Sunil Nagaraj on “Failing Well” & Investing in Early-Stage Deep-Tech Ventures

When Sunil Nagaraj started his first venture, Triangulate, made some mistakes nearly all founders make. After he learned to navigate those hurdles, and he built a committed team who created impressive technology, Nagaraj made the difficult decision to close his company. In a candid conversation with Shikhar Ghosh, Nagaraj reviews how and why he decided to exit and “fail well,” despite some encouraging signals. His thought process for making the decision would interest any founder considering whether to remain afloat or close. After sharing the facts that influenced his decision and how he proceeded to exit, Nagaraj shares his decision to become an advisor and a VC. The founder and Managing Partner at Ubiquity Ventures, a seed-stage institutional venture capital firm that invests in “software beyond the screen”™ startups, Nagaraj has unique insights into investors’ mindsets. He also shares the criteria he uses for investing in “nerdy and early”—early-stage deep-tech—ventures. A lightly edited transcript follows the video.

Sunil Nagaraj interviewed by Shikhar Ghosh on November 13, 2019, at Klarman Studios, Harvard Business School.

Sunil Nagaraj on “Failing Well” & Investing in Early-Stage Deep-Tech Ventures

SHIKHAR GHOSH: I’m Shikhar Ghosh, a professor of management practice at Harvard Business School. I have Sunil Nagaraj who’s a graduate from Harvard Business School, but has founded his own venture firm, Ubiquity. So, he’s the founding partner of Ubiquity and they focused on deep tech. He works with early-stage companies that are focused on technology and usually on difficult technologies or complicated technologies. So Sunil, you’ve been an entrepreneur, early-stage entrepreneur, and then you went on to become an investor, a VC running your own fund. Let’s go back to when you were an entrepreneur. Your first venture didn’t work out like 90% of seed-stage ventures. Most people will run the venture right till they’re out of money. So that becomes an indication that I have to shut down. You chose to shut down the venture when you had almost a third of the capital that you’d raised still in the bank and you had prospects at least some glimmers of hope that the technology you were building had some chance of working out.

So could you describe how- It’s one of the hardest decisions that an entrepreneur has to face, which is do I shut this thing down or not. Do I have a preponderance of evidence that this is probably not going to work? How did you go about making that decision?

SUNIL NAGARAJ: That’s a great question.

SHIKHAR GHOSH: So he has focused his investment activity on companies that are nerdy and early. So that’s an interesting combination of attributes. So could you describe how you made the decision to shut down the company well before you’d run out of money?

SUNIL NAGARAJ: Yeah, so we’re talking about shutting down a company, but in many ways it actually depends on how you think about starting a company. At the seed stage—this is for me 2009 when I launched Triangulate an online dating company—that used behaviors to make more accurate matches. I viewed—and I still continue to view—a startup with the Steve Blank definition. It’s a temporary organization built to search for a repeatable business model. Another definition, another version of that is that it’s a set of hypotheses that I’m going to validate with careful use of capital to run these experiments, these experiments, there’s no way to resolve the answer without running the experiments. And if you resolve them, you’ve uncovered economic opportunity.

So against that backdrop, it’s a different mindset than saying that I’m going to launch a company and fight till the death to prove that I was right. Those are sort of very different quantities as I think about it. So now when you asked the question about how did I decide to wind down the company, I think back to now eight years ago when I was shutting down Triangulate, and while it was running, I was full steam ahead, right? In my mind, and in every entrepreneur’s mind you sort of repeat to yourself, “We got to figure this out. We got to push ahead, we’re permanent optimists.”

And against that backdrop, I started to have signals with runway looming, maybe four or five months of runway left. So enough time to continue to fight. But that was one of the signals. Another one of the signals was the metrics that I would see from day-to-day. I could start to see that the metrics were leveling off. That is to say we were adding the same number of users every day. We were seeing the same activity every day and that started to weigh on me. Now, neither of those signals had changed my optimism. At that point, I was still full speed ahead. Layering on additional signals, I noticed that my team started to have fewer and fewer new ideas. We used to be bursting with ideas everyday at lunch, we’d talk about five new ideas and have to call them down. At some point we’d sort of asked them to [do] it, we had gotten most of the early ideas out under the table, which is to say that we had played out a lot of the experiments that we wanted to play out.

