Company Building, Network Effects, and Understanding Different Business Models with Anu Hariharan

Hey everybody, it's Eric Torenberg, co founder partner of Village Global, a network driven venture firm. And this is Venture Stories, a podcast covering topics relating to tech and business with world leading experts. Hey everybody. Welcome to another episode of Venture Stories by Village Global. I'm here today with a very special guest, Anu Hariharan, partner at YC Continuity. Anu, longtime friend of the welcome to the podcast. Thank you for coming on.

Thank you for having me.

So Anu, you have focused on, among other things, the growth program at yc. YC has been famous for a long time for giving advice to early stage founders. That has now become startup lore like make things people want you to talk to your users and lots of other early stage advice. I'm curious, what is some of the lore that you've created around the growth programs? What are some of the mistakes that people who transition into growth typically make as they move beyond their Series B? Maybe let's start there. Some of the things that you teach early stage founders as they're transitioning.

Yeah, so we've seen often that when founder, when companies hit product market fit, which is the product is really valuable, there are lots of users who use it. You're typically around that company size of 20 to 25. And the CEO role actually changes dramatically at that point because when you have strong product market fit, you've probably raised a CDs a and in some cases the Series B. And so you have money now to scale. So it's all about if there are a thousand users using the product, how can I get it to 1 or 2 million? And the CEO until that point is essentially what we call the doer in chief. But for the next phase, which is the phase two of the startup, to really build it into a successful business, your role as a CEO has to shift towards company building, especially around senior hires and execs. And how do you first identify them and get that team to work together? So my partner Ali Rogani wrote a great post about this which is called the second job of the CEO. So I'd highly recommend for those who haven't read that and often we've seen that it's a shock to the CEOs and more often than not they don't even realize that they have to make that shift. What does that shift? So until that point, if you were the typical founder, the technical founder who used to cord, you probably did code until that point almost at least half a day every day, or you at least get into the weeds every now and then. And so Therefore, if you look at the calendar time of those CEOs who are at that 20 to 25 people, it would be a lot of one on ones code reviews, not necessarily on hiring. And the big shift they need to do, given everything is about company building, is switching that time. At least 50% of your calendar time is needs to go into recruiting.

Yeah. And say more about the exec hiring process. Besides not spending enough time, what are other mistakes the founders make or things that they should appreciate more?

So first of all, if you look at the median age of the YC CEO in recent years, it's around 27, 28. Therefore more often than not this is their first startup. If not, even if it's their second startup, they've probably managed a total of only 15 to 20 people before. Therefore they've never done the role of a Chief Product Officer or a Chief Technology Officer or a VP of ENG or cmo. They not only have done those roles, they've never hired for those roles. So they don't know what good looks like. And therefore a bulk of the time that they really need to spend on is in calibrating. So identify who are the best CMOs in the market and go talk to them. Not necessarily to recruit them, but try and get a better sense for who they are as people. What about them makes them really good? What do they do in their day to day job so that then you as a CEO can identify what do you need for your specific company. So first step is actually investing a lot of time in calibrating by meeting the best in those roles, then using that to coming up with a spec. Because too often what happens is you as a CEO feel the pressure. I need a marketing lead. And by the time you launch a search, you're already too late and that is a spec that isn't clear. And so you spend circles and hours trying to find this person when you yourself can't articulate who you're looking for. So that's the first investment in time. The second, once you have a clear spec, it's about accessing various networks or getting the search firms to identify a list of candidates, but more importantly to have a thorough interview process. So this is another big gap that I've seen CEOs do, which is you can't treat it like the hiring you did in the first 10 or the first 20 hires, because this is about hiring a very senior executive leader to your company that you need to look very buttoned up in your process as well as everyone in the interview team. Need to be able to articulate what they're really looking for. And you have to be very responsive because think of this exec that you're trying to recruit. If they're really good, they're probably in a very great job and more often than not it's going to be harder for you to get them. And for an exec, unlike an investor who is taking 10 bets, it is a one bet, at least for the next five years that they're going to make. Therefore, they need really high conviction on why you and why this company. So you need to, as a CEO, you need to make the right investment of time, not just yourself, but in even getting your team within the company that is going to be interacting with this candidate.

Yeah. I'm curious if you can give some advice for CEOs who are thinking about how to put compensation plans together, especially as, as, as, as they start scaling. What advice do you have there? What mistakes do founders typically make around compensation plans?

So in the first two, three years of a startup, you probably are not paying as much attention to compensation. Right. Because it's especially the first 20, 30 hires. Because it's all about having the right resources in place quickly to really focus on is there a thing here which is, is the product even valuable? But when you start hitting around 50 to 100 people, you'll soon run into compensation problems, mainly discrepancies in compensation. And this usually happens if you don't have a structured process or a leveling exercise to figure out where people need to fit. Because the people who arguably negotiated better during their offer may end up having higher salaries than the people who didn't. Therefore, when you're in that stage, the phase two, we often say the first step is getting an HR leader in place. And the first thing the HR leader would do is get a pulse check within the company, but will also be actively looking for close rates. Close rates are the first indicator that your compensation is off. So one of our portfolio companies recently saw that their close rate had dipped to 40% in engineering in spite of having incredible brand perception in the market. And it was quite clear it was because they had not adjusted compensation in 18 months and just adjusting that in one quarter improved their close rates to 75%. But how you do that exercise also matters. But great HR leaders will be able to one spot the problem. There are a lot, lots of surveys available that can help you benchmark and you should have a compensation philosophy. Always paying 90th percentile is not the solution. But if you pick a consistent philosophy, if you're in the 50th to the 75th percentile, have a rationale for that. For example, some companies say I want to compensate more through equity, which is the base compass between 50th to 75th. But if you have a compensation philosophy, it's more important that you use that consistently and don't make too many exceptions.

