Deep Dive with Jeff Epstein

Intro: Welcome to The Daily Bolster. Each day we welcome transformational executives to share their real- world experiences and practical advice about scaling yourself, your team, and your business.

Matt Blumberg: Welcome to The Daily Bolster. I'm Matt Blumberg, co- founder and CEO of Bolster. And today we are going in deep with Jeff Epstein. Jeff is an operating partner at Bessemer Venture Partners. He's the former chief financial officer of lots of companies and public companies including Oracle. I was fortunate enough to have Jeff on my board at my last company Return Path for about eight years. And Jeff, it's great to have you here today. Thanks for joining.

Jeff Epstein: Matt, great to see you, and thank you for inviting me.

Matt Blumberg: So I always love to start these in- depth ones with a quick career journey or arc of your career. So obviously, I noted a couple places you've been CFO, but give everyone a sense of where you started and where you've ended and the big steps along the way.

Jeff Epstein: I went to business school in Stanford and then ended up joining an investment bank. It was then First Boston, it's now Credit Suisse doing mergers and acquisitions in the 1980s in the media industry. So I was representing buyers and sellers of television stations and newspapers. We represented A, B, C in the merger with Capital Cities Communications. And I love the media industry and I love learning about finance. And then one of my clients was a television programming company called King World. They had Wheel of Fortune, Jeopardy, Oprah Winfrey, so big shows but a pretty small company publicly held. They had an accounting CFO, and they wanted a Wall Street CFO. And I thought, "Do I want to spend my career as an investment banker, or would I like to move into the CFO side?" I looked around at the investment bankers that I knew, and they were all very good salespeople. And my strength was being an analyst. I think I was an okay salesperson, but not nearly as good as them. And so I said maybe I should go from the sell side to be a buyer of services where the CFO gets to hire the bankers instead of having to persuade to hire you. And so it's a great decision. I was 32 years old, became the CFO of a New York Stock Exchange company, and that was the beginning of a 25- year career as CFO at different companies.

Matt Blumberg: So you were Advo, DoubleClick, Oracle. What was -

Jeff Epstein: Not Advo. It was King World.

Matt Blumberg: Oh, King World. Okay.

Jeff Epstein: Then I became the CFO of DoubleClick during the internet advertising boom. The first internet advertising ad ran in 1995, and I joined DoubleClick in 1998. So the industry was tiny. My friends and I all felt that we were at the beginning of a revolution in advertising. And in fact, today internet advertising is far bigger than television or any other medium. And we were absolutely right there. And DoubleClick became the largest internet advertising technology company. It was sold to Google and now the Justice Department is trying to get Google to sell it. So it comes full circle. And then I became the CFO of Nielsen Media Research, division of Nielsen. And we had Nielsen net ratings and internet ratings business, and then the CFO of Oracle.

Matt Blumberg: Got it. And along the way, I know you've been a board member of lots of companies, some private, some small, some large and public. What are some of the more notable ones?

Jeff Epstein: So I joined the board of Priceline in 2003, and I served on the board for 16 years. We acquired booking. com and we changed the name to Booking Holdings. It became the largest hotel travel agent in the world. And the market value went from 1 billion to 80 billion over those 16 years. So pretty extraordinary. And what was fascinating was Booking was a very decentralized business, Booking, Priceline, Kayak, Agoda, OpenTable, all separate businesses, all run independently. Oracle was a very centralized business. So while I was the CFO of Oracle on the extreme of centralization, I was also a board member at Booking on the extreme of decentralization. And I saw two very well- run companies with two very different tactics of how to organize large enterprises, both of which were successful. So fascinating view.

Matt Blumberg: Yeah. It's interesting. A lot of execs and CEOs will ask me, " Hey, do you think I should take an outside board seat? I'm really busy with my day job." And my answer to them is always yes that yeah, you come away from someone else's board meeting with a lot of learnings and a new way of looking at your own business. And I always think it's really shortsighted of companies when they have policies that say their people can't serve on outside boards for that exact reason.