Another input is that I had talked to fellow CEOs of other companies, so very disconnected from my own company so we could speak freely about the state of our respective companies. And I shared what I was up to and what signals I was seeing and I was able to start to get input from that group of individuals that we may have scoured all the opportunity in the space at that moment. And that was the first kernel of doubt.

At some level with the startup, you are exploring opportunity while the music is playing and you start to wonder at some point, is it time and by that point with the input from the team, with the input from the metrics, with the input from other CEOs that I know, they were all sort of lukewarm about the future and I still was full speed ahead. Again, push ahead as an entrepreneur. Eventually I had a conversation with my lead investor, Trinity Ventures, fantastic firm. Patricia Nakache was on my board and we spoke in honest terms with Patricia and we always had honest sort of no facade conversations.

And we talked about what she saw in the business, what her partnership saw in the business, what I saw in the business. And it was the– Those precursor signals I mentioned combined with Patricia’s thoughtful conversation with me about the notion that we had pivoted the business quite a bit from where we started. And it was tougher for them to see what was keeping me so excited. I mean honestly, I think that it was an amazing move, an elegant move, for her to word it that way. It was very helpful for me to hear it, not as an accusation because no entrepreneur wants to hear that they’re not working hard enough or they’re not fighting hard enough and that’s not what she was accusing me of.

Rather she was saying, “I don’t see what’s keeping you so excited about this. So help me understand what’s keeping you so excited.” So all those signals started to get me over the hurdle. And a few days later I finally decided it was time to wind down.

SHIKHAR GHOSH: So you also mentioned that as you started to ask yourself the question, should I be thinking about shutting this down? You noticed that you would look at your watch, just little things like that about what your body was telling you or what intuitively you were starting to figure out, just talk to us about that.

SUNIL NAGARAJ: Yeah, so again, the notion of jumping into a startup is never rational. So founding a company, but generally whether through fear or passion or whatever emotion, a founder, myself, I was so immersed in what I was doing that I would lose track of time all the time. I would work until 2:00 AM because I hadn’t checked my watch since 5:00 PM and I’d just been cranking on something. That immersion, that obsession is usually necessary to create great products that reflect customer interest and generate great revenue. Against that backdrop, towards the end of triangulation, I started to notice that I would look at my watch at noon. I would look at my watch at four. It’s a very, very subtle thing, but that was a small signal to me that the obsession was starting to wane.

And honestly, it wasn’t for lack of personal interest. It was more that I had scoured all the different areas of opportunity. And getting back to that definition of startup as a set of experiments, I had genuinely started to feel whether– getting back to that definition of startups, I started to notice that I had played out all the experiments that I could play out, the experiments that I wanted to try and it became more apparent whether my mind wanted to listen to this, the facts are not. That there wasn’t a lot more opportunity to explore. And in many ways we may have played out faithfully all the experiments that we had signed up to play out.

SHIKHAR GHOSH: That in some ways the process of experimentation had worked. You just didn’t get the answer you wanted to, but you got a clear answer or at least an answer that gave you enough evidence.

SUNIL NAGARAJ: Yeah, that’s exactly right. I can only say this with almost 10 years of hindsight now, but at some level, we raised money to try a set of experiments to drive to a result and we did that. So we could say from a scientific angle, we successfully executed those experiments. Now Triangulate was a failure. We did not generate a profitable business. We did not have a successful exit. We did not return capital to investors, so I don’t want to muddy that clarity, but if you take it from a different angle, using a relatively small amount of capital to drive to an answer to retire your highest risk items early, I think we were successful and in the years since 2011 I joined a large VC firm and, in 2017, I started a small one.

I actually think it’s fairly rare to have a thoughtful prosecution of experiments with a really small amount of capital. Too often because of momentum or the ability to raise capital, you see folks retiring larger risk items much later in their life that should have been retired earlier. To say that differently, they delay those big experiments until later for a number of different reasons.

