I built a virtual office for distributed teams – Vivek Nair, Pragli

Show Notes

Vivek started building his first startup, Stacksware, while at Stanford and had already raised $2 million from the likes of Lightspeed Ventures and Greylock Partners before graduation. Raising capital while in college was a huge validation for the team and almost surreal.

Two years later, Vivek and his co-founders decided to exit and sold Stacksware to Avi Networks. Interestingly, the Stacksware team, by this point, wasn’t able to grow at the pace VC funded companies are required to and this heavily influenced the decision.

Vivek’s journey as an entrepreneur and a maker has been quite interesting. Post Stacksware, he worked for a couple of years remotely, basis which he derived inspiration for his current product, Pragli.

Currently only a two member team (Vivek and Doug, both engineers), Vivek speaks about what problem(s) he plans to solve with Pragli, how he plans to reach the target audience of remote teams and also, the biggest challenges he is facing with Pragli right now.

Tune in to listen to the full episode here.


[Hrishikesh: 00:00] Hello everyone. This is your host, Hrishikesh and welcome to the makers podcast by Remote Tools. Each episode we chat with entrepreneurs and indie-makers who are building products for the remote working community. This podcast is powered by Flexiple, a network of top freelance developers and designers. We have a very special guest with us today. He started his own venture immediately after graduating from Stanford and got acquired in just a couple of years. Having worked in the remote setting for over two years, he realized the need for a virtual office for distributed teams and started building Pragli. Hey Vivek, welcome to the show.

[Vivek: 00:47] Thank you so much for having me Hrishikesh.

[Hrishikesh: 00:49] So Vivek, being an entrepreneur, you would have worked in a variety of functions, what do you like doing the most? Is it engineering, marketing, sales, writing?

[Vivek: 00:58] [laughing] Yeah, that’s a great question. So for me, I’m an engineer by trade and I love it so much and I would also sort of characterize myself as an introvert. So, although, you know, this sort of function of a start-up requires that you actually wear multiple hats. I love engineering. It’s what I do when I’m looking to create a new project or whatever else. And you know, that’s essentially my role now and what I intended to sort of transition to, as we sort of scale out with Pragli.

[Hrishikesh: 01:26] Super! Well, so let’s just quickly go to the very beginning. Tell us a little about your time at Stanford and how it all began.

[Vivek: 01:33] Yeah, absolutely. So, you know, just a little bit of context about myself. I was raised by Indian parents and I’m an Indian American. I was introduced to computer science when I was young and that’s sort of how we got into it. Basically, that sort of love for computer science and engineering in general kind of led me down the path to Stanford. When I moved up to Stanford and still wasn’t necessarily sure whether I wanted to pursue computer science or software engineering yet, but there’s just so many brilliant minds there and just all sort of gravitating towards the subject, and all my friends who are really interested in it. And then it would lend itself to actually just wanting to make side projects and all of that. So, I quickly found myself gravitating towards computer science and then, yeah, stuck to all the classes for all four years. And in my fourth year, we actually decided to make Stacksware.

[Hrishikesh: 02:41] Wonderful. And this was a research project at Stanford, right?

[Vivek: 02:47] Yeah, yeah It had a funny origin. So, we were three co-founders and all good friends for the last few years. Basically, we came into the project saying, hey, you know, this is a great opportunity to structure our ideas. We incubated a company and thought we will get as many resources from our corporate sponsor, VMware at the time, get as many customer introductions as possible from them. We sort of really liked that component of it and then, basically, there were set of prompts that each group inside of the class had to choose and we just chose the one that was most flexible. That would allow us the most opportunities to sort of pivot from idea to idea.

So, yeah, so it was a two quarter long research course, about building a company and pushing it out to customers. So, yes, CS210 with Jay Borenstein. And what’s interesting is that Jay was a venture partner at what we now know as Lightspeed venture partners which is a pretty prominent firm right now. 

And when he saw that we were getting some traction and a lot of customers are interested in what we were actually building, basically he just sort of brought us a term sheet one day. So, yeah, being 21-22 years old at the time we were like, Oh my God!, this is crazy. It’s like somebody wants to give us some money to pursue our little project.