Jeff Epstein: I think it always may be a better CFO to be a chair of an audit committee, another company, and it may be a better board member to be a CFO. So it's very symbiotic.

Matt Blumberg: Yeah. So let me take a quick detour on one thing. I knew you had been on the board of Priceline/ Bookings for a long time. I didn't realize it was 16 years. What do you think about board of tenure? Did you feel like you were still as effective at year 16 as you were early on when maybe you had fresher eyes?

Jeff Epstein: Well, I can see the argument that you bring a certain point of view to a board and your world view doesn't change very much. And so if you bring in a new board member, and to bring in a new word member, you have to kick the old one out to make room for them, that over time, that's a good strategy. On the other hand, to take the extreme case, I think activists were trying to get Warren Buffett to leave the Coca- Cola board. If I had Warren Buffett on my board, I would never kick him off. I mean, one of the smartest people in the world, the best investor in the history of the world, why would you ever want to have that person leave your board? So I think, in concept, the idea of board refreshment is a good idea, but I think if you have a terrific board member who's uniquely talented, you certainly would want to keep them as long as you could.

Matt Blumberg: Yeah. That's a fair point. And presumably, good boards have some board members of long tenure and some that do three, four, five years and roll off. So good to have a mix. And if I remember correctly, there were some other people on that Booking board that were there for a very long time with you as well.

Jeff Epstein: We ended up putting in a board refreshment policy and I don't remember what... We didn't have it at the beginning, but at some point, I think it was maybe 10 years or 12 years, something like that.

Matt Blumberg: Yeah. Well, so let's back up and talk a little bit about the CFO role. One of the things that I always thought was so interesting about your journey is that you operated as a CFO in some really different businesses. King World, very different from DoubleClick, obviously, some things in common there, but pretty different. And then Oracle was completely different and even Nielsen was a bit different as well. And I know there's a conventional wisdom is that CFO is one of those roles, maybe HR is similar or legal is similar, where it's maybe easier to switch industries than if you're in product or if you're in sales. So I'd love to hear you talk about that for a minute. Did you find those transitions easy or not easy? How did you go about ramping up on a completely new industry when you come in at near the top of the organization?

Jeff Epstein: Warren Buffett has this concept of a circle of competence. He says you should invest within your circle of competence. And of course, why would anyone invest outside your circle of competence? And I think of the same thing in terms of my career, investing my time the same way to invest money. And so I always wanted to develop a circle of competence and then over time expand the circle. You just learn more things. And so if I started out focusing on the media industry at First Boston, they were a lot about television and entertainment. And so the transition from banking in media to being a media CFO was pretty easy because it was the same industry. And then DoubleClick was a half, and of course, King World had advertising. So I learned a lot about the advertising business. DoubleClick was an internet advertising technology business, half advertising/ media half technology. So the advertising half was something I was familiar with. The technology part was new, but it was an easy transition for me to make because I knew half of it. And then once I... And DoubleClick was a SaaS ad serving company, so I got to learn software to service very early on in the SaaS world or cloud computing. And then Nielsen was data about television. So data was new, but television was familiar. And then Oracle was software. And so there was some relationship there to DoubleClick. So it wasn't a leap from oil and gas to television or vice versa. It was incremental steps each step many years. And I read a lot and learned a lot and was able, I think, to learn quite a bit about the new companies. The biggest challenge, anytime you join a new industry, your company, is actually the jargon and the acronyms. Because I literally would write down... I keep a glossary up here are the hundred things. And the first week people would be talking about all these different words and acronyms that I didn't know. And I would just keep a list. And then a week later someone would explain what it was just in a conversation, and I'd cross it off. And then at the end of the time, I'd have 20 things that I didn't know where and I'd look them up on Wikipedia or ask somebody about it. But over the course of three months, you learned the language. And over the course of six months, I think you can learn the fundamentals of the business. The second difference besides the industry is size. And that's quite a bit different. So King World had hundreds of employees, DoubleClick, we grew to thousands of employees. And then Nielsen had over 10,000 employees, and Oracle had 100, 000 employees. So at each step, it was a different scale. Nielsen was very helpful for me when I started Oracle because it was a halfway scale opportunity to the Oracle scale. And it gave me a sense of how large enterprises, global enterprises work at scale, both what to do and what not to do because there were some challenges at that time in the way Nielsen was organized.

