Episode 22: Why Startups Fail with Tom Eisenmann

This week, as usual, we're trying to cover things I wish I knew when I was 23 and starting out as a first time tech entrepreneur. Today we've mostly covered things that you know, what to do, how to come up with a business idea, how to fundraise, how to evaluate your business idea. And this week we're doing something different. It's like understanding why startups fail. Very few people have actually gone in depth to understand why and we're actually privileged to have someone has done that. So without further ado, let me introduce you to the episode of the week. So welcome to episode 22, why Startups Fail.

So Tom, thank you for joining us this week. Why don't you start?

It's great to be here.

Thank you for being here. So why don't you tell us a little bit about your background and what got you down this rabbit hole of like why startups fail.

So I've been a professor at Harvard Business School for the past 24 years, since 1997 and good chunk of the early part was spent studying platform Strategy. But about 10 years ago I took a turn into working on startup management and started working closely with students who were launching businesses. Around the same time I had discovered Lean Startup, brought my students out to see Steve Blank, and brought Eric Reiss in as an entrepreneur in residence at the school, and had several teams that I thought were just terrific, encouraged them to start the business. One of them, Quincy Apparel, followed Lean Startup principles in just a textbook perfect way, validated demand, had a great idea I thought, had confidence in the team and they failed. And so here I was, supposedly an expert on entrepreneurship. I was a failure at explaining failure. And since depending on how you define startup and how you define failure, maybe 2/3 of startups will fail, that seemed like a pretty big gap in my background, so caused me to question whether I was teaching the right stuff or whether I was really as competent as I thought. So put me on a quest to learn everything I could about why startups fail. Talked to lots of failed founders and the investors who'd back them, do some survey work and so forth. Since things move slowly in academia, took eight years of research to come up with a book.

So here's the book by the way, for everyone, why Startups Fail by Tom Eisenman, which I had the pleasure of reading last week. And it's both a well written and so it's a reasonably easy read, but it's very structured and thoughtful and I think everyone is an aspiring entrepreneur should read it, especially since it Breaks down all the underlying reasons and, and breaks it down in terms of, like, what, what happened, why they felt early stage from late stage, which is rather different. So, so why, by the way, is a Harvard Professor McKinsey background the right person to tell people, you know, why startups fail rather than a practitioner?

Well, I mean, in the first place, a lot of what I know I've learned from practitioners. So, you know, if anybody would go to the acknowledgments, paragraph number one is thank you to Steve Blank and Eric Ries and Fred Wilson and Peter Thiel. You know, you go right down the list. A lot of the best ideas in this field come from folks who've cut their teeth on the problem. So that said.

I've been working with literally hundreds, maybe thousands of aspiring entrepreneurs, have invested in a bunch of them, have coached them, have been on boards. So.

A lot of the best investors, and you're among them, have been entrepreneurs. But it's not hard to find venture capitalists who've never run a startup. I think Fred Wilson would fit the description. Jim Breyer would fit the description. So, you know, if we don't, if we don't require venture capitalists to be successful entrepreneurs, I don't think we should force a business school professor to do it.

No, Bill.

Bill Belichick, by the way, one of our great, greatest football coaches wasn't. Wasn't much of a football player himself, so.

Yeah. And by the way, like, I actually think, you know, writing a book, that's like launching a startup, right? Creating a class.

You bet.

Like launching a startup.

And I've launched 4, 14 courses. It is. I mean, it's not in the sense that if, if it doesn't work, they're going to force me out of the school and I, you know, I won't be able to pay the rent. But it's a long project. Each course takes a couple of years. It involves lots of people. You have to generate demand. You have to guess whether the market's there. And some of them do fail.

Demand, meaning.

And by the way, most books fail.

Like this because you need to attract students to the class, basically.

Exactly. Yeah. Yeah. The students vote with their feet. And to your point about the book, the vast majority of books fail, so we'll see how this one does.

And that's your second book or your first book.

It depends how you count. I taught a course on Internet business models back in the boom and wrote a case and textbook there. I edited a volume of other people's blog posts that I thought were really good ones. So that kind of counts. And then when the first time I came up for promotion at Harvard Business School, I had taken everything I Learned about Internet era.com boom era, get big fast strategies and put that into a book. And the folks who were trying to figure out whether I was professor of material thought was done in a hurry.

It's interesting because if you apply the same principles of entrepreneurship to book writing, uh, you know, like, put it. Be a. Be a shame of the first version you put out there in mvp, et cetera, I think most people, yeah, I can see why people in academia would want it to be slower and different.