So at that point we took that and decided to shop the deal and got some VC meetings with other different firms like Greylock and a couple others; and decided to settle on a $2 million round, mostly led by Lightspeed. And the remaining amount participated by Greylock.

[Hrishikesh: 04:54] Wonderful. We’ll come back to the VC part of it, but just that our listeners understand correctly what was Stackware about? Like what was the product about?

Yeah, it was an interesting product and it was definitely a little bit outside my wheelhouse. So when we were given this problem by VMware to explore the entire customer base that they had, a lot of our customers conversations came down to sort of on premises software licensing.

So, in big data centre environments where software can’t necessarily communicate back to a central server, well within vendor servers actually to meter the software correctly. Basically in that process what happens is that normally these big sort of software vendors negotiate these licenses, like these contracts with, the companies that are buying their software to say, ‘Hey, since we can’t actually meter this directly,  we’re going to have auditors come on premises and then validate that you’re actually using the software and the way that you say that you’re using it. And basically if you are found to be out of compliance, which means that you are using software more than you were actually accepted and expect it to be using it. Let’s say, I have 500 licenses of Photoshop and then it turns out that 505 people are using it inside of the data center. Basically that’s grounds for litigation. But what would often happen is that the leverage point for big vendors like Oracle or Microsoft is to actually sell you some other crappy products that you didn’t actually need.

[Vivek: 06:20] So basically coming full circle, what we did was, we basically just preempted, like provided data and analytics to say, “Hey, you are out of compliance”. And also the flip side of that is like, are you actually using the software you say you’re using? So that was the sort of flip side and it was a cost saving and in a way risk mitigation.

[Hrishikesh: 06:42] Right, Very interesting. And so a $2 million funding round and then, you see great traction. But then after a couple of years you decide to exit, you know, you get acquired. What was the rationale? What was playing on in your mind then? What happened exactly?

[Vivek: 06:58] Yeah. You know, it was a fascinating time for us because we were building a business and getting a decent amount of customers, but for us, we weren’t growing at the sort of VC pace that is expected when you take venture capital from these big institutional investors. And then how they make their bottom line is not having a good business. They need a spectacular business that’s doing, say, 50 million to 100 million ARR after only four or five years. So, we weren’t growing necessarily at the rate that we wanted to or what was sort of acceptable for VC backed business. So, well, with that in mind, we decided to sell. We sold it to Avi networks which is based out of San Jose.

[Hrishikesh: 07:42] Super. So your stint with Stacksware, what would be your key learning from it?

[Vivek: 07:49] So the biggest thing for us is that, if you’re going to go down the Avenue of taking VC financing, which is a big, big commitment, that’s a multi-year long commitment. So if anybody’s who’s listening, is in that position, they should definitely weigh that very, very carefully. So the biggest thing is that if you’re going to go down that route or any business idea, validate the market. I think for us the biggest thing is that metering on premises software, it was a problem, but it was also a declining market.

So when we talk to customers, a lot of times they’d say, yes, this is a problem, but their thought processes behind it was not like, how do I mitigate sort of software out of compliance or reduce my software licenses. It was like, how do I get to the cloud? You know, that’s the future, right? How do I start consuming cloud based software that’s always updating and I know exactly where I stand from the sort of compliance and sort of pay perspective. So that’s one thing – just take the extra time to validate the market, talk to additional customers, see how much they’re willing to pay, and do the due diligence on that as much as possible.

[Hrishikesh: 08:56] But do you believe that, let’s say you hadn’t picked up the funding round or if you had bootstrapped Stacksware entirely, would the scenario be different, would you have spent more time in validating things or moving forward with Stacksware itself?

[Vivek: 09:10] Potentially. It’s kind of hard to say in retrospect. There was definitely a big enough market to have a good sort of lifestyle business and I think we sort of missed out on that opportunity by taking the venture capital route. But the biggest thing is to validate the market and make sure that you know that and the sort of capital that you raise or don’t raise matches up with that expectation.