SHIKHAR GHOSH: Yeah, here we think about failure as coming in three different types. One we call constructive failure, which is exactly what you describe. You have a hypothesis, you test it. If it doesn’t work, you actually have saved yourself a ton of time and money in prosecuting it. Second is we call it either standard or normal failure. And that is the idea looks pretty good. You get a series A, maybe a series B, you go into the market, but the VC’s have funded another a hundred companies in that space and some win and some lose in that process.

So in the execution, things fall behind and some companies get to the top. But any industry has five, 10 companies that are going to survive. And so you’ve done all of these steps but it didn’t work out. And then the third one we call it destructive. And it’s where you lose your reputation, where you end up doing things that are illegal. Where you treat your investors or your employees really badly so that it’s a stain on your reputation. You can’t get hired again. So, and you actually cause long lasting damage with that.

And your example is a great example of constructive failure. And at the end of it, the employees actually did quite well. I think you’ve maintained great relationships with the investors. You went on to have a really successful career going forward. And so it’s exactly a set of experiments that didn’t work but taught you a lot about the process. And everybody said, “Yeah, that’s fine. That’s what we invested in.” Could you tell us about how you actually went about the process of shutting down the company? So what did you think through? What were you trying to optimize for?

SUNIL NAGARAJ: So once I decided to wind down Triangulate, which was a tough decision and required a lot of inputs, I then knew that this would be a major about-face. The momentum of the company was towards pushing ahead and I had set that culture and I would need to be very careful and methodical about how to communicate this, how to take care of the people that were involved, all the different stakeholders. One of the first things I did was to think through the financial implications from my team, right? This is a team of five people who had followed me to the ends of the earth. We had been working in my home office, living room, shoulder to shoulder for about a year with no salary, and then we had raised venture capital and now we were going to decide to wind down the company.

So I spent time thinking through what their major concerns might be, trying to anticipate them in advance and trying to understand how I could mitigate those as much as possible. Part of that is allocating some of our remaining cash towards severance. Part of that was making sure that I could help them recruit to find new jobs. So they had a soft landing. And part of that was the intangible, just them knowing that in a faithful and authentic way I would be there for them and I wouldn’t rest until they were taken care of because they had done such a good job taking care of the company along the way.

Another element that I then worked on was trying to understand if there was any residual value in the organization that I created, whether that’d be through a talent acquisition or through an IP sale or a larger purchase. So we carefully but efficiently executed a process to try to understand if there’d be interest in a purchase of the company, whether it be purchase of the P&L, the whole business or purchase the team or a purchase of an IP and these are smaller and smaller amounts typically.

And then finally it was talking with my investors and executing the wind down documents and returning the capital back. We had raised $750,000 and in many ways we executed experiments ahead of schedule. It was a failure, but we could say we executed them fully ahead of schedule. So we were able to return a meaningful portion of that $750,000 back to investors rather than wasted on activities we knew wouldn’t be fruitful.

SHIKHAR GHOSH: And you actually talked to the investors about the severance and so on. You didn’t just decide because at some point it became their money that you were using for this.

SUNIL NAGARAJ: Yeah, that’s exactly right. Triangulate grew from a twinkle in my eye to a group of five people in my living room, to a venture backed company. And at each step we pulled in more stakeholders in the company. So when it became time to shut down, it was in the end, my decision. But I wanted to make sure that I included the right people at the right time to make sure their concerns were taken care of and that the resources were carefully managed even through to the end. There’s often a– I know this now at some startups that I interact with, when the music stops, things can fall apart really quickly and things can fall apart unprofessionally, and I knew that I wanted to drive triangulate to completion and take care of the people who had taken care of me.

So along the way I did have conversations with my major investors as to how we should allocate the remaining capital, how much we should allocate to severance, how much we should allocate to other items, how much we should expect to distribute, selling furniture. There’s a whole set of activities and unfortunately it’s a set of activities that don’t place you in the limelight. You don’t get bonus points immediately for putting up the– You don’t get bonus points for doing a great job selling the couch. It’s just something that has to be done. There’s a lot of dirty work but I can tell you that at that moment I felt that I owed it to everyone around me to finish that process cleanly and, in retrospect, with many years of hindsight, I can tell you that it’s fairly rare and it’s dramatically– With many years of hindsight, I can tell you it’s fairly rare and it’s really appreciated that investors can have some return of capital back and can trust the people even when they know that no one’s looking. That level has been something that has allowed me to maintain great relationships with all my employees and my former investors.