Yeah, yeah. The question was, aren't you the guys that were excited about how I filled up the classroom with all these people interested in Internet companies? And, you know, that was done in a hurry too. So I actually put the book out. So I was told that as I came up for promotion the first time, unless you think everybody who reads that book thinks it's the most brilliant thing that's ever been written about Internet companies, you probably shouldn't publish it. So I actually had a big advance from a publisher. I marched back to my office and said, where do I send the check back? And put the book in a drawer. And then essentially 15 years later, pulled it out of the drawer and looked at it and said, wow. My current students were in middle school when all this happened. There's actually a lot to learn from who grew too fast in the dot com boom. And sometimes that strategy paid off. A lot of the times it didn't. So that actually was an exercise in figuring out what separates success and failure. The book is available. It's called Speed Trap. It's a free ebook. So anybody who's really curious about what it was like 20 years ago can go have a look.

The times we are in are also. It's not as crazy it was then. I mean, I was getting like, signed, binding term sheets by fax from people I never met. I mean, that's a bubble. So we might be in a bubble in the public markets now, but it's not a bubble in the tech startup early stage.

There's really no comparison. It was insane.

Yeah, it was completely insane. And you're right, most of the entrepreneurs I back who are in their, like, you know, mid to late 20s, early 30s.

They didn't live through it in any meaningful ways. They don't realize how crazy it was relative to the frothy times we're in today.

Yep.

So why do you.

My introduction as an angel investor and I probably did 15 of them all in late 99, early 2000, that, that portfolio, you know, I think I got my money back on one. Money back, not, not a return. And a couple others I got.10 cents on the dollar is a whole bunch of goose eggs there.

Well, I'm sure most, sometimes worse, right? Like most wanted, like literally zero. So why do startups fail?

Boy, you know, the easy answer is they run out of money and they can't raise more.

And it, I mean, there are some who fail because they raise too much money, but we can come back to that. But, you know, it's not a very helpful explanation for an entrepreneur. It's like the coroner saying person died because of loss of blood. You know, okay, you know, was it an accident? Was it a gunshot wound? And so you have to keep asking why. And what the book does is it isolates for, as you pointed out in the intro, the patterns for early stage startups, the failure patterns are different than they are for late stage. So. So the book lays out three recurring common patterns, reasons why early stage startups fail, and then three more for late stage startups. And I think they really do cover, you know, a lot of observers will pin the tail on the, on the donkey by every. I'll stay with the horse and jockey metaphors. Right. The explanation for a lot of people on why startups fail is bad founder, bad founder fit or founders, plural. And you know, there's a lot to that, but it's really, as I found in the work, an oversimplification. You know, there's usually much more going on. And, and so what these failure patterns do is, is, is isolate the sets of recurring patterns.

I'm sure all the VCs are like, it was the founder. And all the founders are like, no, it's the market.

Yeah. Yes, exactly. And we actually, in academia, we've got a name for that. Anybody who's taken a psychology course has run into this. It's called the Fundamental Attribution Error. It's the first thing you learn in psychology 101 that if a bad thing happens to us, it was not our fault. It was external circumstances, it was the situation. My co founder dropped the ball. My VC pushed me to grow too fast. The regulators did something crazy. My competitor came out of the blue. I couldn't possibly have foreseen it. And if we see somebody else do something bad, it was their disposition, their lack of skill, their lack of commitment.

And so that was actually one of the challenges with doing the research is if you go to first person accounts, you blame the market. If you go to third person accounts and in particular ask the investors, you blame the founder. And the truth is usually some combination of all of that and there are patterns to why this happens. So the case study approach actually turned out because the case forces you to sort of surround the problem and come at it from different angles. Talk to the team, talk to the founders, talk to the investors.

Yeah, no, I actually think in this case it might work well because if you try to just do surveys, you're going to have that attribution error that's going to come through. Like all the founders will say, ah, we didn't find product market fit. And all the VCs will be like, yeah, the founder sucked.

Exactly, exactly. Right.

So you want to summarize the three for the early stage. Maybe we can start there and then. Sure, go through that.

Yeah, so. So the three are.