Right. I think Jack Almond might say something like or could be quoting someone else. There's so many things to innovate on in a startup that the things you don't have to innovate on, it's actually nice to just benchmark for you. Is comp one of those things where, hey, it's probably best to just benchmark what other people are doing? How do you think about the things that a startup should innovate on versus should just do what everyone else is doing? Not.

I think that it really comes down to the founders and the exec team, probably more so the founders. The startups that we work with, we see the founders compensation philosophy have a higher influence in how the company shapes. Therefore the founders spending time on articulating that philosophy and even their own clarity of thought around what's fair is super important to invest time in figuring that out sooner rather than later. I wouldn't say innovation is not smart because especially we live in the current market condition where San Francisco, the cost of living itself is really high and it's extremely competitive to hire engineers. So if you're a young startup that's still trying to build a talent brand, it's very important to have the right people in the seat and it makes a huge difference. So Brex actually has the slightly innovative compensation philosophy which their inspiration was more Netflix, which is you use a sliding scale to decide between base comp versus equity and they give the ownership to the individual to decide and therefore the employees in the company feel empowered to make that decision. Was the company telling them this is the bracket or this is the percentile that we are fixed on so you as an employee can choose. If you want higher base, then you're automatically giving up on equity. But it is a choice that you get to make and I think it's really helped them at least so far in standing out versus the rest of the startups.

Let's talk about board composition. You gave a whole talk this, I believe on the YC podcast or Startup School. It's a great talk. But if you had to pull out some of the main nuggets, what do you think is the biggest mistake founders make that they may not appreciate or the most important, important thing founders should, should think about, or what's their mental model for board composition?

I think founders make two mistakes. The first mistake is the amount of time they spend in looking for a board member. And the second mistake they make is sometimes listening to the board too much. So let's talk about the first one. More often than not, I am shocked that founders spend very little time in choosing their board members. You may run into a CDC process in two or three weeks, and you have signed up a term sheet and agreed to bring someone on the board who you don't know or you may have just met for the first time two weeks ago. And if you think about it, if your company ends up succeeding and staying around, the Median time to IPO is 11 years, and you're expecting this board member to be with you for 11 years. When you make a decision about a partner in life, you don't make that in two weeks. But it's strange to me that we make decisions about choosing board members in two weeks or even few hours, right? Even in those two weeks, it's a compressed process. You may end up spending only four to six hours with the board member. So that's the number one mistake. And I think that's where some of the founders do this extremely well. Mathilde from front, she has written about this quite a lot, and I think she's really paranoid about it because she cares about it as a cultural fit. And she spends a lot of time getting to know the partners at the respective firms very early on, even 12 to 18 months before the fundraise. Therefore, by the time she's about to fundraise, she knows who are those four or five people she would have on the board any day. And so I think it's important to invest time in getting to know them as people and understanding whether you would want them on your board. You also should do reference checks. You know, a lot of times when I talk to yccos, I'm surprised when they, when they realized that they could actually do reference checks. You should talk to pretty much every portfolio CEO that that partner has worked with and get a sense for, you know, a good perception for how that board member works and whether that fits with your style. So that's number one. Number two, when you bring someone on the board, there are times when I've seen CEOs, you know, especially when they do office hours with us, they would come and say, oh, but my board member thinks this, and therefore it must be right. And I Think like all advice, you can listen to advice, but at the end of the day, you as a CEO and as the team, that's the exec team at the company are closest to the problem than a board member. The board member is not there in the weeds working five days a week. So I think the best thing a board member can do is actually ask the right questions to make sure the team has flushed out the thinking and the team should use that to push their own thinking. But at the end of the day, day, the decision lies in the hands of the team.

Yeah, there's just news today. I think brandless is going to zero and I love the brandless team and I hope everything's okay. I just use that as sort of a segue of sort of. I want to talk about the macro environment. You know, with Softbank etc, we're seeing lots of companies who've raised a lot of money and are now either having massive layoffs or going to zero. And some people are asking whether a reckoning is happening. So when I'll ask on the macro environment, what's your take there? What do you think is likely to happen? What types of companies are likely to have struggles and either sector wise or status of company wise. And then on the micro, your advice to entrepreneurs to prevent it or to respond to it, both in a fundraising element but also in a just company building element. Maybe the macro is putting your growth investor hat on.

Yeah. So I think there is a lot of narrative in the market that winter is coming and I do think there will be some correction, but I don't think it's going to be as bad as it sounds. So in the last almost 10 plus years, and it's probably, some investors say it's been one of the longest bull markets we have ever seen. Have valuations gotten out of hand? Yes. Right. Because there's just way more competition, there's way more investors at every stage and arguably it's a fewer companies than investors. And so that imbalance has resulted in increasing valuations to an extent. But it's not completely out of line. If you look at the Internet economy, a lot of people say the Internet economy is of the past. I actually think it's beginning. The last 25 years was about getting everyone Internet access. We had broadband, then mobile, increased penetration and I think 5G will only accelerate it. So therefore the next 50, at least for the next 20 to 30 years, if you assume the Internet economy can even grow 10% a year, we're talking about at least five times to 10 times the size than we have today.

So the exits will be bigger, the.

Access will be bigger. There is no doubt. And so therefore these corrections that happen, people forget. Even though it was a bull market, you know, 2014 was also a tough year for growth companies, especially the ones that were capital intensive like the food delivery businesses. And therefore these type of corrections, I view them as opportunities. There would be unbelievable investment opportunities wearing the investor hat. But also as a founder, it's a great opportunity to even more strengthen the fundamentals of your business. So if use the approach of, you know, is winter coming? I tell founders that think of it as, think of it as, it's not going to be as slam dunk as it was last year to raise your round, but doesn't mean a round wouldn't happen. So use this more as an opportunity to drive more healthy hygiene in the business. So if this is, if you were all about growth the last two years, it's probably a right time to think about contribution margin and focus a little more on how to get it more sustainable and scalable before you turn on growth again. It doesn't mean growth is tapped out. And so I think people get confused in this conundrum of if winter is coming, growth is done, growth is not done. There's still a lot of appetite for demand in the market, but it's this healthy dial that you need to turn. If we were growing for two, three years now, it's good to turn the dial a little more to focus on contribution margin. Once you have that more healthy, then turn the dial back on growth.