SHIKHAR GHOSH: Yeah. It also tells the investors that you actually respect their capital, the fact that they haven’t really given it to you, they’ve given it to you for a purpose and if that purpose has been completed, then you’re going to give it back to them and treat it with respect all the way through, even if the amounts don’t actually make a difference in their fund. It actually says something about who you are as a person, how you think about capital.

You spent six years at Bessemer, one of the best known multistage large storied funds, and then decided to leave Bessemer to start your own fund and focus on deep technology and really early stage companies. So I think you described it as nerds and-

SUNIL NAGARAJ: Nerdy and early.

SHIKHAR GHOSH: Nerdy and early. Why did you make that decision? Do you want a different question?

SUNIL NAGARAJ: No, no, no. I just want to answer it in such a way that someone at Bessemer watching wouldn’t be mad. So after spending six years at Bessemer Venture Partners, I realized that I had learned the practice of venture capital from some of the best in the industry. I also realized that there were certain areas that I could pursue better on my own with a tighter focus. So I took, in many ways, I took what I learned at Bessemer and I skinnied down the focus and I skinnied down the team, specifically Ubiquity Ventures, now invests exclusively in what I call software beyond the screen. That software, not on a computer, not on a phone, but running in the real world. That’s why the firms called Ubiquity and many of these areas you can use the moniker deep-tech to describe them.

In terms of team, I’ve gone from having a large resourceful team of lots of experts down to one person. So now I’ve leaned into this model of being agile, responsive, generally being able to meet a company and get to an offer to invest within 10 days. And I do that because I work alone with a small team and I have a tight sector focus. So generally I’ve been able to recruit and secure diligence experts in my sectors before I even meet the company. So this is a firm that’s sort of good at one thing and can move quickly on this one thing. And I’m not only enjoying it, but I think I can produce great returns with that revised focus.

SHIKHAR GHOSH: So having been an entrepreneur and sort of struggled through the ups and downs of a typical entrepreneurial journey, how do you think that’s affected you as an investor?

SUNIL NAGARAJ: So as I think about my background from Triangulate and how it impacts my day-to-day now as a venture capitalist. I feel that the first six months, the first year of a startup is really a tremendous emotional journey and it doesn’t get enough credit for that. Too many people, including many venture capitalists, view the first year of a company the same way they view the fifth year of a company, how many people, how much in revenue, what are your milestones? And I think there’s a different element here around marrying a vision with a few early teammates with prospective customers and doing that in a way that doesn’t set the company up.

Let’s see here, there’s an element of understanding the nascent stages of a company and treating it like a temporary organization that’s hunting for a business model. We did that with Triangulate and the companies that I invest in now and the companies that I partner with. I’m on the hunt with them. That means that for example, we rarely set revenue targets six or twelve months out before we have first revenue. We realized that we need to be agile. I tried to keep Ubiquity agile. Triangulate was agile and the companies I back, I want them to be agile and I don’t want them to expect that they need to report hitting certain milestones or revenue milestones or metrics milestones if they revise and change paths.

So generally when I’m investing, leading a seed round, I’m helping to fuel the hunt and financing these different experiments that they’re running. And each experiment that gets resolved could take the company in a different direction. So I want to support that learning organization just like we did with Triangulate. We had several zigs and zags along the way and they were productive even though the ultimate goal was not successful. And so now with Ubiquity, I’m generally investing about a million dollars at the seed stage, and that is to fund a hunt for twelve to eighteen months to continue to find and hone product market fit. If you take a series A or a series B mindset and apply it to the seed stage, you sort of force a nascent organization into excessively rigid structure.

SHIKHAR GHOSH: When you look at companies in deep tech, you’re combining two sets of uncertainties, the uncertainty around the technology with the uncertainty around the market. So if I was doing an app, an eCommerce app of some kind, the technology is sort of known, it’s really can I find the fit with the market? This combination of uncertainty, how do you deal with that and how do you even evaluate whether something has a market, has potential?