The first one is good idea, bad bedfellows, and the, the premise there. I'll do them fast and we can come back and figure out which ones you might want to talk about. The idea there is, the founders actually got a pretty good idea. The Quincy Apparel, the company I mentioned the beginning as the motivator here, fits this description. Pretty good idea. But they never assembled the call it a big T, Capital T team. You know, founders, other team members, supportive investors and supportive strategic partners. And for a whole bunch of reasons why they can't quite get the resources together that you need to actually execute on the good idea. The second pattern is almost the photo negative of that. You have a pretty good Capital T team, supportive investors, competent team, and they never find their way to a good idea. They're pivoting, pivoting, pivoting. But you can only pivot so many times before the money runs out and the investor's patience runs out and they won't bridge you further. So, and the underlying cause for this pattern is, I call the pattern false start. And the reason behind it is like entrepreneurs identity, entrepreneurs want to get going. They have a bias for action. It's how they view themselves, how they view the world. And once you sort of make the plunge, sort of quit your day job, your salary job and start going, there's pressure on you to launch. And so these entrepreneurs are eager to build and sell and they've got a vision burning bright of the opportunity, the pair of problem and solution, and they do it. They go out and build the thing without doing upfront research to figure out is the demand actually there and of the Many ways we might try to solve this problem. Have we got the right solution? And we're not talking about years, we're talking about weeks, maybe a month's worth of upfront research that people just skip. And by skipping it, they basically boosted the odds that the first version of the product is going to miss.

I think it's especially true of product people and engineers who just want to build. That's their expertise. And so doing customer interviews and identifying product, market fit, etc. Doesn't come necessarily nearly as naturally to them. So we see that.

Exactly. Yeah, yeah. We can stereotype and say that engineers are introverted. So getting out of the building and Steve Blank style to go interview people can be uncomfortable. But I tell you, Fabrice, you know, Most of the MBAs I work with are not technical. But what they hear from everybody is you need great product. And how do you get a great product? Well, somebody's going to have to build this thing. And so. And because they're skilled at networking, they can find their way either to a technical co founder or, you know, somebody to run the engineering side of things. And boy, once you bring that person on board, there's. Particularly if you're paying them. Right. If they're not a co founder, there's pressure on you to keep them busy. How do you keep the engineer busy? Well, you know, here's the vision, here's the specs. Build this and, you know, if you raise 18 months worth of capital or, you know, in a lot of cases, people are scraping by and they really only raise a year when they probably should have more than that, you know, so if you've got 12 months of Runway and you waste the first four months on a flawed version of the product, you can still pivot. Yeah, but now you've only got essentially two more pivots left before the Runway has expired.

Yeah, for us, that's a false. That's. Episode 10 of my show is like, okay, you have this brilliant idea. Episode nine was like, how do you get the idea? It was like, how do you don't launch immediately? Like, what are the things you need to prove to be true or false before you go and actually launch? Then you hire the engineer, quit the job and pull the trigger before then it's too early.

And that's why there's such an advantage to non technical founders. It takes about 200 hours to learn enough coding to create a crappy Python prototype. And usually that's enough to sort of get some feedback. And by the way, it helps in attracting A technical founder. Like, they'll look at it, they'll laugh, it's pathetic. But they'll respect you for having tried.

And they'll actually go one step further. I wouldn't even do eight, do customer interviews. But B, just do a landing page that's beautifully mocked up. Why code anything like 500 bucks, spend 100 bucks in marketing, you get a sense of CAC, you get a sense of what people want, you talk to those people signed up. I wouldn't even code anything.

Yeah, no, I agree 100% and try to get every student team I work with to do that first. It works in B2C for sure. It works in most B2B markets too.

It works in both cases, I think. And so number three.

Yeah, so false start. Just like an athlete, right? Sort of sprinter or a swimmer jumping the gun, literally jumping the gun. Number three is a cousin of that failure pattern. I call it the false positive. Just like COVID testing, we're all familiar now with false negatives and false positives. The pattern here is, boy, every entrepreneur needs the early adopters, right? If you're going to make the business go, you've got to find some way to attract them and make them happy. And.

It'S often the case that the needs of the mainstream. You need the mainstream consumers too if you're going to build a business with any scale. And it's often the case that the mainstream consumers have different needs than the early adopters. And if the entrepreneur gets beguiled by, seduced by the early adopters and works really hard, which is a totally understandable thing to do to meet their needs and satisfy them, you can actually commit too hard and too far in terms of serving the early adopters and not have. You may have a product that's over engineered for the mainstream. It may confuse the mainstream. Great example of this scene. I saw on your list of portfolio companies that Dropbox is on the list. Drew Houston did an amazing job of avoiding this trap. Now, he knew he had to test the product technically and in terms of just understanding demand for it with sophisticated early software engineers who were basically on hacker news, who really had very demanding and sophisticating requirements for file management. And. But he had in his head, you can read it in his Y Combinator application. I want to build a product that's so easy to use, my mother can use it to store her recipes. So he actually had the discipline to engineer Dropbox, you know, so it was just simple to use. He took a bet that it would be good enough for the early adopters sufficiently superior to what the sort of crappy solutions they were using at the time that they would embrace it. And they did. It was a good bet. But boy did he play that.