And what are the biggest levers that one can dial for contribution margin? And maybe a company like Rappi as an example because they raised the big softbank round and you guys are involved. But another example could be fine too. You can keep it vague, but what are the biggest levers companies can.

Yeah, so I think that most marketplaces go through this. This is not new. I'd say even Uber did this in the early days. So you've expanded from one country to multiple countries. You may have gone from 20 to 1,000 markets. So right now it's about, let's focus on our mature markets and in maybe the next layer of markets that are going mature, what are some fundamental things in our business model that we could do that we can drive higher contribution margin power. So let's take the Rappi example. Their first market was Colombia. It's growing still pretty well and it's quite profitable as a market on its own. But if you are thinking about Experimenting on others, layer on services beyond delivery. Like it could be advertising or it could be samples. This is the time to try that because then you're testing your proof Points of at scale, what is my business model versus Brazil is the newer market. So it is clearly an investment market and everyone gets it that it's an investment market. And the market condition in Brazil is different. Right. It's a newer, they were only 18 months old. There are two, three other competitors in the market and they need to reach a certain scale before they start focusing on business models. So as a founder, you need to separate your core company into different segments, figure out which ones are more mature and try to figure out what is the right business model there and then determine your playbook for the rest of them to come to scale. So combining everyone's burn and saying oh my God, this company burns X dollars and therefore it's bad is not the right approach. So I always use this distinction. Is a discount driven burn or operational driven burn. If it's discount driven burn, this is the time to switch it off. If it's operational driven burn, which is you know that you're scaling and you're about to hit the tipping point and you're making the right investments required for scale, you need to continue to do that.

Yeah. And is it easy for people to know which is which or which they're in?

Well, if you don't know that, that's step one. Right. So if you have raised, you know, a billion dollars and you don't know that, that's worrisome. But more often than not, CEO' know that and they do know what is the business model at scale and they've usually, if they've raised a lot of dollars, more often than not they've at least proven it in one market. So it's about coming up with a playbook to get other markets there.

Yeah. And we're putting your investor hat on what's going to determine whether a company like that, we don't have to get too specific in there, but maybe a marketplace like that or on demand marketplace is going to be a, you know, 10x or 20x or you know, bigger than it is today versus it's peaked because there's too much comp. Like what's, what's the criteria that will determine how it plays out? Like what does it have to figure out in order to, to grow significantly versus versus if it doesn't, it stays the same.

So founders usually post series B, try to articulate how big their company can get. Not just for investors, but for themselves. Because it's also important to communicate that internally to the company. And this is another thing, we often tell founders to do it sooner rather than later because too often we've noticed that a founder that hasn't articulated is usually struggling. Right. So you can look at any market and get a sense for that. You can do it two ways, top down or you can do it bottom up. Now, the mistake of doing it top down is you may look at a fixed tab. This was a classic problem for DoorDash in the earlier time, like two, three years ago. That wasn't that early ago. Right. Where most people will look at grub in the US and but this was the TAM. But the fact was DoorDash was delivering food and therefore the TAM invariably was not Grubb's tam. Because any restaurant that didn't have delivery could sign up with DoorDash. That was not true for Grub three years ago. And so the problem with the top down TAM assessment is you limit yourself. So it's almost you as a founder. You need to understand what is your addressable market size and therefore you. The only way you do that is by looking at your customer base and who are your core merchants today. Right. So for something like DoorDash, they were very clear from day one that I'm opening up Tam. The second strength for DoorDash was it's from the Peninsula. Right. It was pretty much all suburban markets where Uber wasn't strong because a lot of people use Uber in the city. It wasn't as common in suburbs and therefore DoorDash's forte or strength started in the suburbs. Now today, of course, they're quite prevalent in the city as well. So having that clarity of thought on what are your core strengths early on will help you articulate the market potential. And so for all of these companies, there is a lot of market opportunity ahead of them. It really comes down to execution because if you look at it, these businesses are not network effects businesses. DoorDash Food delivery is not. Ride sharing isn't. And so it's not a zero sum game. And I think we talk a lot about everything being a zero sum game. No, it's not.

It's not winner take all.

And therefore it comes down to sheer execution for how big they can get.

Yeah, you gave this great talk that listeners should check out about. I believe it was nine different business models and how founders should think of them. I want to go through some of those business models and it'll be similar, but instead think about how investors should think about it and how you think about mental models for each one. So maybe we could start with marketplaces or around series A or series B. What metrics are you really digging into and what are you, you know, what's the framework you're thinking about in your head as to whether this could be 10x or whatever it needs to be to be a worthwhile investment for you versus hey, it probably doesn't have that potential.