SUNIL NAGARAJ: Yeah, that’s a very good point. With regard to Ubiquity taking two types of risks, both maturity of the technology and maturity of the market. As I think about those two areas, I’ve put in place different mechanisms to help me to resolve that risk. So on technical maturity, said differently, understanding does the technology work, should it work, will it work? I’ve developed a group of individuals, I call it the Ubiquity extended team. This is a group of 40 experts in different technical areas that are underneath the Ubiquity umbrella. I have a relationship where I ask them to pitch in on a same day basis to help me evaluate technical elements of different pitches that I think are promising.

That is not the same as building a 100% confidence that a technology will work, but there are other things I look for in the team’s background, but I pair that with the extended teams input to try to get a sense of if a technology will be able to get to the point that it will function, it’ll create value as we think it will. On the other side of the table, on the market side, I would split that into two kinds of risks. So you mentioned the idea of customer risk, will customers want to bite? There’s another element of financing risk that, will the VCs follow on VCs, later stage VCs also be interested in this.

So at some level I’m looking for the intersection of three different trends. It’s newly possible, customers want it and VCs will want to finance it in 12 months. And that’s a very delicate thing to find that overlap. I do believe that in an efficient market you get paid for taking more risk like that. And if I’m able to assemble an organization and processes that allow me to reduce that risk, that’s how I can create outsized returns with this tight focus.

SHIKHAR GHOSH: So when you do this, you’ve chosen to work alone as the only partner in the fund. A lot of VCs would say, “Okay, I’m going to start this way, prove that I can do it, with a smallish fund and then I’m going to keep growing the fund, keep growing the partnership because the measure of success is how big I can be. You’ve taken a very different approach, right? You’ve said that what you want to do is just remain this way and create follow on funds, so on, but not change the nature, not chase growth. Why is that?

SUNIL NAGARAJ: So I mentioned spending six years at Bessemer Venture Partners and it was during that time that I realized that there would be an opportunity both given my own personal interests as well as given the opportunity in the market to create a agile, nerdy and early, but institutional grade venture capital firm. This means investing $1 million leading seed rounds, but being an empathetic partner that can understand the code and understand the technology that belies a lot of these startups.

As I think about that, the anchors that make that exciting for me are very special. One office, one person, tight focus, the ability for entrepreneurs to know exactly who to call at any time, a hands on approach, a diligence process that can be less than 10 days, which requires that I’m personally on the phone with every part of the diligence, whether it’s customer calls or expert calls. So being able to synthesize that, I don’t believe that scaling is more important than authentically prosecuting that strategy. So in particular, I do think for the moment that Ubiquity, the model works with the high touch centralized process that doesn’t demand doubling the fund size or tripling the fund size. It’s not clear to me that there are scale advantages to doing nerdy and early seed investing. I think the authenticity and the focus and the reading between the lines kind of diligence that’s necessary can be effectively done or maybe most effectively done with just one individual.

SHIKHAR GHOSH: So, since you are involved in deep tech right now, what are the areas that you’re most excited about? The technologies, the particular applications of technology?

SUNIL NAGARAJ: So, at the moment, Ubiquity does focus on smart hardware and machine learning. Two technologies that allow software to run in the real world and understand the real world, permeate the real world. So what we’re seeing is a massive market size expansion where software which was confined to a desktop computer with a keyboard can now run everywhere. Again, hence the name Ubiquity. Within that, the two areas that I think are very exciting right now, there’s a wave within the two areas that are very exciting. There’s an element of machine learning that has to do with adversarial learning, where you’re pushing machine learning to push one machine learning model to push another machine learning model to get smarter and smarter and smarter.

This notion of adversarial learning even hearkens back to War Games, the Matthew Broderick movie from 30 years ago where the computer would play against itself to make itself smarter. Now, machine learning and neural networks take that to the next level and there’s a technique called deep reinforcement learning where a computer can explore a particular space like the game of chess or go, play itself millions, billions of times and build up its own rule set and be able to best any human player. I think there are many other areas and other domains where there’s a confined rule set and now you can train a computer program, a machine learning program to be able to make decisions that are initially a little better and then orders of magnitude better than a human.

So this touches on two different technologies GANs, generative adversarial networks, as well as deep reinforcement learning. But these are, we’re just in the early phases, machine learning came into its own in 2011, 2012 as the computer made it possible with GPU’s, and so we’re five, seven years into this. The idea of GANs is three or four years old and we’re now able to see the impact of a really well-honed decision-making model.