You know what's interesting is I would argue there are some counter examples here. Think of discord, right? Like going after these early adopter gamers and being completely hardcore or twitch and turns out that those audiences are actually big enough that these early adopters pull you all the way through to multi billion dollar companies. And often you want to be pulled by those people because it means you have product market fit. Now sometimes of course they it's not big enough and then you've built something that's not appropriate for the audiences. But finding that gold lilux as you call it, like medium is really really hard and all of these things you're like let's find the gold lilux. Wow. I mean true, but how?

I agree completely. I think the point is not so much that there's one right way to do it, it's that every set of early adopters and mainstream customers and in every product strategy, it's just something that entrepreneur has to keep in mind. So you have to simultaneously figure out how to get and satisfy the early adopters and to make sure that whatever you're doing will work when it comes time to sort of cross over to the mainstream. And there'll be different ways to do that. Sometimes you start with a product that's great for the mainstream like Dropbox. Sometimes you really double down on the early adopters and stay with them and evolve the product or the marketing message over time.

So let me summarize what you've said so far. So. So for early stage companies, three main reasons they fail. One is bad. Bad fellows, meaning either co founders or investors or partners or employees are not the right people. Number two, false starts, meaning you get going and it's actually not the right product market fit. And so you have to restart a number of times and perhaps you don't have enough cash to get to the correct product market fit. And number three, false positives where you think you have product market fit, but what you have it works with these early adopters is not adaptable to the rest of the market. And so even though look promising ultimately was not the right product and you fail as a result of that. Maybe and we should double click using old terms of probably aegis on all three of them, but maybe give the two minute summary as well of the why, the reasons, the late stage sharp sale and then we will, yeah.

So I mean, what was really, I guess, surprising to me, disconcerting too is if your definition of failure is investors don't make money, which is one definition. You can come back to other definitions of failure, but if that's the definition, then a surprising fraction of late stage startups, and these are startups, let's sort of think of them as series C or D and beyond don't make money. About one third don't make money for their investors. And, and these are companies that have had, at least at some point in their history, product market fit. They may lose it over time and they've amassed a lot of resources, team money, et cetera. And when they fail, they leave these steaming craters in the landscape. I mean, these are the ones we read about and hear about. Like when a seed stage company fails, it's painful. You may feel some pain and the mother and father of the founder of ME feel some pain, but not too many other people hear about it. When Theranos goes down.

We get movies. So the three patterns are speed trap, and this is very simply just a company that had product market fit. Great momentum from word of mouth, satisfied early adopters not spending anything on marketing. The thing's just growing. Investors come in at a high price for the equity with an expectation of continued growth. The entrepreneur doesn't need her arm twisted, right. All the entrepreneurs want to grow too. And then bad things start to happen. Clones come out of the woodwork. Incumbent companies wake up to somebody finding opportunity in their space. The next wave of customers is rarely as excited as the early adopters. So you have to spend money to acquire them. They repurchase less. They don't generate word of mouth referrals. So you get an LTV lifetime value and cac, customer acquisition cost squeeze, price cost squeeze, and your unit economics start to wobble at the same time, if you've got humans doing anything. Some businesses are pure software, so they scale smoothly. But many businesses have people packing boxes or answering phones or whatever. And now if you're generating all this demand, you got to hire legions of employees, which is hard to do. Think Robinhood a few weeks ago. And so you get quality problems and customer service problems. You also get cultural problems. You get old guard, new guard conflict. The old guard was present at the creation. They're mission driven, they love the founder. The new guard are specialists who feel that their contributions aren't understood or appreciated by the old folks. The guy down the hall who's got sitting on $5 million worth of stock options and you get fiefdoms within the company. Different marketing points its finger at engineering for being late with the features, sales, blames marketing for generating crummy leads, et cetera, et cetera. Everybody wonders where the CEO is because he or she's off raising money all the time now. And eventually all this catches up with the company. The cultural discord and the economics wobbling. And people will pump more money in sort of under the assumption that people are going to fix things, but they just can't fix them. And it can unravel very fast when it happens. The case study in the book is Fab.com, which I think a lot of people will remember, sort of an E Commerce darling for a while, and then the wheels came off.

But do you think Fab.com could have also been a false positive rather than a.

A false or a speed trap? Because in a way, the early cohorts were amazing. It's just that the early adopters loved it and then no one else after that loved it nearly as much. So it feels like it's as much as that as it is as we.

Grew.

We had competition, we got distracted, we went to Europe, we hired these specialists and was creating these wars. I mean, all this was true, but it just feels like at the end of the day, you know, the early courts were the best, the golden cohorts, and nothing, no, no one and nothing ever compared to it.