So at the CD Z& Series B, I think the risk is still relatively high, which is companies usually have demonstrated early signs of product market fit, which means, you know, they're coming off a small base, they might be growing 20 or 30% month on month, but it's still not slammed up. You have to really wrap your head around what would the consumer demand be and would this go mass market. And so therefore, at that time investors are really looking for a 10x potential because it's worth their time and risk only if they say that that is a 10x potential from the point at which they're investing. So as a founder at that stage there are really three things, right? Especially in series A and series B is slightly different. So I'm going to separate the two. At series A it comes down to three things which is team market opportunity and early signs of product market fit. So team, obviously it's why you and what unique insight do you have about the space and the product that others don't? That is super important and a lot of people miss that. So this example of doordash that I gave, which is Tony's insight into starting from the suburbs very early on and the fact that we are doing delivery and therefore it is not the tam that grub has was there from day one. And that's super important. You should be able to articulate why. What's so unique that you know that others haven't figured out yet? Second is the market opportunity as a result and then the third in early sense of product market fit. So if it's marketplace, let's pick in the case of doordash or Instacart, at the highest level, it's gme. How fast are you growing? But this is where the discount driven growth matters, which is you really look at customer acquisition cost at that stage to the extent you can show all organic the better because that means people truly want your product and you haven't spent a single dollars in marketing to get anyone and you're still growing 30% month on month. That is a real sign of product market fit. So it would involve GMV monthly growth. Really? How much of it is organic, Preferably oil. There's definitely a bias towards all and then unit economics on a per transaction basis. So this is where people look at blended, but more importantly, they will look at show me your most mature segment. So in the case of doordash, it would have been San Francisco. And of course by the A, I can't remember now if they were, if they had proven out everything, but by the A, maybe they were just in San Francisco. And so maybe the blended contribution margin on a per order basis was negative, but it surely declined a lot and almost break even.

Wow.

Right. And so as you go to the B and the C and the D, it becomes more about show me improvements. San Francisco was your first market. How is San Francisco doing today? How are the markets that you launched after San Francisco? How have they done? Hopefully they should have done better because you've Learned, you had 24 months to learn in San Francisco. So surely if you launched in Chicago, it should take you 10 weeks.

What are the benchmarks of what unit economics should look like?

So that really depends on the business model and the company. I can quote here a percentage, but it won't make any sense, right? Because a fintech company looks different from a marketplace. A traditional marketplace that didn't do delivery or was not fully managed historically always had 80%, right? And not contribution margin, but the gross margin. Because if you look at Airbnb, for example, they have a 12 and a half percent take rate and most of that translated into margin. But now marketplaces have expanded beyond just one to one connection, right? There is a lot more that goes on. And if there is a lot more that goes on, then it's no longer 80%. So what you have to really break it down into is today, what is your unit economics. So let's pick, for example, if you order $45 of worth of food, okay, you have to pay X to the restaurant, Y to the delivery person. So your net revenue truly is, you know, something else. And then you work down all your customer support payment processing costs fully loaded. And then what is your margin per order today? And what could it be a year from now or two years from now? That's important to articulate, too. So in the case of Instacart, for example, they've started doing a lot of CPG advertisements because they have so much data on you as a consumer, on your buying preferences, that without having to share it with anyone, they can selectively tell you, hey, if you're Trying this shampoo. Have you tried panty? Right. And so that then layers into X dollars per order. So therefore you can, you need to have your current view of unit economics but more importantly future view and clarity of where it will head. Both of those are important. So I don't think there's a necessary percentage per se. So in the case of on demand businesses, which seems to be in the narrative all along these days, I don't think the margin is that high. But hey, if you can make three to five dollars an order and you know you can make a, if you are on a pathway to get to $120 million EBITDA business, that's a pretty good business.

Totally. And just to zoom out for a second, the businesses you've done so far led the series be, Checker, Convoy, Brex. What am I vouch, vouch. Any other ones I forgot?

Lob, there's quite a few. Yeah, we've done 20 so far. 20 portfolio companies and roughly I think 50% of them are the bees.

Cool. The let's go to fintech. It's something you're thinking a lot about right now. How are you thinking about it from an investor in terms of what's, what's most interesting broadly, what are you looking for and how are you evaluating these companies?

I think Fintech is changing very fast, very fast and there is a lot more innovation possible today. End to end, full stack. Let's look at Brex for example. Brex launched as a corporate credit card for startups. If you look at it, you think well why do we need a new corporate credit card? Anytime someone asks me that I always say have you tried applying for Amex? Even today Amex takes three to five days to approve a credit card for you, especially as a founder and you need to send a bunch of statements and you also have to personally guarantee. And Brex made it really simple by saying in the day and age today where everything happens on mobile and online without talking to a single customer service, it was literally get it approved in 10 minutes. As long as we know you have an investor that has wired checks in your account and you register your entity. So you get a virtual card in 10 minutes and you get the real card shipped to you in 24 hours, the physical card. And when they were beta testing NYC in the first year because they hadn't launched live and around 100 startups were using them and when they launched live, I don't think even they expected, expected this demand. So it just shows the pent up demand that's sitting in financial services due to the lack of innovation from a large number of incumbents. And you know, we can go around and say, well this is about launching a mobile app and writing a new algorithm. It's not you to be able to ship both to approve do the full KYC process and to be able to ship you have to build a full tech stack from scratch. It's not some layering or plaster on top of someone else's solution. The second thing that was very convenient for the Brex customer was each time they swiped the card you got a text message at the time of the swipe, which is super important because if it was few minutes delayed, the experience isn't good, right. And you could just submit the receipt. So therefore anyone who was controlling the dashboard within the company can do real time reconciliation of experience. So imagine as a startup, you're managing your burn, you have lots of dollars in expenses. At any point in the month you want to be able to reconcile all the expenses to know which categories does it fall in. It's as simple as that. And they were shocked that there was so much demand for it. So I think it's like we will see more such fintech business models attacking different levers of every aspect of the fintech sector. The second is a challenger bank, right? And a lot of people, again the popular narrative is I don't understand why we need a challenger bank, but my question always is, are you able to open a bank account in five minutes? Right? And I think that's where we are headed. We are headed where we don't have patience. We don't have time to do a back and forth or an archaic fax process or send something, scan documents by email. So Monzo in UK gives you a bank account pretty much instantly and then it's about, well, if you're the bank and you know everything about me, why aren't you helping me control my finances? Why do I need to have 20 different apps that are helping me do that? And so that's been their vision from day one. So literally today in the uk, if you're using Monzo, if you get your salary into the Monzo account, you can split it into three parts. You can say this is my savings, this is my regular bills that I pay and this is all I have in the month to spend. So in your week, four, if you're running out of cash, it'll tell you, Eric, you can go to this restaurant. I'm sorry, I don't think you, unless you want to override your own rule for savings. And so I think that if your app or tech stack is built from first principles today, which is mobile only, the bank will look fundamentally different. Now many people talk to us about the Chase app or look at the bank of America app. None of them can send real time notifications. I can't use IFTT to set rules to help monitor my behavior. Even the statements I have to do a PDF. Right. So when you give the NPS on Monzo is 80 from the customers and they are at 3 and a half million accounts, which is roughly 6% of the UK's current account market. And majority of this was organic. So that just tells you that what the potential is for a new banking product. Yeah.