I think the flip side is that, if don’t raise money for a product or market that requires you to raise a lot of money, then you’re, you’re kind of shooting yourself in the foot somewhere, right? In certain cases, you can figure out that sort of sub niche market from the bigger market and then craft a product in that sense. But, yeah, I would just say like, what is the sort of capital dynamics of that market? And are you picking the right financing model? Whether it’s bootstrapped or VC.

[Hrishikesh: 10:12] Great. And so the acquisition happened in 2017. What happened after that?

[Vivek: 10:18] Yeah, so that happened in 201. So the first thing we did was took a couple of weeks off [Laughing] because I think at that time we were just grinding man. So we took a couple of weeks off and then moved to San Jose. Basically we started heading Avi’s SaaS side of the business. Yeah, so basically Avi is sort of a data center, you know, enterprise load balancer. And, we were taking our experience from building SaaS side of things and then translating that product into the SaaS world.

[Hrishikesh: 11:00] Interesting. And then you were in a very different setting here, right? You were working remotely for Avi. So how was that experience? It was like for a couple of years starting in 2007?

[Vivek: 11:11] Yeah, definitely. So that was an interesting experience because, at a certain point we moved back to San Francisco after a few months. And at first we absolutely loved it. It was very flexible in terms of having our own sort of structure around our day. We didn’t have to do the soul sucking commute. I don’t know if you’ve ever commuted in the Bay area, but it really, really sucks, especially if you’re going the distance that we are going from San Francisco to San Jose. So you know, the Avi team liked us a lot, so we had this sort of leverage to sort of say, “Hey, you know, can we work with remotely from here?” And a lot of team members actually did work remotely, so, the experience was good.

But of course it’s not without problems. I think the number one is social isolation. I think people take it for granted when they move into a remote setting, how much they actually need that social contact on a day-to-day basis. When you go sort of nine to five without seeing anybody and then transition from working on something to like watching Netflix, and still nobody else is there, it wears on you after a while. So that was the first thing that became apparent after a couple of weeks.

And then the second problem is the depth of communication. In a typical office setting, (and in our case as well since a lot of the team is in San Jose, so it’s the same time zone, but the sort of mechanics of this sort of office), we’re very much like, I pick up my laptop and I go over to your desk and we talk and hash something out at the desk, right? So it’s very easy in those scenarios whenever you’re in a physical office to forget or have a brain lapse about the people who are working remotely and including them because they are outside of your visual space.

So communication was a big problem as well. How do you sort of, replicate the experience being inside a physical office and having those ways of reaching people within the team. Also, since we were a hybrid structure, it was a lot more challenging.

[Hrishikesh: 13:12] Absolutely. And then how did this set up the stage for Pragli which is your current venture?

[Vivek: 13:18] For sure. So, the big thing for us is that, Doug and I are entrepreneurs at heart, so it was kind of a calling for us to make the next company after a couple of years. So, that was a big motivator and we sort of did this roundup of all the different ideas that we had. And, this remote idea was like one of the first I think that we had, and we sort of came in and gravitated back to it just because we loved the space.

You know, we were personally experiencing the problem that we had. So we did a lot of validation with a lot of different remote teams before we sort of dove in and told the Avi team that we were leaving to pursue this next idea. So basically that’s the origin of it. We experienced the problem, we did a little bit more validation with more remote teams and said, “Hey, communication is a huge problem!”. That was sort of a prompt and then we decided to quit and pursue it in a full time capacity.

[Hrishikesh: 14:14] Wonderful. So Pragli is essentially a virtual office for remote distributed teams. And so I was talking to Brenna from Doist the other day and she was stressing on the importance of asynchronous communication. So how does Pragli fit into the entire picture where let’s say, remote teams are focusing a lot on asynchronous communication?

[Vivek: 14:35] Yeah, yeah. So asynchronous matters a lot, but the idea that’s being promoted inside of remote teams right now is, first of all, it’s great, right? And I think you do need far more asynchronous communication when you’re working across time zones. You need to document all your conversations and it is super, super important.