Matt Blumberg: So I want to come back to that, but let me back up for a second. Do you think it's possible and then do you think it's advisable to make a bigger industry shift? Could you have gone from digital media to oil and gas, or is it just bridge too far?

Jeff Epstein: I think it's certainly possible. And if you think about McKinsey and consulting firms like that, often people early in their careers work on oil and gas one year and then retailing in the next year and they get a broad exposure and they just get up to speed quickly. Of course, at the partner level, they probably specialize it. So by the time you're at the senior level at CFO, I would say it's rare, but it does happen. If you remember back to when Jack Welch was having a succession planning at GE, and they were two other people who didn't get the GE CEO job. One I think became the CEO of Boeing and one became the CEO of Optimum or something.

Matt Blumberg: Optimum, I think. Yeah.

Jeff Epstein: Completely different businesses. And of course, GE itself is a conglomerate with a hundred different businesses. So there certainly are certain managers who can do well in different industries. I think it's harder. It's a harder degree of difficulty.

Matt Blumberg: Yeah. So let's zoom forward a little bit to your time at Oracle. So you were CFO there for four or five years, right?

Jeff Epstein: Two and a half years.

Matt Blumberg: Two and a half years. It probably felt like four or five years. Most people in the world of tech and startups don't see a company that large, and it can't even necessarily conceive of what it's like to have hundreds of thousands of employees. So I'd love to hear a couple of stories from your time at Oracle that are either fun, or interesting, or, put things in context, bonus if they're about a sailboat or Larry Ellison, but certainly don't have to be.

Jeff Epstein: Well, I did get to sail on the trimaran America's Cup sailboat, which if you ever get a chance to do that, they go 40 miles an hour and they fly out of the water. It's just a pretty incredible experience. But before I came to Oracle, I had this vision that Oracle's the largest, most successful enterprise software company in the world. I'm going to arrive and I'm going to show up at my desk. I'll push a button and every number I need will be at my fingertips. I don't even have to ask questions, it'll just be right there. And the first thing they did is they say, " Here's your office and here are all the loose- leaf notebooks of the budget and the plans and everything." And they were four- inch thick, black, loose, new notebooks. And we were basically living. It was all computerized, but they printed out the reports. And so I said... This was 15 years ago, so I'm sure it's approved since then. But I was surprised at how manual certain things still were.

Matt Blumberg: Yeah. When did the company get started? It was started in the late'70s, right?

Jeff Epstein: Exactly. Yeah. Exactly. It started '77.

Matt Blumberg: And there probably were some legacy systems and processes and not contemporary dashboard instrumentation.

Jeff Epstein: Well, I think enterprise software for many years was not nearly as good as consumer software in terms of the user interface. And so it did enormous numbers of calculations very quickly and it calculated billions of numbers. But the user interface and getting information out of the system was historically poor. And I think it's dramatically improved over the last number of years. So what I found out was that we had a 100, 000 employees, and I had 5, 000 people reporting to me in finance, which was of course a very different scale than what I've seen.

Matt Blumberg: That's larger than most companies.

Jeff Epstein: That's right. And so the question you have as an executive is, what do you delegate and what do you personally do? So when you're an individual contributor, you do 100%, you have no one to delegate to, and then you become a manager and you maybe do half- and- half and you're a player coach and you do some things, but delegate other things. When I was at DoubleClick, by the end of that time period, we had 2, 000 employees. I probably personally did 25% and delegated 75%. So I would personally write my investor presentations and personally go on the investor calls, but I would delegate to our controller leading the audit committee meeting, discussion of the accounting, things like that. When you have 6, 000 people, it's more like 1% and 99%. In fact, if I tried to do things myself, the person in that area would say, " Why am I here? Why are you doing that? It's my job." And so of course, there were many things where I felt my subordinate was better than me at. I'm not an accountant, and so I wouldn't want to do the technical accounting. On the other hand, on The Wall Street stuff, I spent my whole career doing that. So I felt pretty capable there and I had to hold myself back and say, " Well, yes, Jeff, you can do a very good job on this, but why don't you let the other person do it for a while? Try to coach them, train them. Maybe they'll end up being better than you, but they'll never be better than you if you let go inaudible."