And for zooming out where companies are growing organically, besides having a beautiful product, what else are they doing to drive all that organic growth? Like what is most effective or how do you advise?

Yes. So I don't think it's build it and they will come. Right. I mean this is another common assumption that oh, if I build it, my users are going to come. You have to invest. All these companies invested in acquiring their users. So I think it comes down to your philosophy of how you want to acquire users. There is no right or wrong answer. In the case of Monzo, for example, Tom Blomfield, who is the CEO of Monzo, is an excellent guy at growth. So he actually led growth for a couple of startups prior to Monzo and is a great community builder. So both Tom and Jonas, the co founders, invested a lot of of time in the initial years hosting town halls and community events for the community in London and they would talk about every feature. Even the transition when they made from prepaid to checking account was done through town halls.

So it's community, it's content, it's.

Yes. And they also basically created this whole genesis of we are building the bank you want. So consumers often gave a lot of input. If you look at, I mean how many banks have a block log? Monzo does and they have ten thousands of users who log in daily.

Yeah. So there's new bank in Brazil. Is there going to be one basically for every major region? Like are you, are you sort of looking for the new bank of X or how do you think about.

I think that fintech will, you know, there will be global players, but there'll also be regional players. First of all, again, it's not a network effect business. It's not going to be zero sum game, that is trillion dollars of capital locked up or market cap across banks. Therefore there will be a few players in every market and it will take much longer for each of them to scale because banking is not easy to scale.

Yeah. What are the, the metrics by which you're evaluating? Presumably Monzo has a bunch of people who are doing something or trying to do something similar. Brex2 perhaps. What are the metrics by which you're evaluating?

Yeah, so at the end of the day, first step is coming down. It comes down to, to number of active users that you have for something like a Monzo. If you're expecting people to use it as a bank account, that card needs to be used weekly. Right. So it may sound like a social network, but it's not. It's being used for transactions. So therefore you do track weekly active users. You also look at the percentage of users that put salary in the account because it's important. You may not start there. Surely in the early days of Monzo people didn't put salary, but today they do. Right. So you have to track that evolution of what percentage of users because salary is like the ultimate customer stickiness. And then in terms of the business model, you look at revenue drivers and cost drivers. So revenue drivers for a bank are usually interchange. You have lending, you have cross reference services, There are a bunch of new products like subscription is a pretty big thing for banks these days. So you can look at the different revenue drivers and really look at revenue per user. And especially you want to look at salary reduce user versus non salaried user. And this is where I mentioned you can't look at a company blended and say this is their blended revenue or oh my God, their burn is blended this much, which is the salaried users invariably are going to be way more profitable. Right. And they're probably some of your earliest users. And so you look at that and go, okay, this is that segment. This is the business model. I need the rest of the customer segments to.

And is the discount driven versus the operating, you know, sort of dichotomy you mentioned earlier for marketplaces, does that apply.

Everywhere or is that yes, even for challenger banks? Right. If you are spending, you know, 300 to $600 in CAC to acquire and your LTV is thousand dollars one year LTV then you know, I'm not sure, I mean the payback is pretty good. But you have to question yourself, is the $1,000 LTV from salaried user or non salaried user? So you have to. The discount driven is basically saying I'm growing my user base, but 2x but it's at the cost of everyone being negative and my payback being longer than 12 months. So that you ride on the borderline of Is it discount driven? Yeah.

We hear some venture capitalists who are dubious about lending businesses and their ability to be venture scale after a certain point. Do you have a take on.

I think it's going to change dramatically and people are going to be surprised. I'm very bullish on lending business. If you look at the classic archaic lending business, the word lending itself sounds archaic and therefore it's very easy for us to conclude it's going to be boring. I think the world is going to move towards buy now, pay later and that is going to happen through lending and it's going to be more innovative forms of lending. And this is where I come back to why startups have an edge on this. Because since they've built the tech stack in recent years and full fledged, it'll be way easier for them to finance someone at the point of sale sale or at the time of transaction than it is for an archaic bank to figure out how to make sure ANU has money at the point of indebted.

Yeah, we did this company adi, which is like a firm for Latin America. It'll be interesting to see global opportunities there too. Let's talk about consumer, how you think about it in terms of from a macro perspective of what's interesting. Where are you looking to invest? Where is there opportunity? Some people are down on consumer right now with particularly consumer social as it relates to Facebook, its prominence. And then let's talk about what metrics are most important that you think about in terms of valuation.

Yeah, I love that people are down on consumer because I'm very bullish on consumer.

Yes, say more.

So I think that we are seeing a shift in types of communication networks. So I'm going to go a little theoretical at the start because I think it's important to set that context to then say what what I expect in consumer. So if you look at historically, I've talked about this before, there are really three types of communication networks. The first was broadcast which the value of the network is proportional to the viewers you see. So Yahoo in its classic example. The second was Metcalfe's law which is the value is proportional to N squared, which is Facebook because there are more peer to peer connections but they're peer to peer connections and so the value is proportional to N square. The third, which is what I think we are going to see in the next 20 to 30 years is Reed's law, which is v is proportional to 2 to the power n. That is the value of network is proportional to the subgroups and the groups that are formed, which is slack and what subgroups. And that has happened only in the last five to six years. I think we're going to see a lot more of that. What does that really mean? I think that we're moving towards a world where we want to connect with users with vertical interests. Be it gaming, crypto, remote work, you could pick pretty much any interest area. Could we have done that as easily in the last 10 years? Probably not, because we didn't even have everyone online to do that. And it was incredibly expensive to go target that one vertical niche. So it was easy to go horizontal. So therefore, I think we're going to see a whole host of consumer startups that target niche customer segments with broad interests that have the potential to be global. And if you look at it, the value of that network grows exponentially faster than the other two. And so it's going to be easier to create them and therefore it'll be possible to have more global vertical networks as a result.