Yeah, no, I. You've got it. Exactly. I mean, the root cause is a false positive. But the thing was Jason Goldberg, who's a pretty charismatic and amazing entrepreneur, was so good at harnessing the momentum and. And, you know, he raised basically a couple hundred million dollars worth of capital. So after the false positive, you could still play the thing out. Most companies that are killed by a false positive are killed.

Before they reach series B. You know, these guys raised enough money to keep going and going and enter Europe to basically go after the samvirs.

And the clones there. And so we see the false positive actually leading to the speed trap. In this example, the second late stage pattern I call help wanted. And this is, you know, again, a bit of the opposite. Whereas a company like Fab lost product market fit for exactly the reasons you say, sort of the late. The later customers weren't nearly as good as the early cohorts. Sometimes a company can actually sustain demand, can sustain product market fit, but they're missing some key, let's call it a resource. In some cases, it's a key manager, a key specialty leader in a function, and the case in the book is another online home furnishing company called Dot and bo. And they had a great formula. They actually.

Had.

Fantastic demand repurchase, et cetera, et cetera. And they just. I mean, it turns out to be really hard to ship. Couches across the country can't arrive two days early. If your Amazon books come two days early, you're thrilled. It's got to be there when you're home from work because, you know, that's. Well, I guess we're all home from work these days. But.

This, this was. It goes back to 2015 and it can't arrive late and it can't be beaten up, which things that big objects can get beaten up. And so they went, they ran through three vice presidents of operations looking for somebody that could actually handle the logistics. And they bought the wrong ERP system because nobody really understood the requirements and they couldn't keep track of their inventory. And so they were selling things that were actually stocked out and they were failing to sell things that they actually had in stock and they couldn't staff up customer service fast enough. So orders were being canceled and they weren't keeping track of them. So it was just a mess. And the insult to injury was the other side of the help wanted patterns, which is around 2015, you'd have been in the middle of this investment in a lot of E commerce categories just dried up really suddenly.

Well, I mean, Fab.com, gilt, ShoeDazzle, all those failures led to all the VCs saying, wait a minute, I don't want to be in this category. It's not particularly compelling. So, yeah, I do remember that very well.

And this is just when Dot and Bow had stepped on the gas and was trying to raise its Series C and you know, it just, it wasn't going to happen. It's also interesting.

And then when, you know, DTC in a way is like E Commerce, you know, coming back. Right. So there's like this lull and then it comes back.

Yeah, happens a lot.

Delivery services have sort of seen the same sorts of waves of up and down and up and down, satellites in low earth orbit, I mean, all sorts of things.

Nanosatellites, you know, we had Cosmo and Webvan and people probably the audience doesn't remember those, but like Webvan was going to be fresh Direct, not really Instacart, really fresh Direct. And Cosmo is going to be V1 of whatever DoorDash or Postmates.

Doordash.

Yeah. But with that smartphones and geolocalization and the ability to be asset lighter. I mean, zinjikart is a better way to build it.

Chewy versus pets dot com.

Exactly.

And it keeps going.

Those are good ideas.

Yep.

Okay, so, okay, so number one, we have trapped, you know, it's difficult to keep growing. You have a specialist versus generalist battle. You can't hire the right people. You have clones and competition. Number two is like getting the right people help wanted. So yeah, like you have, you need someone to do something like in this case, ops. And they couldn't get it. And then number three is.

Number three is Cascading Miracles. I love this name. I stole it from a guy named John Malone. Was the entrepreneur who built tci, which was the biggest US cable system company. And the concept here is lots of things have to go right for the business, for the venture to work. And if any of them go wrong, you're dead. And the chance that all of them are going to go, you need a cascade of miracles to make sure and the things, so. So these are big, audacious, bold visions that often, almost always they will take have an incredibly prolonged product development cycle just because you're doing, you know, this was Webvan, it was Segway, I'd say.

Iridium in that satellite space. Magic leap. Yep. And of course, Better Place, which is the example you give in the book.

This is the example in the book. But you know, so it's important to understand sometimes it works. Federal Express in the 1970s was the biggest time, the biggest venture capital startup in history. And people thought Fred Smith was insane. You know, why would you put a package on a plane in Buffalo, you know, send it to Tennessee in order to sort of send it back to Boston, Massachusetts, which is what they did. You know, Hub and Spoke was a newish concept.