There’s another element of GANs which is to sort of run a machine learning network in reverse. So historically we are able to train a machine learning network by showing it a cat or a dog and with enough samples, it’s able to train its brain, its model, so to speak so that it can now discriminate, it can now tell between a cat and a dog. There’s a method of running it backwards. Where we’re actually able to create synthetic data, you’re actually able to train a model on something and then have it spit out new stuff. At the moment, we’re just on the edge of these seeing technologies like deep fakes where you see President Obama and his mouth is moving perfectly and his voice sounds perfect and it’s a synthetically generated technology. But I think there’s a lot more productive and high-value use cases for this technology.

SHIKHAR GHOSH: So actually you brought up the deep fakes. But increasingly what we’re finding is technology is getting so powerful and so widely used. Facebook is a standard example for this where it’s no longer just a technology to make life easier. It has effects in civil society and politics and global policy. The effects go well beyond. How involved do you think that founders need to be particularly with these kinds of technologies on the broader social implications? Or is that something that, when you look at a technology, is this anything you consider or is it something that is really left to regulators and laws?

SUNIL NAGARAJ: Yeah, let’s see. So I think the history of technology is one where we open up new vast arenas of possibility and they can sometimes thrive without any regulation or without any forethought about what might occur. But more recently we are running into issues pretty quickly, just because we can doesn’t mean we should. Another realization is that the technology we’re creating has massive impacts and can turn financial markets, can turn elections, can turn other major events in our society if we’re not careful about how we roll them out.

Now in the last year or so, we’re starting to see serious debates about technology companies, including the startups that I invest in, moving from being– In the last year, we’re starting to see technology companies, including the startups that I invest in, start to embrace their role as more than just dumb pipes. There was a notion for the 90s and 2000s that if you put something on the internet or if you put something onto the copper wire leading to your home, the telecom company, the telco had no responsibility for that. They were just the pipes. It was a similar notion with Facebook. Facebook was just posting what other people posted. Now because of political pressure, which is, I think, a good thing, because of the technical ability to do moderation at scale. We’re now switching to this regime where technology companies can and are starting to actually take an active role in ensuring that the implications of their technology are actively managed. I think that’s a helpful thing and I think we’re going to need to see more of that over time.

In many ways, this was inevitable. As technology began to run our lives and run our society, we needed to adhere to the same standards and–that’s a really dangerous word here. As technology begins to run more and more of our lives, we need it to come into the same umbrella of societal norms. We can’t shirk responsibility. We can’t have technology companies shirk responsibility over the impacts of their platform. And I’m glad to see that we’re seeing more of that. Part of that is the pressure to do so and part of that is the technologies are now making it possible to enforce proper norms at scale.

SHIKHAR GHOSH: You’re in the VC community. Do you see any changes occurring in the kinds of conversations that people have behind closed doors when they’re evaluating something or they’re thinking about their own responsibility? Or does this have to come from the outside in? I’ll just add to that question.


SHIKHAR GHOSH: So if you look at the pharma or the biotech industries that have similar kinds of things that their effect could be quite broad. They’ve created a bunch of self-regulatory norms. Same thing with what you can do with experiments and what you can’t do with experiments. These things are sort of well settled in some ways that there’s a set of guardrails around how people think about them. In technology, partly because of the nature of technology, we haven’t had that. But now slowly there’s sort of this pressure building up as technology affects, affects societies in ways that no one predicted even five years ago.

SUNIL NAGARAJ: So I hear about things where Google employees convince Google to not work on programs with the military that have to do with killing individuals. I hear about things where Facebook starts to accept responsibility for misinformation campaigns with WhatsApp in Southeast Asian countries. So these companies are starting to see the impact of their technologies, with regard to your question about if investors at the seed stage or series A stage are starting to talk about the impact, for many of the technologies we invest in, the impact is unclear at that stage in the company, but there are some conversations around surveillance, around military applications that occur much earlier and VCs are having a second thought about it. 