Yeah. So what are examples that come to mind? Paint us a picture of WhatsApp group.

WhatsApp. Right. If you draw the curve literally of every social network that was ever launched, Facebook, Twitter, Snapchat, and if you compare that, WhatsApp, I think hit the same user scale in two years.

Yeah.

But while they took four or five.

Right. So you're saying, are we gonna see more messaging apps or more things like TikTok or whatever?

It could be anything. Right. It could be. I mean, discord. We're already seeing it. Right. And so I think there will be new startups that we haven't. You and I haven't heard of that are going to start in various interest groups. And I think that's the way the world is moving towards because right now it's too horizontal and people don't have clarity of consumption around it. And it's going to be easier to form those vertical groups and connections. So it could be discord and gaming, it could be newer gaming startups, it could be new startups in remote work that, you know, I think remote work itself is growing and therefore there will be someone that targets just remote work. It could be something in crypto. So I think that based on interest group, we'll start seeing a lot more.

So the interesting thing with consumer social is we've seen a bunch blow up and then not sustain. So Meerkat Houseparty two separate apps, same company HQ trivia. Presumably you looked at a bunch of these at the A or the B and they were done by the world's best firms. What determines whether they sustain versus sort of flash in the plan, plant in the pan? You know, TikTok is another example of it's blowing up right now, but is it going to be, you know, musically was an example too. Is it around forever? Is the growth going to continue? How do you evaluate that for consumers?

Yeah, so I think in terms of one of the, in terms of like early days growth and metrics, obviously you pay attention to in anything social network you have to understand, you know, daily activity and weekly activity of the user and how that's trending and the retention. Right. Retention is usually a pretty good indicator of whether people are sticking with it or not. Now some apps struggled with it, they had very good user growth. But when you looked at daily activity retention of their earliest cohorts, it wasn't as strong. But the other element, it's not just on Daus, is are you launching it on the back of a platform? If your app is not the destination app and you didn't grow as a destination app app, the odds of it becoming a destination app is far less. And that was one of the challenges MeetKit had. It was launched on Twitter and the day Twitter shut it down, it dramatically dropped. This is not a new problem. We've had this. I mean look at Zynga on Facebook, right? So it's not a new phenomenon. It has existed for years. And so I think it's very important to pay attention to that, which is are you able to build a destination app from very early on? And the second thing I think is I am a bigger believer of utility. I think if we compete for just time, there are a lot of other apps that already have time. Like people are on Instagram for a long time looking at others photographs, people are on Facebook to stalk other friends. And I think if you're playing the attention game, it may be harder but if you have a true utility. So I'll give you an example of a recent YC company snack package. Now people would think, oh, one more food app, like do we really need one more food app? But they were extremely smart in the way they launched. Their frustration was they were all from Yale and when they went to university, in spite of all these food delivery apps, they had to wait in line for such a long time, every time during lunch, which was a one and a half hour break to get food and they said, well, no one has solved our problem. So they literally targeted universities and launched order ahead because, you know, delivery is still pretty expensive, whether it's Uber Eats or Doordash or any of them. And so not everyone can afford it. And probably as students, you're more price sensitive. And so they said, all we need is we can order when we are in the last class, so we can just go pick it up. Right. So that was literally the genesis. But if you look at the product and say, oh, it's just one other order ahead, you'll be missing the point because now it's turned into this enormous micro social network around gifts within universities. So if you want to send a gift because one of your classmates helped you, you can just send a snack pass and they can pick up tea in their favorite location. If you want to play a Zynga kind of game on the app with other friends, like hatch a chicken, you can do that on the app. Right. So you need to have a micro community with a utility. So I'm a bigger believer that if you target something with a utility that is not just competing for attention on time, and attention for time is just a layer on it just helps strengthen the network.

How would you evaluate a company going back to Marketplace for a second, a company like Thumbtack at the A or the B? And I think that's an interesting example because my understanding of the company is it grew super fast on the back of Google, then I believe had some issue and then raised another strong round for Sequoia is now on the upswing again. But the extent that you're familiar with and it's a horizontal marketplace as opposed to a vertical one, which is more in vogue now. How would you have evaluated a company like Thumbtack and whether it would have made sense for you at the A or the B?

Yeah. So in the case of Thumbtack, they did rely on SEO search. So I would say it's a little different from launching like on Twitter, like how Meerkat did. And the way to look at it is at the end of the day, do you want a marketplace for services? The absolute answer is yes. Right. It was so hard to find a plumber. It was so hard to find a mechanic. It's so hard. And these are not high velocity use cases. So if I found a plumber, the chances are I'd use the same plumber again, but maybe once in two years or three years. Right. And so. But I may need another service. So I think Marco's articulation of that was really good in the A and the B, which was we need a horizontal platform. I'm not building A services for just Plumber. And so their articulation from the day one was I'm going zip code by zip code. So if I start with San Francisco, I have to build all services that people in San Francisco would ever need. And they quickly figured out that the quality of service would be measured by the number of responses to their leads. So they tracked something like, are we able to generate at least three good responses for one query? So if I looked for a plumber, am I at least able to send anu three good plumbers in the area with enough reviews and ratings? So that liquidity is very helpful and probably proprietary to thumbtack and that is enough to overcome the risk of the dependency on Google SEO.

