a16z Podcast - a16z Podcast: Cash, Growth, and CEO ❤️ CFO

Episode Date: August 25, 2017

with Ben Horowitz, Scott Kupor, and Caroline Moon “The only unforgivable sin in business is to run out of cash” [so said Harold Geneen], yet startup CEOs “always act on leading indicators of good news, and lagging indicators of bad news” [according to Andy Grove]; after all, it requires a certain stubborn, headstrong optimism to start a company. So how to reconcile these views? At the very least, pay more attention to leading indicators of running out of cash, “because there’s just no going back”! But doing all this — while also trying to balance growth, advance planning vs. constantly changing strategy, revenue vs. margin, coordination/communication/culture, and so on — is a lot harder than it seems on a finance spreadsheet. It requires understanding that the “math is not the terrain, the spreadsheet is not the business”… yet also knowing when to trade rose-colored glasses for darker rainy-day ones. And that's where a CEO partnering productively with a CFO comes in. In this episode of the a16z Podcast -- moderated by (and based on an internal event for CEOs+CFOs hosted by) Caroline Moon, who leads the financial operations for startups practice on a16z's corp dev team -- Ben Horowitz and Scott Kupor share their personal insights as well as advice for founders: How DO you do it all?

Hi everyone and welcome to the A16Z podcast. Today we're talking about how CEOs and CFOs work together productively and balance it all from cash and growth to advance planning and changing strategy. The conversation is based on an event from last year, which was moderated by A16Z partner Caroline Moon, who was a CFO at a number of startups before joining the firm a few years ago to run the financial operations practice on A16Z's corporate development team. The discussion starts off with Ben Horowitz talking about his classic blog post on cash flow and destiny, and Scott Cooper joins in to provide a perspective from the finance and operational side of what it takes. The topic today centers around what founders and CEOs can and should do to drive their

companies for profitability and things that CEOs deal with on a day-to-day basis. Ben, I was reading one of your blog posts, cash flow and destiny. And when you look back on that period when everyone, yourselves and your competitors included, were going big and not going home, you know, if you could do things over again, what would you have done differently? And I guess more importantly, could you have done things differently given what everyone around you was doing? Well, there's kind of two things that were very deceptive, at least to me as a founder CEO at that time. One is when money is easy to come by, it's hard to visualize what it's going to be like if it's ever hard to come by. And the dangerous thing about that is you don't actually own your strategy until you're generating

money. And that's what I meant by cash flow and destiny was. strategy for how much money you want to spend, what you want to do, like, what direction you want to go in, you know, how fast you want to expand your market opportunity. You may not think it affects your thinking. It absolutely affects your thinking. And it didn't really strike me as how corrupting that was to my thinking until we start to generate cash and I, like, could set my own strategy and do what I wanted to. And that freedom is worth a lot more than almost any incremental money I spent. So like that was probably big lesson number one. And then the second thing, which is so obvious when you think about it, but you can get carried away is,

look, more money doesn't mean more growth. It can mean more growth. It's not like you can grow without spending money, but there are limits. And then it's different by department. If you more than double engineering in a year, that gets pretty hard depending on what you're doing. Because a communication overhead. The coordination is so intense in engineering. Sales is much more scalable, but if you scale sales without keeping very close track of productivity per rep, that can just be a total waste of money. That happens all the time. So if I had to go back and play it again, I would have thought more when I plan growth starting with, well, what are the limits of my ability to operate this company and keep communication where it needs to be and

know what people are doing and so forth. Whereas what I was doing was going, well, here's our goal. Get the team together. What is it going to take you to hit that number of that kind of process? Don't do that. Don't do that because it sends actually the wrong message to the team when you give them an unconstrained like resource to do whatever they want. The flip side of the bubble, of course, was the crash no one could have foreseen that. And certainly, nobody could have predicted 9-11. Those are really extremely exogenous events and hardly something a CFO could really plan for or CEO could plan for. But with that said, what sort of advice do you have on that front? Is it possible to actually expect the unexpected? I don't know that you can plan for

wax one events by definition, but I would say there's a couple things you can do. One is if you start from the concept of, look, we need some constraints in the system, period. So we may have raised $100 million, but maybe the answer is, look, can we do what we say on $75 million? Even if you end up letting go with a pressure valve over time, I think putting the pressure valve in there around the budgeting process and forcing people to actually come back to you and feel like you're turning the screws on them before you actually give the go ahead on some of those costs. The second thing is there were always signs. Nobody could foresee the crash, but there were precursors to problems. I don't think we understood and appreciated the exposure we had. We could have said, okay, where are our dependencies