Matt Blumberg: Right. That's so interesting. I was wondering what number you were going to pick. I was figuring it would be five to 10. I didn't think it was going to be one.

Jeff Epstein: Yeah. Well, what does any senior executive do, they fundamentally are communicator. They make the decisions, they pick the people, they come up with a strategy, but they don't write presentations usually. They don't do the actual work. They delegate it, they lead whether it's a CEO or a CFO.

Matt Blumberg: Right. Or any CXO.

Jeff Epstein: Right.

Matt Blumberg: Yeah. Yeah. That's fascinating. Anything else of note from your time at Oracle?

Jeff Epstein: Well, Oracle had 40% profit margin, still does. And I said, " Well, how do we get there?" And I talked to the internal team and realized that we had this philosophy, which I think applies to many businesses, which we called... It's a four- step process. Simplify, standardize, centralize, and automate. When you're in a large organization, for whatever reason, any process becomes complex over time, maybe make acquisitions, maybe you're going to different countries. So one person leaves, another person comes in, changes something. If you take a fresh look, and you say, look at all the ways we're doing, let's say, credit and collections, everywhere in the world, we have a thousand people doing credit collections, " Are we doing it the simplest possible way? Can we standardize it, do it the same way? Can we centralize it and have it all being done in one building, in one shared service location? And then can we automate it to try to make sure it's mostly done by software and not pay people?" And if you do that, simplify, standardize, centralize, and automate process for dozens and dozens of processes, you end up with 40% profit margins. So that worked pretty well.

Matt Blumberg: That's really interesting that you probably spend a huge percentage of your time, especially in a G& A function like that, working on yourself as opposed to working on either the business or working on customer facing things.

Jeff Epstein: Well, certainly, the finance organization was very efficient, but those processes are everywhere. If you think about sales famously, if you ask an enterprise salesperson, how much time do they spend with customers and how much time do they spend doing internal, filling out salesforce reports and meetings and things like that, they're going to say a third of the time with customers and two thirds of the time doing administration. And how can we as a company change that to two thirds instead of one third and double the sales productivity of the team. Same thing with engineering, same thing with every function.

Matt Blumberg: Yeah. Here's a random Oracle question for you, and then we can pivot to another topic. I have often heard that the ending of the fiscal year in January as opposed to December was an Oracle creation that then moved into salesforce. com and other enterprise software companies. I'm assuming that was in place before you got there, but what was the logic of that? Was it around administrative things, the holidays, buying cycles, a little bit of everything?

Jeff Epstein: Trying to be efficient. When I first saw that... There were a lot of downsides of this because the last two weeks of the quarter and then the last two weeks of the year, everyone is working very long hours. The salespeople are trying to get the sales done. And before midnight and the last day of the quarter, all the contracts are coming in. So the contracts department's very busy, the revenue recognition team is very busy and it creates this enormous amount of work in a very concentrated period of time. And I said, " Why don't we try to smooth it out and make it more even and then even more efficient?" Which it would be. But as I got to see it's not the way human beings work. And if you think about an NBA basketball game, are people playing 100% the entire game? Maybe they are, maybe they're not, maybe they play really hard the last two to five minutes of the half or the end of the game, if the game's close, and maybe they're resting a little bit because human beings cannot sprint for 48 minutes and, in a business environment, you can't work all out all year long. And so the normal human cycle is to work really hard and then to relax and work really hard and relax. So Oracle's fiscal year ends in May, and so May, everyone's very busy. And then you have a couple of weeks in the first week of June to clean things up and then the whole company goes on vacation and not much happens in July. And that's perfect because it's the summer and you want to take a vacation in the summer. And also at Christmas, you don't want the fiscal year to end December 31st because then everyone's working over Christmas holiday and they're not spending time with their family. So by having a May year and not having a quarter end on December 31st, or means that everyone can have Christmas with their family, they can have a reasonable balance. And then the first quarter of the year, things gear up again, and then the fourth quarter becomes very busy. So it's actually an optimally designed system, not for administration, but for the way human beings have this very intense focus and then relax and focus and relax.