But ultimately when you look at the problems, the top problems that a lot of remote teams are facing right now are social isolation problems and sort of depth of communication problems, right? And like you could say that asynchronous solves a lot of the second problem. But there’s another component which is just like how do I reach somebody quickly? Like the reality is that you can sort of like think of a technical spec in your mind and say, “Hey, here are all the details that I have. I’m going to give this to my team lead or another engineer and they’re going to go like deliver on it, right?”

But the reality is that engineering is, is very iterative. You know, like inevitably there’s going to be some problems and sometimes you just need to hash that out of our call. And a lot of times the calls are less than sort of two to five minutes long. And what we do with Pragli is enable those communications, have a lot happened a lot more quickly.

And the second point is like how do you facilitate those impromptu social opportunities that you normally have in a physical office? Like the sort of water cooler level conversations where a lot of the sort of beautiful conversations happen about bonding with your team or thinking of new creative ideas for projects that you can go and implement. So that was a prompt. And this is what I think, the asynchronous culture does sort of provide for structured happy hours over, you know, like at 5:00 PM on a Thursday, but it’s not enough. You need that sense of presence to really feel bonded with your team and that’s where Pragli fits into this sort of equation.

[Hrishikesh: 16:42] That’s true, actually very rightly said that a virtual office goes hand in hand with asynchronous communication rather than sort of competing directly with it.

[Vivek: 16:51] Exactly.

[Hrishikesh: 16:51] So also help us picture how does Pragli look like? As in, you spoke a lot about, let’s say, water cooler conversations, mimicking them or let’s say impromptu ones, official meetings or even there is a concept of silent rooms in Pragli, right? Just a quick picture of how it looks like or what is the structure?

[Vivek: 17:16] Well, that’s a really, really great point. So our inspiration for how the product will look like actually was pulled from Discord. Are you aware of Disboard Hrishikesh?

[Hrishikesh: 17:25] Yes, absolutely.

[Vivek: 17:26] Yeah, so during a lot of our conversations with teams, we realized they are sort of hacking Disboard to do what pragli is doing right now, which is sort of getting that sense of presence with your team. And the reason why they’re able to do this is because Disboard has this idea of audio channels. And even if you weren’t inside of audio channel you can see who was in channels and participate in them if you were available. So basically we translated that idea from Disboard into pragli.

Just giving a high level view of pragli – it’s kind of a standard three pane view, similar to slack where you have your teams on the left hand side and you have your audio channels in the second pane and your third pane will have all the people inside of your office. And you can click on one of your teammates and start a conversation with them, a direct conversation with them. And then basically your audio goes through to them immediately and they’re muted on the other side. The corollary in the Slack world or the Disboard world would be a direct message. We also sort of leave it to teams to figure out the mechanics that work best for them in terms of having it like DMs or otherwise .

And this sort of impromptu synchronous dynamics of the product have it actually in the audio channels. So basically if I go into the silent room, or the water cooler, or my virtual office hours room, basically I’m sending out a social signal, which just says something like, “You know, I’m inside of the water cooler, I’m available for a break and if anybody wants to come and hang out with me, they can”. If I’m in inside of the silent room, that means, “Hey, I’m heads down right now but any one else wants to join me” and feel like they’re actually working together towards a shared goal more so than they were before, They can also do that. So it’s almost like the idea of going to a coffee shop to get work done versus actually being inside of your room alone, even though you’re not talking to anybody necessarily. That sense of presence is super important to sort of solidify the mission.

[Hrishikesh: 19:26] Right. And I also saw a lot of gamification into the product as well, what’s your take on that?

[Vivek: 19:32] Yeah, yeah, yeah. So, what we quickly realized from analytics is that, we were running into a problem that a common communication problem, which is the empty room problem where when somebody signs into pragli, they look around, they’re excited by what they see but to get the real value from the product, you actually really need to start a conversation with somebody. You need to see the sort of communication dynamics that happen when you have a team of at least two people or more. So the biggest challenge for us was like, how do you actually get people to invite their other teammates into the product.