from a customer perspective and tried to think about, okay, if the environment changes, what does that mean for the life of our customers. Therefore, not only can we not raise capital, but our revenue that we thought had goes away. Those are things, again, which seem obvious now in retrospect, and it's worth thinking about in the context and doing some planning around, you know, if things dramatically change there, would there be knock-on effects to your business that you might not have otherwise anticipated? So that actually takes me to this notion of building a company vision, really putting it down on paper, and then going through an annual planning process. So when do you guys think that company should really go through that process of putting it down on, on sort of codifying

it on paper and then going through a formal plan. The easy way, at least, is think about what's the story you're going to want to tell investors of the next fundraise? I think it's a very useful model for how to think about kind of what is the, what the goals do you want to articulate? The day you close that series B financing, the only way you can actually have a rational strategy for what you're going to do for the next stage in 24 months is what your assumption is on when or if you're going to get that series C financing or if you're going to actually

be cash flow positive at that point in time. And so I think you have to kind of, you know, you have to almost work backwards from there, and that becomes your guidepost. And then you can say, great, this is the amount of money we can spend. You've got to have some guidepost and the most logical guideposts that next round of financing or a point where you actually get to cash flow positive. Then it's obviously a question of course correcting all the time, you know, if you're on or off plan. The thing about plans change. I mean, how many you guys have like a plan that lasts a whole year? And there's two different things in that. One is the strategy. It's not the way like big companies do it or management

consultants tell you, right? Like the way that you all do strategy, I'm quite sure is you think about the company all the fucking time. And so at some point, you've articulated to the employees and whatever, like, this is where we're going. But every day, you're going, oh, that's a data point. Competitor is doing this. Oh, this new, like, chip came out. This is happening.

And you're modifying the strategy. And what most CEOs do is they never actually kind of fully articulate out the updated strategy. So a very important discipline is to, at least, at least once a quarter, just take whatever your old strategy is and re-articulate it with all the new things that you've added in your mind. And you've told your staff probably, but they haven't told their people. Just to keep people on the same page, then how do you relate that back to the plan? Generally like the more formal, the better.

In the old days when I used to work on compilers, it used to be like this two-pass compiler was the thing that you had to do because you couldn't actually get all the optimizations done on the first pass. And I don't know if it still works that way, but budgeting is very much like a two-pass compile. So what you want to do is have like a very constrained thing with like the goals that you want and run everybody through it and then like see how much they burst into tears and like where their problems really are. And then do the second pass saying, okay, team, this is what you got. This is what I want.

And having that negotiation, there's certain things that are good to be formal on budgeting, raises, stuff like that, really early in your history. I think we all know in this room, in particular, that founding and running company is incredibly difficult work, and it takes a tremendous amount of optimism. So, you know, the ability to kind of wear rose-colored glasses, if you will, and yet there are time for things don't go well, right? And so how do you know when to take those rose-colored glasses off? I think human beings tend to fall into that sunk-cost fallacy quite often, and you just say,

oh, I just need one more quarter, just one more quarter. And then suddenly, like, a whole year has gone by. So, you know, what are some things that you can share with us in terms of your learnings on that front? And Andy Grove had the best line on this. A bunch of CEOs got fired right in a row. And I was like, Andy, like, why did they all get fired? And he said, well, because like CEOs only listening to leading indicators of good news, never leading indicators of bad news, only lagging indicators of bad news. And I was like, well, why is that?

He's like, well, because what you take on to go build a company is like ridiculous. It's impossible. So you have to be an optimist to go think that you're going to go build a company. that's going to take over the world. Of course, you're not going to look at leading indicators of bad news. You would never built the company in the first place. The thing that I would say is that the one leading indicator of bad news, or there's a bad news thing that happens in the macro environment when prices reset, that you just have to, you actually are better off adjusting immediately to that. Peter Drucker said one of the greatest lines ever, which was

the only unforgivable sin in business is running out of cash. When it comes to leading indicators of running out of cash, I think you do have to pay more attention to that because there's just no going back. And like we definitely would not have survived as a company if I hadn't just admitted to myself that things might not get better in the kind of funding environment. So when we went public, if you go back and read the articles, they all say, these guys must be desperate. They're going public at the bottom of the market. And at that time, the NASDAQ was 2000. And the bottom of the NASDAQ was like 1,200. So like we weren't at the way. It had fallen from 5,000. So like 5,000 to 2,000. And they're like, they're going out at the bottom.