Matt Blumberg: That's a great way of thinking about it. I hadn't actually heard that before. The thing I had heard was that it was designed to give sales reps two bites at the apple. They had the end of their own quarter to make a big push and the end of the normal calendar quarter where buyers were maybe more likely to be making a big push.

Jeff Epstein: Good. That too.

Matt Blumberg: Yeah. All right. So let's talk a little bit about Bessemer. You left Oracle and you joined Bessemer and eventually became an operating partner. I can't remember if you had an interim step there as venture partner or something. I'd love to hear what inspired you to move to the venture side of things? What inspired you to join Bessemer? And what role do you play as an operating partner?

Jeff Epstein: I retired from full- time work 10 years ago, and I had my one board Priceline. And you go from a very active corporate job where your calendar is completely full, and you have people calling you all the time. And the day you leave, everyone stops calling, and you say, " What am I going to do with the rest of my life?" And I wasn't sure, but I talked to David Rosenblatt, who you know from DoubleClick, who had a similar period before his current role. And he over time just accumulated a series of board roles and advisory roles. And I tried to follow that. So I basically said yes to everyone whenever they asked me to participate in something. And a friend of mine, Byron Deeter at Bessemer said, " Why don't he come by Bessemer and work part- time as," I think, he called it, " executive residence for three months, and see if you could be helpful to our portfolio companies? You got a lot of great experience. We have at the time 100 portfolio companies. Today, 200 portfolio companies. I'm sure there are ways you can be helpful." And so I came up with lots of different ideas. We tried a lot of things. And the thing that worked was I realized that each of our companies had a head of finance, either a CFO or VP finance, and none of them knew each other. And so I created a peer group, we call it the Bessemer Venture Partners CFO Council. And now we meet once a quarter on Zoom, twice a year in person. We have online discussions of various topics. We have an online platform using software where people can ask each other questions and answers every day. And now everyone's part of this community. So I've created, and I lead this community of CFOs and it's turned out to be extremely valuable. And now Bessemer hired a whole community team to do this with engineering, and product, and sales and through trial and error, I sort of created this idea and it worked. And now Bessemer's rolling out to all these other functions. So that's worked out very well.

Matt Blumberg: That's great. And do you lead investments, or do you take board seats when someone else has led the investment?

Jeff Epstein: I don't have any investment authority. I can invest my own money side by side with Bessemer, and I have done often in Bessemer companies. And then I am often an advisor or a board member at a Bessemer company if the CEO thinks that I can be valuable, independent in Bessemer. So for instance, Twilio was a Bessemer company, Byron Deeter from Bessemer on the board. And the CEO wanted an audit committee chair and wanted someone with my background. I was introduced by Bessemer, but I'm an independent board member there.

Matt Blumberg: Got it. So the boards you're on, even if they are Bessemer boards, you function as an independent?

Jeff Epstein: Yeah. I'm never the Bessemer representative. I'm always an independent board member.

Matt Blumberg: And how many boards are you on today?

Jeff Epstein: I'm on four public boards and three private boards. And then the non- profit, Kaiser Permanente, which is very large, 90 billion non- profit health company, which is both a health insurance company and 39 hospitals, 20 plus thousand doctors, large. We provide healthcare for 25% of all Californians plus other states.

Matt Blumberg: So if you think about those three different types of boards. So you're on Twilio, a large public company, you're on some smaller private company boards, you're on a non- profit, but that's a very large business at Kaiser, even though it's a non- profit, but regulated sector. What do those three types of boards have in common? Do they have a lot in common or a little in common, especially as you think about your role on them?

