trying to trying to build tools to solve them, especially on the operational side. So we've covered a little bit of how you got into this space and your first couple of gigs, what what led you to start sendbird.
So kind of like, I would say, it's not a fun story, but but background is like still to this day, the majority of my career was in b2c, like my first started was a b2c company, web 2.0. And then when financial crisis hit, all of a sudden, nobody wants to invest in b2c company or web 2.0 companies anymore. But because I have background in gaming work for a gaming company, or software engineer, our investors, like, Hey, if you build games, we'll invest. I'm like, Okay, now I have to make games now, because I didn't want to fail. So we pivoted to games. But instead of just normal games, I build social gaming. And then with the second company after selling my first company, I started this company with a buddies from my first startup. So we've been working together 1213 years now. And we started out building this social network for moms basically find other moms in your area with similar aged kids buy and sell used baby products do Q and A's set up playdates and things like that. And it was around 2015, when like the entire world was talking about messaging app, conversational UI chatbots. Remember, like, everybody started talking about those things. And so like WeChat, took over the world, Telegram WhatsApp, like, everyone's like talking about missing apps. And we want to like add a missing feature for our own application. But we had built like chat four times in our gaming company before. So our CTO was like, I'm not gonna do this same thing over again, like, Why reinvent the wheel? So we went out on a buyers journey, like, Okay, well, first, let's try to all the free stuff like open source, etc. A lot of open source for like, you know, optimized for desktop experience, like the JavaScript plugins that you find on random websites. So we didn't really it wasn't really mobile optimized. So then we went on to use like Firebase, we built our like version, point eight on top of Firebase, it was kind of easy to build like a POC and prototype. But once we got to, like really sophisticated, sophisticated user experience, there's a lot of work around, so the hacky things were to layer on top. So we ended up ripping that out again, and then built this entire chess set from ground up. And then we're kind of running out of money. But I saw this like, notification from yc. It's like, hey, apply to yc. Now, it was like, a week left? Oh, no, actually, it was a day left. Basically, I was helping out under a friend to apply to yc. So I'm like, okay, maybe I should apply to. So we applied to YC with that idea. And then our partners turned out to be Michael Seibel and Justin Kahn, obviously, CO for as co authors of Twitch, they knew a thing or two about chat. So they had a lot of questions about chat. And we had scaled to like a million users back then. Like, okay, we solved this. Here's how. So that's how we kind of got into yc. And then yeah, that's 466 and a half years, you'll be our current quarter billion users.
And that's timber. The last minute YC see application Did you recruited did you have to record your videos back then and submitted the same day or
so we actually apply to YC with the mom's social network app, once we we got to interview but we didn't get in. But second time was I was actually trying to help out my friend to apply to YC was interesting. I saw basically a copy pasted my mom's application application, but it just changed the business part, the team and everything else was the same right. So we got an invite to the interview again, but this time with just a different idea. But it was like last minute they kind of funny sort of like my friend who I was helping actually did not get in and we got in. So because of this kind of awkward moment. Like okay, sorry, man. I'll buy you beer drinks. But, yeah, so we were pretty lucky there but just kind of kind of this story back story.
Yeah, I can't bear to watch our or YC video application, It's too, too awkward to go back. And so, you know, you mentioned, do you have, you know, massive amount of users at sendbird, you're sending, you know, a massive amount of data. And, you know, one of the things that's interesting about sendbird is right, you're doing real time, you know, conversational data, where, you know, most traditional data pipelines today are our batch. And so I'd love to hear like, what are some of the unique data challenges that you've run into? Over the years at sendbird? And how'd you? How'd you work to solve them?