And Anyway, so Tesla, SpaceX, Federal Express, sometimes they work, but boy, when they fail, they fail in a gigantic way. And what you need are basically fundamental changes. So better place for the listeners. Plan was a network of charging stations for electric vehicles. This goes back 10 years, more than 10 years. And when range anxiety was even greater than it's still great today. But these electric cars at that point could only go about 100 miles before they needed a battery recharge. So the entrepreneur Shai Agassi was going to blanket the world with charging stations and then these battery exchange stations where you'd pull in like a gas station, a robot would pull out the depleted battery from your car and pop in a new battery. And all in five minutes they built this, they built it in Israel and Denmark and burned through $900 million proving that this was not a great idea. And so you need fundamental change in customer behavior, thus a product support in this case of car manufacturers, right? Big incumbents who benefited from the status quo, who may not care to be steered around by a startup government because you're often on fuzzy legal regulatory ground and huge amounts of capital and often just technological engineering breakthroughs. And you know, if you flip a coin, so if each of Those is a 50, 50 chance and you flip a coin five times, the chance of getting heads five times in a row is 3%. It's like spinning a roulette wheel and landing on a specific number. And you know, so, so what you need is like an astonishingly charismatic entrepreneur who can spin the tail of why this, you know, we're going to usher in a dazzling future, put a dent in the universe, Steve Jobs style and so forth. And the charisma only lasts so long when the business starts to get in trouble.

It's interesting. Like that one struck me as a terrible idea from day zero. I remember when it was first fetch of batteries are going to get better. You're re. You know, and, and I expected by the way that they would get better faster. I mean the range has increased. Now you can get like 4 or 500 miles. I actually thought that the graphene type batteries or V flow batteries or you can recharge your batteries in five minutes would be here by now. And they, and not they're in the lab, but they're not in, in mass deployment. So I'm like, you know, this is like an idea who's like completely temporary and it's super complicated, right? Like you need stations, you need like robots, you need a standardized system for removing and putting them in. Agreed by the manufacturers. It's like over engineering for something that's going to get solved automatically by improving battery tech over time.

Yeah, yeah, a bunch of things happen and batteries got better, but again, fast charging didn't. You can still damage your battery if you fast charge it too many times. The interesting thing with Tesla is they actually, they did a bunch of things, they put bigger batteries in bigger cars. Right. If you compare the Tesla sedan, it's just a big car, which means bigger Battery could go 300 miles when everybody else can only go 100 miles because your battery was three times the size. And they built battery exchange stations, same as Better Place. Turned out nobody wanted them. What people wanted was the fast charging stations which they'd use on a Long trip between San Francisco and LA or something like that. You know. So they tried a little bit of everything and much more flexible approach as opposed to as you say, sort of committing to one solution and over engineering that's.

Well that's the problem with the misnaic founders. It's like it starts being loyalty. It's like you have you believe in me or you don't and if you don't believe in me, you're out. And then it's like self reinforcing cycle of like you only hear what you want to hear and you think it's working when it's not.

Yeah. Sometimes it's just you get your counsel from the mirror. Right. Basically you surround yourself with yes people and then in the worst of instances it, you know it can concern to sort of ethical lapses. And we saw that with not. Not with better place but Theranos. Exactly. Yeah.

The. So I actually did not double click on Theranos. Like did she know that it wasn't working and she figured they'd eventually figure it out or not really.

You know, it'll all be settled in court I think. Okay. But if you read the book is amazing. The the Wall Street Journal reporter did Bad Blood. Everybody listening. I'd recommend if you like these stories. I can't imagine a better research than more fascinating. We teach it now in the MBA program at Harvard and I thought it was a stupid idea because isn't it obvious what the problems were? But boy, the students just love talking about this because you get into really interesting issues of the board's responsibility and.

There was a whistleblower who made a huge difference, you know. So it's a fascinating tale.

Interesting. Well it's definitely in my two read list. So why don't we go back to the first idea. So bad bad fellows, you're an early stage startup. What are good betfellows like? Like picking your co founder is really hard. But you didn't mean just co founders, you actually meant early employees. Maybe you can talk about a what Quincy did wrong. And by the way, I was also an investor in Quincy and when Christine pitched me it was like literally the best pitch I'd ever heard. Like I gave her a check in the meeting. Like you're going to figure it out. Like you're extraordinary From a visionary storytelling. You seem to have identified product market fit. The early traction is really good. Why not? Let's go do this. And I was shocked that not that much later, nine months later, it was over.

Yeah, it was fast. And you're exactly right about Christina. She is now my colleague on the Harvard Business School faculty where she's just doing a great job and it's fantastic to have her. She's very charismatic.

She was paired with a co founder, Alex, Alexandra Nelson. And Alex was just what you'd want to pair with the charismatic, the outside person on the team. Alex was inside, a disciplined engineer from mit, you know, who would be very analytical, who would run the inside stuff. And so the two of them looked like they were balanced inside, outside. They had a vision. They knew how to sell the vision. They also were best friends and resolved to never let the disputes over the business get in the way of their friendship. And that's, that can be a big problem because basically you're going to have disputes between co founders over the business and if you can't work them through, you're going to slow things down or not make the tough calls. And so they clearly fell into the trap of just not sorting through. And they also, like so many, you'll have seen this a lot, sort of teams of MBAs all want to be the boss, right? They named Christina the CEO, but they actually functioned as effectively as co CEOs and they sort of had an agreement that any big decision, including all the product decisions, they would both have to agree on. And their tastes were different. I mean, yeah, fundamentally the aesthetic which is so important in a fashion apparel business.