Jeff Epstein: Well, the things they have in common are the governance. So the board is legally obligated to have a duty of loyalty and a duty of care. And the main job is selecting the CEO and evaluating the CEO. And then there's the strategy part. They're all, to greater or less extent, invited to participate in the strategy and ask these questions about what decisions have you not yet made that you're thinking about and give opinions about that. The larger companies, the CEO pretty much knows what they want. They have huge staff, they're asking for the input from the board, but the chances are the board's only going to have minor influence in a well- run large company. At a small company, the CEO really wants... Doesn't have anyone to talk to. So at a small company, the board is much more actively involved in defining strategy and has more influence on strategy because the CEO needs more help at a smaller company. In the large companies, I'm typically the audit committee chair and the large companies, it's almost like a... It's not a full- time job, but it's a very defined job. The audit committee chair has a role and you're actually doing a particular type of work. And the smaller companies don't even have audit committees. So those are the variety of types of things.

Matt Blumberg: So that's interesting that there's different degrees of use of the board. Are there different time allocations on things as well? I know in a separate conversation you and I had, you talked about board meetings have some show and tell component and some strategic decision- making discussion component. Does the ratio of those things change, small to large?

Jeff Epstein: Yes. Well, part of it is just how many hours you spend. So a large board typically has a two- day board meeting four or five times a year, plus committee meetings in between that. The very small companies might have monthly board meetings, and they might move to every two months or every three months. They might only be three hours long. So the large companies typically are more hours. The smaller companies, you might actually spend more time because they want you to come in and they want you to meet with customers. They might want you to help recruit employees and do things outside of the board context because they're already paying you to be a board member. They say, " Why not take advantage of this extra person who can help with all these activities?" The other difference I would say that is whether the company's going through a big change. So if you're doing a big acquisition, there might be a series of meetings. During the financial crisis, I was told the Citibank board meeting had 70 board meetings that year. So every week or twice a week, there was something we had to do, so that's of course an extreme case. If you're making a senior leadership change or a change like we have at Twilio this last week where we're dramatically changing our organization structure and reducing headcount, there were a whole series of meetings over the last two months where the board was actively involved in that. And then there might be a period where things are pretty much on track, and we have fewer board meetings.

Matt Blumberg: A lot of our audience are earlier stage founders that probably can't contemplate what a two- day board meeting looks like because they're doing one- hour board meetings or three- hour board meetings. What does a two- day board meeting look like?

Jeff Epstein: Well, it's a one- day board meeting with one day of committees' meetings and then dinner. So the board meeting itself would typically between, for large companies, it's probably six hours long. And if you have a big company, every part of the company is you're talking about each business. So at a company like Twilio, we have many different businesses within Twilio, and we're talking about each of them about US and international, and we're talking about our software business versus our communications business. We might be talking about competition, regulation, new products. If you think about each segment of a board meeting is 30 to 60 minutes long, it's not anymore-

Matt Blumberg: More segments. All right.

Jeff Epstein: And then there might be a board dinner with the board and leadership team together, so 20 or 25 people. And then committee meetings, there are typically an audit committee, a compensation committee, a nominated governance committee. Something at Kaiser Permanente, there's a health committee and the community benefit committee. So there might be other committees as well. And each of those committee meetings can be two to three hours long.

Matt Blumberg: All right. One of the things that I think is so interesting about some of the big boards in general, and in particular some of the ones you've been on, is you end up with some board members who you wouldn't necessarily think of as, " Oh, that's an obvious choice for that board." I think the governor of Massachusetts is on the Twilio board with you. I think you've been on boards with four- star generals with Jeff Immelt. How do you find some of those directors adding value or functioning in a high growth tech board where they don't necessarily have the domain coursing through their veins?

