How Coda Reached Tens of Thousands of Users and $60M Raised Pre-Revenue (Interview with Co-Founder and CEO Shishir Mehrotra)

Introduction and Company Overview

Nathan Latka

00:02Hello, everyone. My guest today is Shishir Mehrotra. He's the CEO and co founder of a company called Coda. He was formerly an executive at YouTube where he helped grow the platform of the world's largest video destination, helping lead operations there. He also held leadership roles at Microsoft and served as founding CEO of Centrado, which was acquired by Launcher. He's also an MIT graduate. Shishir, you're ready to take us to the top?

Shishir Mehrotra

00:21>> Yeah. Sounds good.

What Coda Does and the Alpha Launch

Nathan Latka

00:22Alright. So you launched this thing, I think, in 2014. You're still pre revenue today, but you're learning a lot based off kind of your alpha and beta launches. Help us understand what the company does and what what you've learned from the alpha launch.

Shishir Mehrotra

00:34>> Sure. So we we started the company in 2014. The company we built a new type of document. It's an all in one document that blends the best of documents, spreadsheets, presentations, applications all into one new surface. The thesis is that anybody can build a doc as powerful as an app. We ran the company a little bit nontraditionally, and and we're in quiet stealth for about three years building up the product with a small number, a few

00:55>> dozen alpha customers. We launched the beta in, late twenty seventeen. That grew quickly into about a hun hundreds of companies, and we launched the 1.0 this past February, about five months ago. We're now tens of thousands of users, thousands of companies around the world. We'll launch pricing in the next couple months.

How Many Companies Are Using Coda

Nathan Latka

01:13Okay. So sorry. How many companies are on you now? About a thousand?

Shishir Mehrotra

01:17>> Thousands. Mid thousands.

Nathan Latka

01:18Okay. So call it maybe four or 5,000, something like that. And where did you tactically, how did you sign up these folks? Going all the way back to 2014, what channels did you use?

Growth Channels: Word of Mouth, Product Hunt, Reddit

Shishir Mehrotra

01:27>> Yeah. Most of it was word-of-mouth. I mean, there was a handful of things that worked well in terms of communities we dropped into. The Product Hunt community is probably one of the best for for Coda. It tends to tends to draw a lot of people. A lot happened on Reddit and various forums. A lot of the demand for Coda tends to come through, sort of two different mechanisms. One one is people that I think of as

01:48>> sort of tool seekers that are coming after Coda as a new type of document purely. But the other type tend generally come after a set of use cases, and so they tend to they tend to arrive first in our template gallery. I kinda think of it as front doors and side doors. And in those cases, we tend to be relevant in communities of particular types. Know, we show up in community forums for, you know, people trying

02:10>> to track fitness or people trying to track project management or people trying to run better meetings or so on.

Product Hunt Launch Results in 2017

Nathan Latka

02:14So I wanna actually quantify some of this, because my team did some research on the distribution channel. So back in October 2017, you launched on Product Hunt. It got about 5,000 upvotes. Quantify what that meant for the business. So how many new signups did you see that day?

Shishir Mehrotra

02:29>> Let's see. So, yeah, 5,000 upvotes is a pretty it was pretty good first launch. Tens of thousands of sign ups that day or that week, I should say.

Nathan Latka

02:39Yep. And these were all individual seats, or did you see a lot of kind of at Google and at Uber and at, you know you know, Airbnb signing up?

Shishir Mehrotra

02:46>> You mean as opposed to as opposed to Gmail?

Nathan Latka

02:48Yeah. As opposed to individual users using it, like, personally for fitness.

Shishir Mehrotra

02:53>> We let's see. Early on and back then mean, it's probably about the same percentage. We're about 80% businesses, about 20% individuals and consumers.

Funding: $60M Raised Across Two Rounds

Nathan Latka

03:03Okay. Now anyone listening to this is gonna go, wow. This seems like they've done a great job signing people up. It's is this like you see betting on a freemium plan, and and also how the heck has he supported himself for the past four or five years with no revenue? So you've either raised a ton of money or you've sunk a bunch of your own capital in. Which one is it?