The gamification that we built in, is generally a direction that we’re trying to go with channels. It is to enhance the sort of single user mode experience of the product. So we actually even built in something like a trivia channel, which is basically an audio video channel as we normally have inside of Pragli, but just enhanced. So it’s like a trivia game that’s actually inside of that particular channel. And they can play that in a single player mode. So it’s some additional value that gets you more hooked with a product that reminds you, let’s you remember a little bit longer and perhaps you’ll actually invite people.

[Hrishikesh: 20:48] So let’s talk a little about the current stage that pragli is in. It’s still in beta, right? So how many customers do you have? What’s the traction that you’ve seen till this point?

[Vivek: 20:59] Right. Yeah, so we’re in beta right now and we aren’t asking our customers to pay right now because we’re in growth mode. We want them to be using it in all different sorts of use cases. There will be a free tier between like probably 3 to 5 people priced and then beyond that we’ll pricing per user. 

Right now we have around 50 weekly active teams and of those, 10 to 15 are using it fairly religiously per day. And those teams using it in the sort of mechanics that we expect them to, which is staying in eight hours a day and actually using it fairly frequently.

I should say since we started around five months ago, the first three months of those, we really spent a lot of time developing the actual product. So only over the last two months have we really barely been pushing on the sales and marketing perspective. And I think around a month ago we had less than 15 weekly active teams, so we’re really, really trying to grow right now.

[Hrishikesh: 22:00] That’s amazing. And the entire conscious decision around not having customers pay, that’s also really great to hear. Now what are the core marketing channels for you at this point? As in like what is your plan moving forward in terms of reaching out to remote or distributed teams?

[Vivek: 22:16] Yeah. Yeah. So we are trying a bunch of different stuff and  I think a super important thing is to try different things and just put it all out on the board and see what sticks. But for us, what’s really worked has been blogging about our journey. And putting it out there on Hackernews and developing really high quality content, where it can be picked up by the sort of Hackernews outlets and then featured. Because once it’s featured then you get sort of tens of thousands of people coming to your site, which is awesome. Yeah, that’s been a great source of traffic for us but of course we definitely know that it’s not a sustainable path towards a continuous adoption.

That’s why we’re investing a lot in content, actually on SEO (Search Engine Optimization) for particular keywords. And hopefully we get that organic search going for us. Also, we’re starting to see an element of sort of word of mouth. I know certain teams have actually heard about us from other teams so that’s an element there. We haven’t sort of gamified that in terms of referrals or anything. But the biggest channel we have is sort of developing these blog posts or interactive products that we can then market. So to give you an example of that, actually Doug, my co-founder pushed out an interactive application, which was basically a product called ProductHunt leaderboard. That got to the number five product of the day on ProductHunt yesterday, so we’re super excited about that.

That got us a bunch of traffic and a lot of that traffic went back to the site. So, some of those translated into sort of team usage. So, the big thing is like sort of providing value to your target audience, but it doesn’t necessarily have to be direct value in the remote space. So, like for example, ProductHunt people, a lot of them actually work out of offices, but a lot of them actually work remotely as well. So it allows us to target at least some part of remote people inside of that sort of indie hacker community of course. So if you provide value to that community, eventually you’ll develop your brand and people will start to recognize you a little bit more.

[Hrishikesh: 24:26] Absolutely. And you also have your own newsletter, which is targeted at the remote working community in general. Where do you see it going?

[Vivek: 24:37] Yeah, so currently with our newsletter, it’s more designed around people that we have had some sort of interaction with whether it’s because we did try a bunch of mailing as well, which didn’t have the best return. But even still, like every time they would interact with us like sort of positively or negatively, we add them to the list after we ask them, of course. Also, a bunch of people who are organically signing up after they’ve sort of found our interactive applications or blog posts.

One thing that we realized with this is that just because somebody sees your product and doesn’t sort of sign up right then and there, it doesn’t mean that they’re lost forever. As a customer, I think brands take like a multi touch approach, that’s the only way it’s going to work now. So, the newsletter is mostly designed for providing more than one touchpoint. Like every stage after that to reinforce the product thing in their minds. So every time we push out a newsletter, like people who we thought either sort of churned or weren’t interested all of a sudden start signing up. So I think you capture the base of people who showed interest and then start re-engaging them, so that’s the primary use case for the newsletter.