Yeah. Or that they will be unbundled from, you know, people building vertical solutions.

Yes. And I think that articulation that, yeah, I'm not just going to go for an app, I'm not going to go to an app for Plumber, different app for a mechanic, a different app for some renovation versus I'm going to a horizontal platform. And they're actually building enough insights and data and over time they'll also know pricing information. And I think they've launched an app. Right. They launched an app more than two or three years ago, but they knew that they had to launch an app. So therefore, if you have strong assets that are proprietary to you, but you're using SEO as a way to scale, that's totally fine.

Yeah. How about SaaS companies or enterprise companies? Where are you excited? Where is the white spacer opportunity and how are you evaluating these businesses? For a long time, I think a million ARR was the Series A Metro imputer that stayed the same or changed or I you think about it, I.

Feel like the millionaire ARR is probably still the Series a, but the 2 million ARR has become the B and the 3 million has become the C. So we're definitely seeing, you know, rounds happening sooner. I think this is largely because cloud took off. You know, we've been, I mean, as VCs probably feel we've been talking about cloud for a while, but in reality the mass adoption has happened only in recent years. And therefore with that adoption, people need a lot more B2B tools and the large companies weren't prepared for it. So therefore, I think for our B2B, from an investor standpoint, what we look for, evaluation is really at the highest level, obviously. What is your KPI, right. Usually it's revenue, but we want to understand is it recurring revenue? People use the word ARR, but often enough it's not recurring. So it's super important to distinguish between recurring revenue versus services revenue or one time revenue. And then you. The other important metric. I think there are a bunch of SaaS metrics that have been discussed before, so I won't repeat the same like the magic quotient or the ratio, the sales efficiency. But one of the things I really like to look for is how much are your existing customers expanding. That's usually a strength of your business and that translates to your net dollar expansion rate. So meaning if you didn't add, you know, you would have added some people. But like I like to look at your existing customers, how much on a quarterly basis are they expanding? And there's an equivalent metric that Twilio actually started including it in their 10k called the net dollar expansion rate where they measure that and if it's like usually greater than 140% on a quarterly basis, that's top quartile. That means you not only landed, but you're expanding so fast and companies really need you with it. So the switching cost is another measure to figure out. How sticky is this offering?

Yeah, how about hardware companies? Where are you excited? People have said hardware is hard for a long time. What are the metrics or things unique that you're evaluating from hardware perspective?

Yeah, hardware companies are really hard. I think it's not easy. Again, it's not a zero sum game, but there is a lot of competition. If hardware is your niche, it could easily get copied. Right. And there are competitors, not just from the U.S. but people who have incredible supply chain experience like in China or Taiwan. And therefore many companies approach hardware with the software component. And there it's important to understand what is your software moat. So one company from the YC portfolio recently that's been doing extremely well is Flock Safety. And they basically provide camera sensors in zip codes in dense locations and they work with police and city councils to help catch crime. So their moat is actually the software layers, which is there is a network effect in that if 10 counties in say Georgia are using flock safety, the 11th county also wants Flock Safety because if someone's committed crime and you're using the license plate of the car, you need all the counties to use the same software to be able to track the driver out and figure out what happened. Right. So and then they also have, you know, over time they'll have data on behavior, what sort of signals, red flag behavior versus normal behavior. So I think if you are going down the hardware path, it's important to focus very early on what is the software layer and what's your moat. And so just the hardware device. As investors, I don't think we would value that much the premium value of the pure hardware sale. It really comes down to what is your software mode and therefore what's the software software revenue you can derive.

Yeah. How about e Commerce or D2C brands?

Yes. So E Commerce and D2C brands. So the highest level, they're selling goods. So it's revenue. Right. And then you have all the cost of goods sold as well as the shipping and logistics to get to what your gross margin is. The number one thing there is organic growth because we've seen enough E commerce companies step up at 50 million, at 100 million.

And whether just because they become commodities.

Or it's hard if you've relied on paid marketing and it's going back to that niche community.

We're seeing Casper in the news right now.

Yes. Or even beauty. Beauty. I mean if you look at E commerce penetration, actually what's interesting is a lot of people think oh, everything is E commerce. No, actually media, electronics, books are all E commerce. But fashion beauty is also like very thinly penetrated. Even grocery. Right. Grocery is like 3% online. So there are lots of categories that don't have high e commerce penetration. And I think if you're trying to reach these niche users or users who have the willingness to do that purchase online and you sort of tap out after a certain point, the customer acquisition cost doesn't make sense relative to the marginality. And plus they're not high velocity products. You know, it's like, do I use those products every day? Yes. But how often do I buy them? Maybe quarterly. Right. So you're not top of mind. And therefore I think that if you're doing the pure E commerce route, you have to ask yourself how do I get in the high velocity use case purchase? Or I think you have, you have this unique opportunity as I mentioned earlier, to form vertical communities. I think that is something in beauty. Like you buy products like we all buy makeup or any of those products by getting recommendations from our other girlfriends. The problem with pure E commerce is I don't get that recommendation. So I'm not sure that I want to buy that product. And I think glossier is really changing that. And so I think this is where approaching vertical commerce in a way which reflects buying patterns which is similar to what pinduodor did, did by the way in China it was literally people think oh it's group bargain, group buying and therefore it translates to low cost. Of course that's the benefit. But if you look at markets like China and India there's a lot of trust deficit in even buying regular day to day products because we never had brands growing up. I never had a, there was no Costco, there was no Best Buy in India. I couldn't just walk into a store and rely that the produce was actually good. So you would ask your friends, your neighbors hey where can I get mangoes? Right. And that's literally what Pinduodua did. It said why don't you know, if all five of you as neighbors are willing to buy it must be good. And then the sixth time and it was self initiated. Right. By one of the neighbors. So you trust them invariably. So therefore I think in vertical commerce it would be important for any E commerce to reflect the buying behavior and not just just focus on selling the product.