Shishir Mehrotra

03:18>> We raised a bunch of money. So we raised about $60,000,000 across a couple rounds financing.

Nathan Latka

03:23Six or six zero?

Shishir Mehrotra

03:24>> Six zero.

Nathan Latka

03:25Okay. So 60,000,000 raised. Where most of that capital, where has it been deployed?

Shishir Mehrotra

03:30>> Well, most of it's in the bank, but the the capital's mostly been deployed to engineering. I mean, we were we were probably 80% engineers for until

03:40>> probably in the last year.

Nathan Latka

03:42Mhmm. Okay. So so what's the team size look like today?

Shishir Mehrotra

03:45>> We're about 50 people.

Nathan Latka

03:46All in there in San Mateo?

Shishir Mehrotra

03:49>> No. No. We're Mountain View, San Francisco, and Seattle.

Nathan Latka

03:52Okay. Very cool.

Shishir Mehrotra

03:52>> And then and then we have a few people spread out, Florida, New York, so on.

Freemium Model and Pricing Philosophy

Nathan Latka

03:55So, you know, I guess, is it fair to say that you're betting hard on a freemium model? You wanna get up to kind of millions of users quickly and then try and go four or five, six bucks kind of a seat per month, something like that?

Shishir Mehrotra

04:07>> Yeah. So it's definitely a freemium model. I mean, I think products like this, new documents, so on, they they they spread quickly, and so you want it you want as few frictions as possible in that spread process. And as they stick, people tend to find more and more important use cases and upgrade through it. In terms of the pricing model, we're we're going through that now with all the different companies that are using us, and, you

04:27>> know, we'll roll it out in the next couple months. But the the current thinking is that it's a creator centric model so that the the you don't pay for many parts of the space pay for all types of users. Our current plan is that we charge for creators and not for editors. That incents a certain type of behavior of allowing people to share more freely. It might mean the per seat price of the creator is a

04:48>> little bit higher, but when blended together with all the editors, it tends to work out to to pretty similar price.

Nathan Latka

04:54So a team of a 100 might use you, but maybe there's only three people creating docs and the rest are actually editing. You'll charge for the three that are creating.

Shishir Mehrotra

05:01>> Yeah. So the ratio tends to be eight to one. 10 to one tends to be the the general creator to contributor ratio, and that's actually something that's part of the Coda thesis from the start. Was before Coda, spent a while at Google, before I spent a bunch of time on Microsoft working on Office. That ratio of maker to user is a is a ratio we've seen across the Office suite as well. Mhmm. The the way those

05:24>> products are used to.

First Round of Fundraising and Pitch Deck

Nathan Latka

05:25And then so take me back again to kinda some of the earlier days here. So your first round of you leave the company, you leave YouTube, you jump directly into this. Your first round of capital, how much did you raise? And the follow-up question on that is, was it simply you pointing to your historical resume at YouTube that allowed you to raise at a valuation that you weren't getting I mean, you have no idea what the

05:46valuation is going be that early. How do you even think about that?

Shishir Mehrotra

05:49>> Yeah. I mean, I would say our fundraising process was pretty different than most. I mean, this is my second company I've started. My first one was twenty years ago, and that one, it took us, you know, six years or sorry, six months to raise $500,000. And this one, you know, managed to raise. The first round was about 25,000,000, and it was done basically over a weekend. And, you know, I think there's probably a couple different things

06:08>> that went into it. Certainly, the quality of the team was was high on the list, as myself, my cofounder Alex, and I both have pretty good track records, both with startups as well as with with larger organizations. And then it was the market and the space and and our approach to it. And at the time, hadn't really built anything yet, but we had a pretty good idea of what we wanted to build. The pitch deck was

06:29>> basically 69 slides of of of product mocks. And it basically had the the the view of what we were gonna end up building back then.

Nathan Latka

06:38Is that have you published that anywhere?

Shishir Mehrotra

06:41>> I haven't.

Nathan Latka

06:41No. Come on. You you should release that now. That's a really interesting story.

Shishir Mehrotra

06:45>> Yeah. At some point. At some point, will.

Nathan Latka

06:46Is there something sensitive in there? Right? It's just early it's early UI stuff. Right?