[Hrishikesh: 25:45] It’s great to see that you are trying out channels like Hackernews, ProductHunt, Newsletter, content plan in order to reach to the target audience. Now coming to Pragli, what is the biggest challenge that you see at the moment for pragli? Is it in terms of product,  marketing or something else? 

[Vivek: 26:11] Yes. So I would say the biggest right now is how do we get people to stick inside of the product long enough to invite their teams? There can be product improvements, there can be gamification improvements, that’s the first one. And I think we’re planning to add a lot more features around sort of user engagements, like ways for people to interact with the product more so as a single player. Even if it’s setting it up for the team that comes later on after they’ve been invited, things like that.                         

The second is how do you sort of get that organic traction without continually having to make another app or another blog post, right? So things like ranking on SEO are something that we really, really want to do, right? So, I mean, I think that the golden keyword that everybody’s trying to go after is work from home right now just because that’s what everybody searches when they think of remote, right? So like what are other permutations of work from home or other longer tail keywords that we can sort of like hone in on and sort of rank for that gets us in sort of consistent signup usage without having to, you know, necessarily put additional effort beyond that. Of course you have to make sure that your blog is all up to date and then you keep maintaining your domain authority of course. Otherwise you’re going to slip on those rankings. But how do we get to that initial amount where we’re going to rank in the top 10 for those keyboards?

[Hrishikesh: 27:46] Absolutely. And, so it’s just you and Doug currently working on the product working on pragli?

[Vivek: 27:51] Yup.

[Hrishikesh: 27:53] So given both of you are engineers, who do you think would be your next hire? As in what function?

[Vivek: 27:59] Yeah, we always fantasize about this, right? So we have one very serious one, which is a QA person. So you know, for us, we’re constantly shipping out product. Both Doug and I, even though I gravitate more to the engineering side than Doug does, we’re both engineers. We are constantly pushing out new features, new sort of improvements of product. And it becomes very, very laborious, especially if you want to keep pushing it out at a fast cadence than to manually test everything.

And honestly, it’s really kind of untenable though to have those sort of automated tests without taking a serious hit to your bandwidth. So I would probably hire a QA person who just looks at the product and we give them a laundry list of things that they need to test edge cases and we just push products. And then basically come back to us with a little list of bugs that we introduced or you know, maybe it’s perfect. That would probably be the number one biggest thing that we can probably do to sort of improve our bandwidth because that’s a massive context switch for us right now and we’d prefer not to bear that.

In terms of the next type of person who we would hire, and you have probably noticed this from our homepage but we actually have these sort of configurable avatars. Basically the idea is this sort of gives a quirky element to the product. And we actually, you know, sort of show presence based on those avatars.

So when we actually initially made those, we thought that they were going to be pretty gimmicky, but it turns out people loved it. You know, they love the ability to sort of craft their own digital identity and sort of show their friends and in a quirky way. So I think we could even make serious sort of improvements to that. Like how do you make additional assets for those little avatars? That additional improvement makes that personalisation stronger.

Maybe people can get a balloon on their birthday or like set their background to be some cartoon, the background of where they are, like whether it’s LA (Los Angeles) or SF (San Francisco) or India or whatever else, you know. So, a lot of illustrators, you can get for fairly cheap. So that’s, that’s more on the sort of fun side. And of course, you know the additional marketers to sort of help us with sort of content creation. I think a good quality content creation that we can start ranking all of those important keywords.

[Hrishikesh: 30:32] Wow! So at least three new hires you’re looking at, well, hope you find the best of talent in each that brings us to the end of our conversation with Vivek. It was an intriguing chat and we learnt a lot in the process.

[Vivek : 30:44] Absolutely Hrishikesh. Any parting words for makers would be just to go out and validate the market and make sure you do that and they’ll pay serious dividends.

[Hrishikesh: 30:54] Wonderful. Thank you so much for being a part of the show. For our listeners, you can try out Vivek’s product at Pragli.com, it also hosts an engaging blog, newsletter and a couple of other interesting products, so do take a look. Until next time.

Leave a Reply

Your email address will not be published. Required fields are marked *