They must be desperate. And I was like, well, one, I am desperate. And two, this is not the bottom. And that was probably the single best decision I've ever made just to go like, how do you know you at the bottom? You're not sure at the bottom. What's some advice you have on how do you as a CFO, work with a CEO who is by nature just a tremendous optimist, you know, and I'd like to think that, you know, I think we got a bad rap as finance people that were Debbie Downers, but the end of the day, we're realists. And so how do you preserve the relationship you have? Because it is very much a partnership, you know, by dousing some stuff in realism and having those difficult conversations, but still, you know, sort of handling and managing the CEO's

personality. I think actually the mistake that a lot of people in this, in, in the finance will make is to kind of be, look, I've got to always, like, look at everything from the opposite of, you know, rose-colored glasses. Instead of having, instead of, like, coming in with that negative argument, I think it's better to say, okay, look, here's what you just told me you want to do, right? So let's just lay out, let's lay out in, you know, math numbers, like, what are the assumptions you have to believe in order to get to that plan? And then you can, then we can at least have a rational argument around whether we think any of those assumptions are real or not, right? I think it's a bad idea to go in and say, look, you know, you're

fucked, you're wrong, like, this doesn't make any sense. If you figure out a rational way, to say, okay, look, if you think we're going to do $100 million next year, that means we've got to have this number of sales reps at this period of time doing this amount of quota. And it takes them six months to ramp. And you just kind of walk through those. And then at least you can have a rational discussion around, okay, do I believe that or not? Or maybe the answer is, look, it should be willful suspension and disbelief and that's okay. there was a great actually metaphor for this you guys remember the group the time where prince wrote all the music there in purple rain and there was like morris day and then there was jerome and morris day would do his dance and jerome would hold up the mirror and dance with him like that's kind of like what you want to do is like hold up the mirror like make them look in the fucking mirror because ceo is like as ceo is start like you're used one like you have better knowledge of the whole operation than anybody so like if somebody if cfo comes to challenge you

like, if anybody challenges you on anything in the company, first thing, it's like, well, you don't know what you're talking about. Like, my CFO, I'd be like, you know, he'd show me the spreadsheet. And I'd go, the map is not the terrain. The spreadsheet is not the business. Let me tell you what's inside that fucking spreadsheet that you don't know about. What happens with most finance people is they back down at that point because they're like, well, fuck, he does know more than I do, even if he's wrong.

Whereas going, okay, like, let's just draw out. Like, here's all the things you want to do. Like, this is what it means. Like, look at this. That, I totally agree. that's a much more effective way to manage a CEO is to show him what he's doing. It's supposed to tell him he's wrong. So you've used the analogy before in the past about HR being QA for a company.

I like to think of finances sort of the scrum master of the company. We don't necessarily know how to do it, but we know what things are running on time and to course correct when there are sort of things that change, whether it's good or a bad. When should a company really think about hiring a CFO? Like, when are they ready? Silicon Valley went from this hire a CFO immediately. the late 90s to the era of the lean startup where you, you know, you just outsource finance forever. Look, I think as soon as you start spending significant money or getting like significant revenue,

it gets to be dangerous not to have a CFO. So like if you're increasing headcount at any kind of like fast rate, or if you're generating revenue, particularly if it's something where there's real like legal liabilities of not having proper controls structure and so forth, that becomes very important. We found in 100% of the cases that, having done it, they all kind of now say, gee, like, had I done that six or 12 months earlier, it would been better. It could be a VP of finance, right? But somebody who is that kind of business partner, I think the other point is right from a timing perspective is if all you're doing is, you know, you're hiring engineers, you're doing product

development, you know, it's, you can probably, you know, you're, you need to just know, like, do I have enough cash to make payroll and stuff like that? And probably you can outsource that kind of stuff to, you know, an outsource firm. But as soon as you think about go to market, right? And you start to think, okay, now I'm going to spend marketing dollars. Do I understand what my lead generation costs? Now I'm going to hire sales. What's the throughput and conversion rate and quota and all that stuff? I think then you've got enough moving parts where you're probably doing yourself a disservice as a CEO. If you don't have someone whose job it is to actually put all that together and say, okay, you know, if we spend this amount of marketing dollars, it ought to

yield this amount of lead generation, which means I'd need X number of sales to go prosecute those efforts. Who the hell is going to keep an eye on the head of sales if you don't have a CFO? Great. We'll turn it over to all of you. And I do have some questions that were pre-asked. Assuming you can control net burn perfectly, how do you think about determining the ideal burn rate and runway? What factors go into determining the Goldilocks-style middle round? That is the really hard question to answer. Part of the challenges you're tilting into a bunch of unknowns. Like, what are the capital markets going to be like at the next round?