Shishir Mehrotra

06:51>> Is there anything sensitive? I have to go back and look at it. It's the it it I would say it's a it's a road map that we're still following. So it's a I go through it with the with new employees, and I think, you know, as many companies that that start, and you end up doing something very different than when we started, this one is pretty true to our original path. We're still basically executing on that

07:07>> plan. So it does include a lot of stuff we haven't shipped yet, but other than that, it's pretty close to what we built.

Nathan Latka

07:12So 25,000,000 raised kind of on day one. Now when I look up your the actual docs you filed from a government perspective, it all looks like it's in one round. So did you basically set terms and then let it roll, and you solely added more on the 25 on the same terms going up to 60 over the past couple years?

Shishir Mehrotra

07:26>> No. No. We we raised the second round about a year later. That was done. So the first round was led by Greylock and General Catalyst, and then we we raised another round about a year later led by, Khosla and NEA.

Nathan Latka

07:39How much on that one?

Shishir Mehrotra

07:40>> About 35,000,000. That's the rest of it.

Nathan Latka

07:42Yeah. Okay. Oh, got it.

Shishir Mehrotra

07:43>> In both cases, I I squeezed in about 15 angels, friends of mine, and so on that that were interested in joining the company.

Nathan Latka

07:49So so what do you I mean, you were at YouTube, so you get this. Right? What do you have to scale to okay. This is gonna sound like a crazy question. It's such a San Francisco question, but I'll ask it because you're the perfect guy to ask. Okay. Assume you never have any revenue. Right? Let's assume you're gonna go public with no revenue. Right? How big do the numbers have to get? How strong does the viral

08:09coefficient have to be for analysts, stupid Wall Street analysts to actually understand the future potential revenue opportunities?

Shishir Mehrotra

08:17>> I I well, that's not our plan. Okay. We're we're gonna launch pricing. So I mean, I can answer a theoretical question. And I I think I mean, I think SaaS businesses and YouTube are pretty different. Mean, I think I think the I mean, YouTube was bought for $1,600,000,000 I showed up, you know, ten months after the acquisition. We were doing tens of millions in revenue, basically big sponsorship deals. It was basically nothing compared to what the

08:41>> scope of the business was. It took us about three years to get to a billion dollar revenue run rate. And, you know, that process, like, at the time was seen as as not obvious in the consumer space that you could take you could take a scale of what we were probably tens of millions of daily active users at the time, and, you know, turning that into a billion dollars in in revenue was was something that hadn't

09:05>> been done many times before, but but had been done at least a couple times. Nowadays, I think in the consumer space, that's seen as somewhat obvious, that if you can get yourself to tens of millions of daily active users, then then you've got a pretty meaningful then you can you can build a pretty meaningful business out of it. The SaaS space is different, though, because the the SaaS space, not not not having pricing actually stunts growth.

09:26>> I mean, it's definitely you know, we launched in in February, and it's certainly one of our top customer complaints is we wanna know what it's gonna cost us. So I don't I don't think it's really that comparable. So I'm not sure I would trust a company that got to great SaaS traction without pricing for for very long.

Activation, Retention, and Daily Active Users

Nathan Latka

09:46So what what currently what are your do do you care more do you look more on a daily basis, or do you look less frequently? Do you measure monthly active users? And if so, what is the act like, what's the activity that says, yes, they're active?

Shishir Mehrotra

09:58>> Yeah. I mean, we look daily. I I I have a whole this is actually similar both consumer world and business world. I I think monthly active users are really easy to they're just a very they're easy fluctuate metric where you can take very light hearted users and and end up counting them. Yeah.

Nathan Latka

10:12Yeah. Like, they logged they logged on once, but, like, did nothing.

Shishir Mehrotra

10:15>> Yeah. So, I mean, our our basic view is we we think of users as a pretty typical growth funnel, but we our our activation period, we give people two weeks of what we think of as our trial period, whereas they're getting started, they they tend to either stick or leave within that two week period.

Nathan Latka

10:31Do you have to go and do, though, in that first two weeks to drastically increase stickiness?

Shishir Mehrotra

10:35>> I mean, the best signal of it is sharing a doc. And so the moment they share a doc with someone else, that's generally

Nathan Latka

10:43Which means they to create it too. Right?