Yeah, so, I mean, obviously, at this point, I, there's so many hurdles that we have to jump through. You're absolutely right. I mean, the Data Data Challenge is real. And also, we are dealing with real time socket connections to scaling a concurrent connection is non trivial compared to scaling, you know, HTTP, stateless connections. So there's that. And then the way the data flows in, if you think about the moderation capabilities, everything has to be real time. Then there's ton of web posts related to the message for every single message that gets sent. There's so much metadata, right deal receipts, we received having indicators, everything that's surrounding that, plus all the metadata. Plus, if you're doing if you're sending a message, there's a prevent workbooks that trigger certain amount of data, and post event were posted tournaments, certain amount of data, then there's the audit logs, a lot of analytics, that get piped to other systems. So there's just we never had an issue of lacking data, we have an abundance of data. So we're dealing with a lot of different infrastructure. One thing that kind of be super transparent, we kind of stopped stumbled into sendbird idea, right? Our main business was consumer application, social network. December was kind of like a side gig almost within same entities and members and co founders, but it was like psychic. So internally, we had a much more sophisticated data, structure, database architecture, all those things. For the consumer business. cember was more of a hacky product that we've kind of did on the hackathon positive test case, or selling a sideline. So we had to really rebuild our architecture multiple, multiple times throughout the past six and a half years. Now, I think we're pretty good stay, we have a data team that, you know, use nickels BigQuery, you know, five times all those things. And then we use like Looker for the dashboard side of things. But initially, it was It started out as a purely a database. And what we call the Insight is internal tool was like operating tools, a little bit of data. And then obviously, there's always that like business guy asking the engineers like, Hey, can you like do a database quarry and palisade owl, and you have to wait like, you know, keep harassing them, basically, to give us a data or wait like 24 hours, and then copy paste it into spreadsheets, of course, spreadsheets, runs out of memory, stuff like that. So there's a whole whole bunch of messy stuff that we had to go through. Now, obviously, at our scale, that does not work. So we have a data warehouse project, then instead of running, and a lot of them, obviously are connected to different parts of the system wouldn't be a Salesforce looker. And, of course, sometimes there are some manual spreadsheets. So it's still to this day is kind of messy. I mean, obviously, we have gotten way better at it compared to six years ago. But still, to this day, there's a lot of manual work, a lot of kind of collaboration that needs to happen. Some missing data that we know is in the logs that we have to then pull it out and connect it to more of a sophisticated system like Looker, etc. So that took good years, and we're still working through it. But it was never like a straight line. Because he kind of accidentally stumbled onto this timber as an idea. It just might actually be my first time doing b2b business. So what do I know?
And real quick to like, what was the what was the moment that you knew? You know, because it kind of reminds me of like the slack story, right, with a game in the chat on the side. Like, when was the moment that you knew that sendbird should be the thing that you focus on?
Yeah, I think there are a couple of milestones or episodes that I remember right one. For our mom's application. We had like zero revenue for two and a half years. We had users, right, we had like a quarter million users, but we had like zero revenue. Now because we're running out of money, we had to figure something out. My only way to convince our investors in our own team was like, hey, there's something here and how do I show that is by having either sign agreement, something to do in the pipeline or actually have revenue, right. So I started like pitching random pricing point, because I've never done b2b before. I started pitching random pricing points to my friends where we also tried to like build chat and we started generating revenue before even we had a product or who started signing deals, right? Okay, if you give me this SDK by then I'll pay you this amount. I'm like, holy cow, we never seen dollar coming to our bank account for 20 years, now we're starting to see dollars, or at least committed dollars, before we even haven't had a proper product. So I came back, like talk to our teams, like, Hey, we got to build this now. So that was kind of like, oh, this is like, kind of how customer development process by Steve Blank or you know, the Agile sort of Sir built. This is cool. So we kind of did that. And then once we got into YC, there are moments when you feel like there are prospects or customers that come in down, that you clearly know that you do not deserve. Right. They're always this customer that you never heard of, but you look on CrunchBase, they waste hundreds, hundreds of millions of dollars, wait a second, we recycle million bucks, we clearly cannot take care of their traffic nor their use case i, we don't even have a proper office. And yet we jump on a call with them. They know that we are a tiny little startup, but they're willing to take that bed with us because it's so critical that they get this with a builder with a team that's like super dedicated to this. So once you there's a constant, like a step function increase in terms of like deal sizes, but also traffic also, like when we signed with Reddit, I think Reddit was probably about 200 250 employees back then. We're like 2515 25. So and they literally helped us walk through what is needed to sell to large companies like Oh, you gotta do penetration tests, and you gotta have compliance. So it's like, oh, like, basically, they walked us through. And they were willing to sign a deal and says, We'll sign the deal. And it's almost like after the fact, like, within X number of days, you will do this and this. And this. I'm like, so like, normally, you've asked all those things to be done in advance of signing a deal. But they're like, we'll sign this. And there was like, almost like a question that came system. And they also paid us annual upfront, I'm like, This is amazing. You paid us upfront, non diluted capital. So basically getting seed seed check with no dilution, plus, they gave us a clear path to success. So that's kind of when you start to realize, okay, there might be a clear need to, you know, building a solution like this. So, I think we had a couple of things. Instances.