You know what's interesting is though, here there are counter examples, right? Like there are many. There are a lot of co CEOs that have worked out and there are also a. I've even seen husband and wife teams actually work really well. When they don't work, it's a disaster. And it's like the end of the relationship is the end of the business. But when they work, they work really well because they can kind of finish each other's thoughts.

Yeah, you're right. It's easy to find other examples. You know, here I'll lean on research done by my, by my former colleague Noam Wasserman. He wrote, he wrote a great book called the Founders Dilemmas. And he did a lot of statistical work that basically shows, you know, obviously you can find examples of anything. But on average teams of close friends or family members are less stable, less likely to hang together. And he would say it's because they push away the conflict, they defer the conflict. So that was one problem they had. The second problem they had is, and boy, I fell into this trap because I'd seen a whole bunch of Harvard Business school teams without deep domain expertise, industry expertise, do just great. The Rent the Runway founders had launched a year prior to they didn't know anything about fashion or apparel, and they built this amazing disruptor.

And in fact, the Quincy team had actually gone in and done project work for the Rent the Runway team. So that's funny.

That seems to be kind of an incubator for other fashion companies, I think. Rebag Charles also worked at Rent the Runway for a while. Yeah.

Yep, exactly. And there's a lot to learn there. And. And Jen and Jenny, who started Rent the Runway, are great entrepreneurs. So I looked at them and said, okay, well, yeah, you don't need fashion. This one's working great. We'll. We'll run this movie over again. But it turns out that as opposed to renting the dresses, actually designing and making them is an incredibly complicated task that takes a lot of specialty expertise. You know, there's all the. There's designing for manufacturability, there's sourcing the fabric. Fabrics have different elasticity, so you may think that the thing you used last time will be the same, but, you know, if it wasn't made in the same mill, it may just not fit. And if your promise, as it was for Quincy was better fitting a barrel, you know, and the stuff doesn't quite fit exactly right, you've broken your promise. You know, fabric, pattern maker, quality control, and there's like 10 different steps, and they have to fit together and coordinate the flow of work. This pair of founders knew nothing about this stuff. So what do you do? You try to find a co founder who knows you're not credible, so you can't attract anybody to do that. You try to surround yourself with advisors who can fill in the gaps. And they did some of that, but they should have done more to build out the team. You then, of course, if you don't have the specialty expertise yourself, you go for a specialist. They hired people out of existing apparel companies. Go to Ant Hill to get your fashion cutter and Taylor to get your fabric cutter. And those folks had no idea how to work in an early stage.

Those are the specialists you bring in when you're in late stage, not the. They don't fit in the early stages.

You know, in an early stage, you want jacks of all trades who can sort of shift from one task to the other. Whatever fire, you know, come and help put out whatever fire is burning. And so they hired for skill and not attitude and ended up with a team that sort of sat on its hands when when they should have been, you know, sort of pitching in and making things work. So the team was off target. The founders lacked domain expertise and were fighting with each other.

I think other investors.

The folks who came in early, I think were convinced by a great set of minimum viable product tests and these charismatic, competent founders, the next wave could sort of see that it wasn't coming together. So they had trouble raising the next wave of capital and they, they targeted raising a million and a half. Only raised a million. That meant two seasonal connections. Apparel comes in seasons, so spring and fall and so they could do two of those, but what you really wanted was a third to start to shake out the production bugs. They had a bunch of production bugs, just didn't, you know, while the sales were pretty strong and the repeat purchases were good, they didn't have enough momentum or enough of a clear signal that they could fix all this stuff to raise what essentially would have been back in those days a Series A. Now it would be a seed extension.

And so they ended up with investors too that were too small, couldn't bridge, weren't inclined to bridge, and really didn't bring the kind of domain expertise that they probably needed. This business probably shouldn't have raised venture capital. They probably should have found smart angels or even factories will sometimes put money into and into new apparel companies.

Yeah, I mean, look, ideally inventory, you want to be financed through debt. Though in the early days you're not credible because they don't know if you can make money. So in the early days you can never get that for it. So more likely than not, sadly you have to use equity until you can prove it out.

Exactly.

So, so what's the correct gold lilux match of like, you know, the correct betfellows? If those were the wrong bet fellows.