The capital that venture capitalists invest, in turn comes from a different set of investors, institutional limited partners, so LPs and many LPs are adding clauses into the investment documents with venture capitalists that preclude investing in certain areas, one of the areas that comes up is assault weapons and the other one that comes up is climate change related technologies, thermal coal. So our own investors can enforce some regulation as to where we’re, should not be spending time and then VCs are starting to appreciate that, including proper community management or incorporating the downstream impacts of the technology upfront can result in a more robust business and have better financial return from the Gecko.

SHIKHAR GHOSH: When you invest in a really early stage company and this combination of nerd and what was the?

SUNIL NAGARAJ: Nerdy and early.

SHIKHAR GHOSH: Nerdy and early. That’s a combination that’s fraught with a lot of uncertainty and you’re going to be on the board or at least going to have a strong say on the governance of the company. How do you see the role of investor governance in a company like that? When you get to later stage companies, there are a set of norms and structures and so on that are sort of well-respected and then when you get to IPOs it’s really well structured, but in the early stage there seems to be a lot of sort of make it up as you go. And what have you seen that works well given the uncertainties?

SUNIL NAGARAJ: So as a board member of a seed stage company, I think there’s a few roles that I play. The most important one I think is to force an entrepreneur and a management team to pull up out of the weeds on a periodic basis. Sometimes it’s monthly. Sometimes it’s every two months and recap what’s been accomplished and what remains. This is one version of looking back at the set of experiments. Without that periodic check in, it’s possible to continue working and working and working and lose sight of the bigger problem and lose sight of where you are along that progress.

So I’m an advocate of having board meetings without a lot of formal preparation. The spelling mistakes and beautiful slides, they’re not as important. But what is important is spending an hour or a few hours preparing for a board meeting by reviewing what’s happened, checking in, getting a status update for the CEO’s on mindset and then the report to the board is helpful. But a lot of it is that preparation work to pull out of the weeds. So that’s the first element of I think a board role for a seed company.

Another element is, as you mentioned governance, which at a seed company, a lot of it just has to do with making sure a few small items are done properly. This would be things like stock assurances, invention assignment, because we know as seed investors what the next hurdle would be when we raise a series A, the full diligence process that will be taken through. So having a few things buttoned up is helpful and that’s not a time consuming operation, but many first time founders at the seed stage aren’t aware that these things need to be done.

Another element of being part of a board of a seed stage company and this should really apply to every investor in a seed company is helping with recruiting, pulling in the right set of resources, whether it be consultants or full time hires or co founders or CTOs. That’s something that we do a lot and it ends up being extremely helpful. So as I think about it, forcing the team to pull up out of the weeds and look ahead, making sure things are buttoned down properly and recruiting a great team. Those are the reasons why I do think it’s a great idea to pull together board practices at the seed stage.

SHIKHAR GHOSH: What about, very often, right at the beginning you have a person who’s deeply technical, who knows the technology really well but might not know very much about markets or business things. And as the company progresses, when the technology develops a little bit, that person, his or her greatest value might be to continue with the technology, but you need someone else to be running the company in a business sense. And one of the rules that BCs play in series B, C, so on is to evaluate the CEO and if necessary replace them or move them aside or coach them or something. Do you see that same role in the really early stage companies?

SUNIL NAGARAJ: When I invest at the seed stage in a company, I like to try to box in and bound the risk that I’m taking as clearly as possible. This includes my work in advance on talking to prospective customers, evaluating technology, but I want to have a sense, this is not the same as me saying I don’t like to take risk. I just like to know the risk that I’m taking. At the seed stage, there is a torchbearer, the spiritual soul of the company, the one who’s driving it forward and typically within a small runway of 12 or 18 months of a seed stage company, it doesn’t make sense to add that as another risk item, sort of reorienting the team around a new leader.

It could be helpful to have a business helper and you might call that person COO, you might call that person head of BD, but if there is a CEO in place when I invest in a company, I’m rarely going to advocate that we make a change during the seed runway because there’s so much work to be done on honing product market fit and finishing out of V1 product.

SHIKHAR GHOSH: Probably just say to the previous question, one of the things you’ve done that you did at Triangulate was you really sort of go through a thoughtful process and how a company whose experiments have not succeeded can fail well, at the end of that process, the investors are actually pleased with the way the process was handled. All of the employees and the founder end up in a strong position having learned from this.