Yeah, let's actually segue into mental models for investing in China, investing in India, even Pindodo. I'm curious why, why how you think about it from an Investor you previously a 16Z you guys thought about it a lot there. YZ is also obviously focused on global investing. How do you think about it?

I think that in 10 years from now the billion dollar company concentration would probably be roughly divided between India and China with sorry between US and China and with India being a close third.

Wow.

And then Latam being the fourth.

And so is it literally just hey what's worked in the US it's going to like a rocket Internet model or are the dynamics going to be fundamentally different in terms of what types of companies are merging?

It is going to be fundamentally different and I think the learning is going to be from both sides. I think gone are the days where we can sit around and say oh yeah, the Amazon of India is launching in India or the Google of India was Google, the Facebook of India was Facebook. I don't think the future innovation models will be from the U.S. it'll be local. And so if you look at the recent large companies out of China, Pinduoduo, bytedance with Todiao and TikTok and Douyin, they are fundamentally different and don't look anything like any US company. And that's because culturally it's a different place and the buying behavior is just very different. As I mentioned Pinduoduo, they literally tapped into the group buying behavior because there is a lot of trust deficit in the market. These are not markets with 90% brands like you US. These are markets which are 90% unorganized. Right. And therefore it's first, how do I build trust? And two, how do I initiate consumers to be both the seller and the buyer. Right. And so now Pinduoduo in its scale also has manufacturers listing and incentivizing people to start groups. But when they started, it was purely mom and pop sellers. So that's one touti ao. You know, we had. They had a Twitter equivalent in China that is WeChat. So if you look at the US you'd go, well, we have Facebook and we have Twitter. Why was there a need for Totia? Well, it literally started with, you know, the founders of Totia actually were inspired by Reddit and it started with the Reddit concept, which is how do we make sure we get the most relevant content locally to people quickly? And then they shifted to video very quickly because they realized that people wanted to consume more information in the format of video. And within four years, it's an amazing story. Within four years, people were spending close to 76 minutes daily on todia. And so we don't have a model like that in the US And I'm not saying we need that, but I think it's just important to understand that the cultural nuances and the needs of the user are not the same. So therefore, as an investor, you have to understand the local market.

Right. But China is so unique in its sort of insistence on internal companies and relationships with governments and the companies and sort of the uniqueness of the culture. And WeChat, of course, is super. Is India going to look more like China or more like the US in terms of global versus local giants? And, you know, is India going to have its own version of WeChat or how do you see?

No, I think WhatsApp is already India's WeChat. Right. It is the main messaging medium. I think a lot. I mean, if you look at similarities in market condition. When I say market condition, this is more about shipping infrastructure, logistics, ability, access for payment, infrastructure. I'd say India is a lot closer to China. Not necessarily in government or policies, we're very different. But in terms of the actual infrastructure that would enable the Internet economy, India is closer to China, behind China. So if you look at the GDP per capita, India is way behind China. So if people are expecting the next Alibaba to happen in India soon, it's not going to happen soon. It will happen, but it'll take time, right? Because the GDP per capita is behind. But if you look at the shipping logistics, like, you know, there is a famous quote from Jeff Bezos from many years ago where he said, yeah, when I first launched Amazon and wanted to ship a book. He talks about the story of where he personally drove to the UPS and put it in the postal box and it went right in India. In most places in zip codes, you can't just go to the post office and ship it, you know, more often than not to. If you want to guarantee a same day or a two day delivery service, you need to have local infrastructure that you own. Right. So these companies not only just enabled E commerce, they had to innovate the local logistics, shipping infrastructure to meet the demands of the consumer. So therefore, if you look at opportunities in India, both in consumer and B2B, I think it'll be more like China. Especially in consumer, it's going to look more like China in B2B. I think India is taking a different lens. There is a lot of appetite for B2B tools, but Indian founders that are working on B2B offerings want the company to be global. Look at Freshdesk, it's global. Look at Mu Sigma, started in Bangalore, but serves a lot of customers in the US and so I think in the next 10 years you'll see a lot more B2B companies in India that are global.

Lastly, you have a few talks and a bunch of writings on network effects. And so want to tell listeners to check that out, but maybe to give a preview. What is sort of the debate when you talk to James Currier or Mike Maples or Jeff Jordan or people who really understand network effects and you agree on so much, but what's something you have a different take on or unique take or think the mainstream sort of doesn't fully appreciate as it relates to understanding network effects?

Yeah. There is a narrative right now that maybe network effects are diminishing. And so how do I to put that in context? Right. So if you look at Facebook and Google, they were pretty much winner take all globally, except China. Right. But you look at the next set of companies, Uber, Didi. So take ride sharing. Fragmented. There are big companies, but they're fragmented. They're by no means, they are winner take all. Look at food delivery, same. You see they're not winner take all. I'd say even in vacation rentals probably surprise people more than not, which is we have Airbnb, we also have booking.com. right. And so there is this question going on around, hey, are network effects diminishing I actually don't think it's diminishing. I actually think it's going to come back more powerful than it ever did. Because of this concept that I just explained before. The first one being Sarnoff Broadcast and Metcalf being point to point. I think we have seen very few startups exploit the Reed's Law and I think we're going to see more. And Reed's Law is one of the most powerful laws in the value of the network. And so I do think there is a lot more potential and a lot more companies. They won't look like Facebook and Google because it's not horizontal, but they will still be global for the interest groups that they are found.

On that note of optimism in terms of a lot more opportunity left, we'll close this podcast. My guest today has been Anu Hariharan. Anu, thank you so much for coming on the podcast.

It's been great. Thank you for having me.

If you're an early stage entrepreneur, we'd love to hear from you. Please hit us up at VillageGlobal VC network Catalyst SA.