Shishir Mehrotra

10:44>> Which means they have to do all the steps before that to create it, make it theirs, and so on, but sharing is really that moment. There are people that start with share. It's actually quite interesting that that a a product like ours will say, hey. I'm checking out Coda, and they'll share it with a with a teammate. And that tends to be just as effective. So whether they've created a lot of it before or not, that

11:01>> share moment seems to be it's and, you know, thinking about it in human terms, it's your moment of endorsement. It's your moment of, okay, I think enough of this that I wanna try it out with with my colleagues. And then after that period, after that two week period, we we tend to see people either embark or not embark in in the in the product. At that point, our frequency tends to be pretty high. It's about

11:24>> it's 80 plus percent of our users are active three or more days a week.

Nathan Latka

11:27And what does active mean?

Shishir Mehrotra

11:29>> They've opened the docs. So the the the we I tend to so, actually, maybe backing up a moment. I tend to think of metrics, try to define them as simply as possible, because I find that if you put a lot of nuance in it, then you'll spend all your time explaining the nuance. So our to be a daily active user on Coda, just have to open a document on a on a given day. We then layer

11:48>> on, did they take an edit action? Did they do take a what we call a maker action or just viewer action. Those are sort of three categories of actions we look at. And so on a given day so 80 plus percent of our users are active, meaning they open the doc at least three out of seven days in a week, which for a product that tends to be mostly during the work week, that's I I think

12:10>> basically means we're we're used through through the week. So so that that sort of tends to be the the second phase. And then third phase we look at is how it spreads, and how did those set of makers turn into bring on their teammates and contributors, and at what rate did they then become makers themselves and create their own docs. And so those are sort of three phases of how we think about the Teams' And right

Nathan Latka

12:33so what is today, daily active users?

Shishir Mehrotra

12:36>> It's tens of thousands of users.

Nathan Latka

12:37Okay, 10,000. When do you think you break a 100,000? 100,000 maybe next year?

Shishir Mehrotra

12:42>> Okay, I think maybe early next year.

Nathan Latka

12:45What is that, I guess, that to number of docs created over the past, call it, thirty days on the platform.

Shishir Mehrotra

12:54>> Docs created is thousands as well. Okay. So it tends to be and actually, maybe to to back into that a little bit so you can understand the dynamics. And I think this is I think for your listeners it's interesting because Coda is sort of two different things. One is it's a product that, you know, competes as a document, and people tend to use it a lot like you might use, you know, Google Docs or Office or

13:17>> so on. But the things people build in Coda tend to feel like applications. Many of the they By

Nathan Latka

13:21the way, you feel way more to me. Like, I I I moved from Google Docs or processes. Actually, now I use Airtable plus Zapier, and I can get a lot done. And it makes it feel like English that I'm typing in Airtable is actually codified and actually feels like an app when it all works together. I feel like what you've built is you've essentially been you've taken it out of Airtable and it's more like actually more

13:42like a word doc, but it works with all the back end kind of coding parameters.

Shishir Mehrotra

13:45>> Yeah. That's right. So, I mean, I think and and to use that analogy and go a step further, many of the use cases that people describe their Coda docs to us, they tend to think of them like applications. So people use Coda and say, okay. I've rebuilt my CRM system or I've rebuilt my my, feedback tracker or task tracking system or so on. And so the way the product spreads tends to mimic these two different behaviors,

14:07>> how docs spread and how apps spread. And and so you you depending on the use case, you will end up basically seeing a similar spread, like, depending on the app use case, we'll see a spread pattern that's very similar to whatever that that category of apps is like. So if you end up picking up Coda as a CRM tool, we'll spread like a CRM tool does. If you end up picking up Coda as your way to

14:29>> track meeting notes, we'll end up will end up scaling and spreading a lot like how you Evernote. Like Evernote. Right? Or or actually, probably meeting notes probably closer to, like, Google Docs is probably closer. Yeah.

Nathan Latka

14:39So so over the past thirty days, how many kind of apps or docs have been created, would you say? Thousands?

Shishir Mehrotra

14:44>> Oh, thousands. Yeah.