SUNIL NAGARAJ: Okay, great.

SHIKHAR GHOSH: Maybe one of the things that– I don’t know if we covered this last time, but we could just do it again. Take two minutes. What is your day like or your week like as a founding CEO of an early stage company versus as a venture capitalist? No versus yes, range of things that you don’t– Let me ask the question. So you’ve been a CEO of an early stage company and now you’ve created your own fund, but you’ve also worked for several years as a VC in an established reputed fund. What’s the fundamental difference in what your day looks like? What you spend your time thinking about, the emotional ups and downs between being an entrepreneur and being an investor?

SUNIL NAGARAJ: That’s a great question. In many ways you couldn’t find two more different jobs. As an entrepreneur, I was spending my time assembling resources in the face of adversity, trying to figure out how to say yes, I was a eternal optimist and working on building something day after day, I was going a mile deep into a single subject, into a single domain and I was working alongside teammates where we had arms locked and we’re fighting to solve a single problem together.

As a venture capitalist, each of those items you could say is on the opposite end of the spectrum. As a venture capitalist, I’m meeting many, many companies close to a thousand per year. So I’m way more broad in the number of interactions that I have. Furthermore, as a VC, I’m spending time trying to figure out why something may not work. What am I missing? What are the diligence items that I haven’t asked? Am I going to make a bad investment by mistake? So while entrepreneurship is fundamentally about trying to say yes, venture capital is fundamentally trying to avoid saying yes, unless you’re certain that this is one of the few bets that you’re going to make over the course of the year.

With venture capital, I only play a limited role. I mean I like to be involved with the companies that I’m in but I can’t take anything away from the founders. We’re really building something. So I’m an assistant on the sidelines and so I don’t have the same ownership of product diving deep into the weeds. On the flip side, as a venture capitalist, I also don’t have the same roller coaster highs and lows that an entrepreneur has. I feel them in a mitigated way. But as an entrepreneur, you could go from having the single best day of your life to the single worst moment in 15 minutes and that did happen to me and Triangulate.

And now as a venture capitalist, that never happens. I have several different balls up in the air. I take care of each one, but it’s not the same level of intensity. So they’re very different but you might summarize it as saying entrepreneurs are all about yes, VCs are all about no, it doesn’t mean that VCs are negative people, just means the fundamental mindset is spent trying to figure out how to say yes and trying to figure out how to say no.

SHIKHAR GHOSH: And entrepreneurship is a team sport. VCs, you work in teams but it’s a very different kind of culture. Can you talk a little bit about the way you interact with other people on your team?

So within Triangulate and I think within many startups, small teams are intensely collaborative and they’re co-creating something, watching each other’s back, building something, adding to each other something. But there’s a single thing that’s being created together. That may or may not work. Many startups, the vast majority fail, but during that window of time where you’re building something together, everyone’s on exactly the same project and everyone has the same incentives.

SUNIL NAGARAJ: At a venture capital firm, you have a firm, let’s say of five, ten people and each person is independently prosecuting a different set of investments. In many ways, you get the most leverage as a venture capital firm by having your ten people as ten different tentacles looking into different projects and opportunities. Whenever one of those ten people surfaces something interesting, they create the business case around it, create the investment case. They may pull in other people to help them create the investment case but fundamentally there’s one champion for an investment. So then when the group comes together, typically venture capitalists have Monday morning partnership meetings for a few hours to review the most exciting deals. The champion will present this investment and they’ll actually have a purposeful adversarial dynamic.

The other partners in the room will say, “Have you thought about this? Have you thought about this?” And the goal is to have a more robust set of thinking, but it’s fundamentally a different process than what happens in a startup where everyone’s co-creating. I hesitate to say that one’s independent versus collaborative, but the fundamental work style in a startup is alongside and in a venture capital firm it’s pushing each other against each other to sort of get to better investment results.

Shikhar Ghosh

Posted by Shikhar Ghosh

Shikhar Ghosh is a serial entrepreneur, angel investor, and Professor of Management Practice at HBS. Named one of the "Best Entrepreneurs in the US," by Businessweek, Ghosh has led some most innovative tech-based companies in the US and advised hundreds of entrepreneurs.