Jeff Epstein: Well, Jeff Immelt, the former CEO General Electric, and Deval Patrick, former governor of Massachusetts, are both extraordinarily talented individuals who have led very large organizations. And I would say the first thing they add, the first value they add is strategy. So they've led strategy at their large organizations that figured out how do you make these trade- offs? And of course, board members don't set the strategy, but board members, if they're good, they'll ask terrific questions, which can help the company come up with a better strategy. Secondly is people. They've hired and fired hundreds or thousands of people over their careers and seeing what works and what doesn't and have a great instinct for people. And so giving advice to the CEO on individual people just general advice, but also specific advice about these leaders that we meet during the course of our board service has been very helpful. And then they've also, at scale, seeing processes that work and processes that don't. So that's very important. Scale is very important when you're a company like Twilio, growing from a couple hundred million dollars in revenue to$ 4 billion in revenue. There's new things that we've never done before that both Deval and Jeff have seen at scale. And then of course, there's the governance, the legal obligation of being a steward of the company and complying and regulatory compliance and things like that as well. So it's a pretty broad spectrum of skills that senior executives like that bring inaudible on Jeff extraordinary executives.

Matt Blumberg: When you've seen... I don't think you've ever been on a board that where the company has had a spectacular failure, but if you think about boards like Theranos or BTX that had a lot of good board members, what do you think goes wrong?

Jeff Epstein: I was on one of those boards, it's called Global Eagle Entertainment. We competed with Gogo, we provided in- flight entertainment to Southwest and other companies. And the company had a fundamentally difficult business that was low margin. If you're selling something to airlines and is only a handful of airlines, it's just a tough business to be in. In fact, Priceline started out selling airline tickets and fortunately figured out that the hotel business was much better than the airline business and is now we have 90 plus percent hotels and only a small percentage airlines. So we started out with a difficult business at Global Eagle, and then we wanted to diversify away from Southwest. We bought a company providing similar kind of internet connectivity to ships, and it was a big mistake. The company we acquired had not been well run, they didn't have happy customers, the revenue didn't grow the way we expected to and the merger integration was poorly done. And so we took on a lot of debt, we missed our numbers, and then we were in the fault of the covenants of the debt. And then I ended up resigning for the board after a number of years. And then when COVID happened, airline travel stopped, and the company went into bankruptcy. So what did I learn from that? I learned the number one thing is we should have executed our existing business better before we took on an acquisition. We were concerned about the Southwest risk, so we made the acquisition defensively, and I think that was a mistake. We just should have said, " Look, let's continue taking the risk until we could fix our own business and then make sure we had the ability to integrate the acquisition." So I think we were too impatient to grow revenue without making sure we could do it efficiently and effectively and have a business model that worked in our core business.

Matt Blumberg: Interesting. And probably some very unhappy board meetings along the way.

Jeff Epstein: It was a challenge.

Matt Blumberg: Yeah. All right. Let me close with a final quick question for you. If you had to give one piece of advice to CEOs that are watching this or listening to this, one piece of advice for founders about how to most effectively scale, whether scaling themselves as a leader, scaling their team, scaling their company, their board, what's one thought you would leave founders with?

Jeff Epstein: Well, I'll leave two thoughts. One is, it's very good idea to meet people who are at the scale you want to be and recruit them to your board or meet other CEOs and try to just get a sense of what the future's going to be like. But then the famously Paul Graham, founder of Y Combinator said, " Do things that don't scale." So when Airbnb started, the two founders personally took images of lots of apartments to make the image quality better on the site. And they obviously couldn't do that with a million apartments. But the idea is initially do it things that don't scale, but then you have to do the second start set, which is then scale it, which means then hire photographers and create software for the photographers upload their photos and do something which where if you double the revenue, you're not doubling your headcount. And so what is the definition of scale in my mind? It's exactly that. It's wherever you are now, if you double your revenue, will you double your headcount or only grow your headcount by 50%? If you can grow your headcount by 50%, then you're scaling. If it's 100%, you're not scaling. So what are the systems and processes that you need to put in place to achieve that scalability?

Matt Blumberg: I love that. Thank you. And thanks for talking to me for a while today. First of all, it's always good to see you, and I appreciate your stories and wisdom for our founder audience.

Jeff Epstein: That was a lot of fun, Matt. Thank you.