How are we actually going to perform against the plan? So it would be a hard thing to do if you knew if all those things were set, and then with all of them moving and having variance, it's even more complicated. but you 100% have to get to the next milestone. A really good way to think about it is, is this starting to look like a good business? If we just continue growing like this, given the track record,

this is going to be a business that's worth way more than it is now. You've got to get to something you can scale before you try and scale it. That's the kind of, the big mistake in the 90s was all of us would go like, oh, we've got something that people want to buy, like, let's build a gigantic sales force and just blow it out. And then a lot of people didn't quite have the total product market fit in the company got too big and it just collapsed.

It's the fallacy between a dollar reduction in cost spend and a dollar reduction in revenue growth. There may be some companies for which that is in fact true, but maybe you're just spending money where it just doesn't yield anything or you've just gotten fat in other areas just because the capital was there. There are probably lots of things you can do to have a rational plan plan that doesn't actually, you know, literally take a dollar out of revenue growth. How do you think about revenue versus margin? What's interplay? between these two priorities and what should be the arc of priorities over time? When I was public, I used to have this conversation with investors, and they're like,

you gave us two pennies and earnings. Why don't we get a nickel? And I'm like, you don't get a nickel because I'm trying to build it so I can give you a dollar. So like, stop asking me about a fucking nickel. You're not going to get a nickel, sell your stock if you want a nickel. Like you're going to, I'm trying to get a dollar. So there's that where at a certain size of revenue, if you're generating like 20 million in revenue, like generating a million in cash, doesn't make your money.

your company very valuable, a million in cash, you know, times whatever earnings of 40, you're worth $40 million, like with a great earnings multiple. It's just not. So, like, you can't over-focus on margin. And that's why we talk a lot about like cash flow break even as a big thing just because it means you don't have to raise money. But margin against incremental revenue is more complicated. Having said that, if you hire more vice presidents, what does that vice president do? Hires people and builds a big-ass organization because, like, that's what they do. just have to be conscious of culturally, are you building something that is efficient, productive, or are people stepping on each other where you carve it up into teeny, teeny, tiny pieces like

you're Henry Ford and you're running the Model T. People hated the Model T job so much that Henry Ford literally doubled minimum wage overnight because his attrition rate on the assembly line was so fucking high. But to me, it's like, how can we have a really kick-ass company where everybody's stretching themselves and everybody's, you know, and like when we're asking a lot of the managers to get a lot done with a little money, and then we push as hard as we can into the market.

You've got to be honest with yourself in establishing gross margin and operating margin, right? And so, right, you know, gross margin has lots of accounting definitions, but the simple way to think about gross margin is like at the unit level, does the business actually work, right? So, like, for every incremental, like,

piece of output, do we actually generate profit? And if you think about it, right, operating margin, in theory is, it's things like sales and marketing and stuff. And right, so, look, you could argue, I might grow faster or slower generating two cents for a dollar. Maybe it's important for me to go to APEC because that actually builds a better business. And so in theory, in an economic downturn, those things are easier to change.

Going back to our Loud Cloud days, right? Like, you know, we had this concept of unit economics where, you know, like a data center had to be able to actually generate profitability. And the model worked well on paper, but when we actually put it in practice, you realize, like, you know, you don't get the efficiencies you thought. I think that's the real important distinction, at least on margin, is like, you got to, like, you got to hold yourself to a gross margin and an operating margin. What Amazon is doing, I think it's a lot the right approach, which is you have one business

and you get it working and it generates money and then it can generate enough money for you to add another business, but you don't like have one business and it's not making money and you had a second business and it's not making money. Google very early on in their history had a product that just spit out money. And so Larry, like I think, you know, was smart about it, said, like, I just want to own like every dope engineer in Silicon Valley is going to work for me. And that worked for them, but they had the monopoly. Like, you have the monopoly, like, that's a good method. If you don't have the monopoly, it gets you into trouble. My experience with engineering has always been that if you freeze engineering

hiring for a year, engineering gets way better. Integrating new engineers is super time-constraining. And then interviewing, for engineers to interview people, saps their energy. Hiring just slows you down. Facebook has maintained very high product velocity at like a pretty high scale. And I think to me, the thing that they do better than anybody else is onboarding and training of engineers like that is their boot camp. Commit to not hiring, but onboarding, training, getting to productivity and measuring that you've gotten there, then you grow much better. How do you think about building the biggest and best eventual business?

business versus demonstrating the traction along the way so you get to keep fighting. Well, one, if you want to build the best and greatest business, then generate your own cash. And then I would just also add to that, keep in mind that more money doesn't necessarily mean bigger and better vision. It can actually mean smaller and worse vision and bankruptcy. Some of it is like, God, I fucking hate not knowing if we're going to survive, can we get faster to the destination? Can I like just get there already? But you've got to resist that. You have to kind of walk a balance line.

Thanks, everyone, for coming. We really appreciate having you here with us, and we'll hope to see you again soon.