Yeah, so it's, uh, you know, you're solving a real problem with real design partners, if they're laying out the process for you. That's It's so cool.
So sore, so forever grateful to Reddit guys. I'm still our loyal customer. Yep.
So, so awesome. Switching gears here now a little bit, you know, back to sendbird. And she talked about some of the data challenges that you've had with how massive the datasets are, and how real time it needs to be. How would you describe the data culture today at sendbird? And how has this changed over the years?
Yeah, um, to be honest, we're still trying to figure this out. I mean, yesterday is always important. And we have no shortage of data. But I think it's getting the data into the hands of the right people who need them at the right time, right? You don't want a data that's like too old that you requested, like, two months ago, you need things in a timely manner. So just like if you think about, like, you know, some people like engineers are perfectly fine using JIRA. Right. But if you give JIRA to a lot of non technical folks, they're like, Oh, this is too messy. Can I just use like Trello or Asana? So there are tools that are more accessible, that comes with compromises or like, you know, some people may prefer notion over Confluence for several reasons. So how do we give more data access to kind of non tech folks is always a challenge. I think there are a portion that we've done well, like you're getting the data through, whether it be Looker or Salesforce or Keysight. So different infrastructure in ingest different kinds of data. That is kind of usable to non tech folks. But still, to this day, we're still building it and building data, I think we got our data warehouse up and running, like probably four or 5/5 period. So it's fairly recent. And again, because our growth can kind of came from organic kind of accidentally, but if I could go back in time, I certainly started differently. Assuming that we had intentionally tried to build Scentbird from from the beginning, we probably because my previous background was in social gaming, right? social gaming is all about data, understanding econometrics, you know, user acquisition, you know, potential monetization, the the AR framework,
then getting organic growth, right.
Yeah, it's all about data optimization and everything, but we didn't have that with sendbird. In the beginning, now we kind of we kind of deal. So I would probably put the foundation upfront. And then so that we could say years and years of pain. Now Now today, we have like realtime database that kind of connects to Okay, our system the shows are archived right path, you have different ways to measure it, we have a weekly basis of, you know, looking at Pipeline, looking at OKRs, looking at different kind of metrics, product matrix, uptime data, all those things, right. So it's now gotten to whoever's permission, all the data we have is can be processed on real time, if not, at least on a weekly basis. But it took us not trivial amount of effort, I was still trying to figure it, how to make it more streamlined. Totally,
no, and that's, I think that's something that we see all the time. It's, it's totally circumstantial to the business of when you choose to build out the data stack of when you choose to hire a data team. And those things can can vary so greatly from company to company. I guess just to add, connect back
to my original story of working for this company called MC soft when I was working in the business unit building this operating foolery, a lot of people's manual work was about getting data either getting from the the I guess a data team from consensus or Python was like 2000 3000 employee company. So working through the engineering team to get the right data. And then also internally, the the business users always had to like, put in manual work putting together like spreadsheets back then the goal sheet or online, especially wasn't a thing. So everything was done offline. And they you'd upload to the central repository of Basecamp, Outlook server. And then one person who takes a turn would download all the spreadsheet, copy pasted. And then somebody finally read something about using macros. So automatic automatically copy and paste into certain cells, which should then render into your graphs and things like that. It was a it was a massive, massive waste of human energy and effort. Now, one person accidentally deletes the entire outlook server by hitting the delete button. So then they would have to go back to the Data, Data Team say, Hey, can you restore the backup from like, couple days ago, they're like, Oh, we don't have it. They use like two days of data, which is like, horrible. So that's why I kind of like, okay, I want to build this automated like data tool where people can punch in, you don't lose the data is automatically doing the daily rotation of backups, plus the graphing all those things can be done in real time, because web application, basically. And I saw this opportunity, because like, non technical folks, it's hard for them to understand what's possible and what's not possible, what is easier, what's not easy. And then you can't just expect everyone to learn how to program right? So how do we kind of bridge that gap between data and turning that into insights that can ultimately turn to a different actions, right. So I think there's a huge opportunity or tests still not being resolved, even through companies like Tableau and Looker did a fairly complex thing to use. We have multiple session internally training look ml, like trust me, like no business folks is going to learn look, ml, maybe a couple. But most people like, just give me the link to the dashboard. They will go and look at it and ask questions. So it's been hard.