Nathan Latka

14:44Yeah. But less than 10,000.

Shishir Mehrotra

14:46>> Yeah. Less than 10,000.

Nathan Latka

14:47Yeah. Yeah. Okay. So it's a little less than a one to one ratio between kind of active users and apps created over the past thirty days.

Shishir Mehrotra

14:53>> Yeah. It tends to be pretty close.

Nathan Latka

14:55Yeah. Okay. Very cool. Yeah. Very good stuff. What about churn? So like, how do you measure if people are getting more or less active and kind of the equivalent if you had revenue would be expansion revenue, right? Or contraction revenue, but for you it's kind of expansion usage or contraction usage.

Viral Spread Inside Companies

Shishir Mehrotra

15:09>> Yeah, that's right. Right. I mean, I think all those same metrics sort of apply. You know, the way we tend to spread, there's sort of two different parts there. One, how we spread inside a company tends to have a very high viral spread rate. Basically, group of just to to to think of it somewhat specifically, you know, one one maker picks it up. They, on average, end up sharing with 10 people. Of those 10 people, something

15:34>> like three of them end up being makers themselves. They share with another 10 people, basically until you run out of people in the company.

Nathan Latka

15:39Then Those are your actual metrics about?

Shishir Mehrotra

15:42>> Approximately. Yeah. And in the good cases. So as long as that first person is successful. And so if the first person's not successful, then, you know, that that tends to you know, that doesn't happen at all. But when we're successful, you know, we spread very quickly. And so inside of small companies, you know, have a long list of companies that run sort of everything runs on Coda. You know, in larger companies, you know, we'll end up

16:02>> growing to hundreds or thousands of users as they start to see that spread process. And then the the flip side of that is the churn rate, and as people churn out of it. We tend to find the churn is very low after those first two weeks. So in the first two weeks, it can be, you know, it can be eighty percent of people. But after that, it tends to be very small on a weekly basis, you

16:24>> know, less than a percent, but and over the course of a year, know, most of these tend to stay.

Nathan Latka

16:30A little less than ignoring those first two weeks over the over the first year, you'd say churn is what? Like 30%? Something like that?

Shishir Mehrotra

16:38>> Yeah, in that ballpark.

Why Pricing Has Not Launched Yet

Nathan Latka

16:39Okay, that'd be like one or 2% a month after the activation period is kind so I mean, these activation metrics basically mimic SaaS revenue retention and expansion cohorts as well, right? Basically the same. So what's the biggest risk for you? Like, why haven't you pulled the pricing wall trigger yet? There must be you know, like, you're doing some extra planning, you're doing some extra research. I mean, why not just stick it up, see what happens, and

16:58then edit quick?

Shishir Mehrotra

17:00>> Oh, no. I I think we're the for getting pricing out, the big thing for us is we're we're gonna introduce a new concept. We have to we have to introduce a pricing model, and it's gonna be priced around workspaces, which is pretty typical for products in this space. Mhmm. The product doesn't have that yet, and so we're we're gonna introduce workspaces, which is actually a pretty heavily requested feature anyway, and then you'll price on the the

17:20>> workspace itself. Be a paid workspace or free workspace. So not no no no philosophy behind it. Just time.

Nathan Latka

17:25Wouldn't it be going out with Coda Mobile, Coda Automation, and Coda Packs, all things important to you, because you did separate product launches for all of them.

Shishir Mehrotra

17:33>> Yes. Yeah. I mean, I think each of those basically, the way to think about our process up till now is we have this thesis that you can build a doc as powerful as an app, and we wanted to launch all the main building blocks of what it takes to be able to build up these apps. And it's sort of surprising to people to take a thing that, like you described, starts with a blinking cursor, looks like

17:49>> you can just write words, and then you gradually can add all these different things. You can add an automation layer that allows you to automate all your common activities. What what we call Coda Packs is our our integration layer. So you can from Coda, you can integrate with everything from Slack to Gmail to Twilio, and you can go, you know, pull in pull in thing, pull in your Intercom data or your Salesforce data, and then do

18:08>> things like automate how you send messages out to people, notifying your team when things change, and so on. And so in that process, people end up building these docs that start really small and then grow from there. A lot of people like to describe it as built in scope creep. Yeah. Yeah.