Right, you know, and that's, that's really the problem that we're trying to solve a canvas here, right? It's bridging the gap between data and business teams and giving them a way to, you know, answer questions beyond the dashboard and use the skills that they're familiar with. So totally makes sense. Talking about talking about those metrics from the dashboards and OKRs, that you mentioned, what are what are some of the Northstar metrics for sendbird? Right now? Have they been consistently those Northstar metrics? And how do you manage how you'd like manage that process for those metrics?
Yep, so we switched over to OKRs, quite a while ago, maybe, or maybe 30, or in, so we kind of build an internal muscle, but we also change towards the here and there. But there are a couple of metrics that we track. The North Northstar metric being the number of engaged users when we say engaged, it's like number of people that read messages who sent messages. So it's a it's an active action that people will take. Now some passively connecting to server but actually measure daily engaged users as well as some of the engaged users. Then we have more of a high level metrics, like monthly active users, number of message being sent, number of live applications, customers accounts, they use custom accounts may have multiple applications. And then we also have a centralized dashboard, which shows this product usage metrics, but also like AR metrics, as well as paying customers and then employees by region by role. So we have this a centralized dashboard as as the snapshot of our entire business. So that's kind of what we track. But from the Northstar metric perspective, we care about the engaged user count that communicates through our platform in a given day in a given month.
And what are some of those milestones? Or what are some of the things that you look at for engage users to know? Okay, this is the right time to you know, to have the conversation for upsell or cross sell what are some of those key things that you look at?
Yeah, right now are billing metrics is around Monday, Monday I focus most of the billing cycles are the monthly basis. Some customer quarterly, some customers are annual based on the terms but so our CSM team has the basically the user insight, our internal tools that have evolved over time plus a key site, a couple of like dashboards to look at from product usage perspective, sentiment, tickets, SLA, all those things in a centralized view, which comes out as a customer like the score, right engagement score. So based on the score, we know how happy they are. And then based on the usage metrics, we'll know are they hitting their limits are not under whatever we agree on the agreement or on the plan subscription plan. If they start to hit or near the near the limit or hit start hitting overages, that's when we know like, Hey, we've been singer growth projects or or you've been eating or overage, then you know, it's kind of the right time to do the upsell. And also customers that are large enough. Usually we have a dedicated CSM team and social engineer versus myself. So a customer may not work through CSMs. And they may just work through a normal self service of grace or ticketing support.
Awesome. Yeah, that makes makes sense in that consumption model, you know, makes it really, really easy to know when the timing is right and be pretty transparent with your with your customers. Jeff Cool. Well, I know you're I know you're a busy guy, John. So we'll let you run. And thank you again so much for coming on the show. But before before you take off, where can people learn more about you? And sendbird?
Yeah, well come december.com. And it's pretty. It's a carrier pigeon, you're sending a birth to send a message. And we also have a new product that we just released called the Unified inbox, which is basically bringing together your marketing sales, alert transactional messaging, as well as user to user all into a single user user experience, or recon, the unified inbox is a better better launch. So we're very excited about that. Because we do think this future of user experience, because today the communication is fragmented, but also want to encourage everyone to check out chemists. Because I just a problem that I personally struggled with a lot throughout my entire career was how do you make the data as more readily consumable by every everyday users like the people with the business side, so that we can turn those into actionable insights that ultimately turn into conversions and better user engagement, retention. And it's still, it's a problem that we as assembler are trying to solve at scale as well. So I'm sure a lot of startups and companies in the world will benefit from it. And also the extra benefits of being able to collaborate, right? So I love those aspects. And I do think there's a future it's like figma, there's a figma for design. There should be something like that for data. So thank you for having me. pretty psyched to see you guys grow. And then yeah, let me know how it can be helpful to everyone in the audience.
Awesome. You did the did the pitch well for us. Well, thank you so much again, John, and we'll talk soon.