Paid Marketing and Burn Rate

Nathan Latka

18:22We're we're we're running here since you're out of time, but last two questions here. How aggressive are you being to get new, kind of that one new user? Will you spend a dollar, $2, $3? What's that look like?

Shishir Mehrotra

18:34>> We haven't done any paid marketing yet, so we're just starting to test where that where that stands. My my goal was, you know, we're running a few tests in the next few weeks just to get the mechanics up and running, but I wanna launch pricing before we get too aggressive on that.

Nathan Latka

18:47Yeah. Yeah. That makes good sense. And then talk to me about burn. I mean, obviously, you're burning your pre revenue. How aggressive are you being on that? Are we talking $500,000 a month, 1,000,000 a month, something less?

Shishir Mehrotra

18:57>> You know, it's 50 or so employees, so you can, you know, do the math. It's about about 1,000,000 a month. Okay. That's what ends, I think.

Future Fundraising Plans

Nathan Latka

19:02Yeah. That's fine. I I guess, the reason I'm asking you is do you have another raise in your future, or are you comfortable with the amount of runway you have in your bank right now?

Shishir Mehrotra

19:09>> Oh, we'll probably end up raising again. I mean, I I think this is a space where as I've been listening to a bunch of your podcasts, it's a space where you not only have to raise to cover burn, but your acquisition cost for customers also goes up as a Good. As you Are you enjoying the show? Oh, yeah.

Nathan Latka

19:22It's great. Anything you wish I asked that I and I always don't?

Shishir Mehrotra

19:25>> Oh, that's a good question.

Nathan Latka

19:27Well, email me if you think about it because I'm always

Shishir Mehrotra

19:28>> I'll looking think about it. I'll think about it.

Nathan Latka

19:30You know I have a pattern, and sometimes I deviate if something interests me. But generally speaking, you know what I you know what I'm after.

Shishir Mehrotra

19:34>> I think it's a super valuable show. It's been interesting interesting listening listening to to people open up about how they're thinking about their business.

Famous Five Rapid Fire Questions

Nathan Latka

19:40Well, and thank you for being transparent. On that note, let's wrap up here with the famous five. Number one, what's your favorite business book?

Shishir Mehrotra

19:45>> Switch by the Heath Brothers.

Nathan Latka

19:47Number two, is there a CEO you're following or studying?

Shishir Mehrotra

19:50>> Daniel Lacks from Spotify.

Nathan Latka

19:51Number three, what's your favorite online tool for building your company?

Shishir Mehrotra

19:55>> Well, we run most of Coda on Coda. So but other than that, I'd say Figma, probably. It's like Google Docs for design. Turns out we were their first customer as well. Oh, wow. We have a pretty good Coda Pack for Figma, so it integrates well as well.

Nathan Latka

20:08Number four. How many hours of sleep you get every night?

Shishir Mehrotra

20:11>> Five to six.

Nathan Latka

20:12And situation, married, single kids?

Shishir Mehrotra

20:14>> Married, two daughters, 10 and 13.

Nathan Latka

20:16That's great. How old are you?

Shishir Mehrotra

20:17>> I'm 40.

Nathan Latka

20:18Last question. What do you wish your 20 year old self knew?

Shishir Mehrotra

20:21>> How important those early deep relationships can be. Had my if I went back to my 20 year old self, I was dating the woman who I ended up marrying, and I just become friends with the person I ended up starting Coda with.

Nathan Latka

20:34Guys, those of you in San Francisco will look at $60,000,000 raised pre revenue and go, what the heck is going on? But Shishir is building a is an idea in a big space, which is essentially instead of, you know, Airtable makes marketers powerful because they don't need developers to build systems. Now he's basically saying you can do all this in document, build applications right inside a doc. They've had successful product launches, tens of thousands of active

20:54users now about to turn on pricing. We'll see what happens. They're currently burning, call it, a million dollars a month with a team of 50 as they look to scale and, again, launch that first paywall. Shishir, thank you for taking us to the top.

Shishir Mehrotra

21:04>> Thanks, Nathan.