Yeah, so, so as I'd say the most important thing here is understand as an entrepreneur, whether you are or are not in a place where industry expertise, domain expertise is super crucial. Compare this to launching Twitter.

You basically had a good idea, a good front end designer, somebody who could run the servers on the other side and boom, the thing just takes it. You don't need to know a whole lot of stuff to, to launch Twitter. You don't need to know about, you don't need to be an expert on how people communicate with each other online. You just need some instincts. So, so some businesses need the domain expertise and then if you don't bring it as a founder, you better have a crystal clear plan for how you're going to get it. You Know either through advisors or an investor who can bring it or a c, you know, a layer of vice presidents who can make sure you're not making stupid mistakes. And in this case they probably should have just before they quit their consulting jobs, they probably should have spent a lot more time studying the apparel business to sort of figure out how they were going to fill those gaps. And then there's diligencing your investors. And I think it's easy to tell somebody to do that if you're desperate to raise money.

At the end of the day, if money's there from one person, you don't love them, but it's there, you probably take it.

Yeah. And then you know, even the partners for Quincy where the stuff gets manufactured, this is not uncommon for apparel in outside factories. And these factories would just push their stuff to the end of the line. You know, Ann Taylor would come in with something that needed to be expedited and okay, those Quincy orders because they had no industry reputation, no relationships, small orders, quirky sizing requirements. And so you know, what can you do about that? Well, you know, they probably, they could have given equity to a factory that might have helped.

They could have maybe used social forward thinking factory owner.

Yeah. And also risk. It's also risky because if it doesn't work out now you got a messy cap table and no way to divorce. So at the core I would say is really. And also the problem they had of not figuring out who the boss was. That's one that so many founders avoid that tough conversation because it's natural. Like at the beginning, you know, there are two of you in a room for three months and you sort of talk through everything. So it's very natural to extend that relationship forward. But it can also get you into a lot of trouble.

So yeah, I know, I figure I don't mind co CEOs but you need to have like different areas of focus. Like one does like fundraising and marketing and product and the other one does operations, you know, more co ish type. But the, and you can agree in the core strategic things where like things like design, there needs to be a decision maker. It can't be because otherwise you end up with like half assed version of both. That is not, it's not distinct.

Yeah. And they did divide the functions up. But just what you're saying it was product design where they both shared, they both felt they had responsibility and yeah.

No, that could be a problem Bastion.

You're nothing but your design.

So number two, false starts. I mean we talked about the people wanting to go and build immediately. But I could make the argument especially for like really smart consultants and HBS types, you could also want to test forever. And so how do you find that balance in cold Lilux? Like many people I know before, they want to make pull the trigger. Like what they need to prove is basically unprovable. At some point you do need a launch. So how do you come up with like the right balance?

Yeah.

I have seen teams that are analysis paralysis, you know, sort of too much studying but boy, that's. I think that's a small. And those folks are just not destined really to be entrepreneurs. And most of them will never pull the trigger. They'll do so much. If you do a lot of analysis, you can always find a reason why your startup idea is not going to work. So when they do too much analysis, they find those reasons and something in their head I think is telling them you're really not destined to take this plunge.

So what is the correct amount of research? What is it you think most people should do before they go and rush to build something?

I think.

20 interviews with people that are in your target customer segment. Steve Blank would want you to do 100. I'll live with 20. And by the way, each segment needs a new round of interviews and usually there's more than one segment you could be targeting.

A surprising thing that I see teams skip, which can be done pretty fast and boy, you learn a lot is competitor analysis. I don't mean like the moon charts where we've got this feature. You should do that of course, but I mean actually sit down with a potential customer. You know, you're building a dating site, have them log on to match and you know, sort of try to find a romantic partner and talk out loud while they're using it. And this is confusing and you know, the onboarding didn't make sense here and you can learn so much and you know, half dozen sessions like that. So I would say this upfront, some places a little ethnography actually watching people in the real world do the thing you're trying to do better. It doesn't always work. You know, if you're designing an online grocery delivery service, you can watch people walk around a grocery store.

I don't love focus groups, to be honest, because you get one person to dominate the focus group and their opinions start becoming the norm and it's like. So I don't know, like maybe watching individuals individually like use the product, but like focus groups, eh, I don't know.

Yeah, I tend to agree with you. It takes some training to do them well. You're right. They can get group think, they can get dominated. So you need somebody who can actually quiet the dominant voice and draw out the others. Take some real skill. Where it works well is products with strong emotional associations. And sometimes you can get people, if you can get the rhythm going, you can get stories triggered by other stories that. And I've seen it work really powerfully. But I agree with you, it's sort of in amateur hands. It probably does more harm than good.

I think we do tell our intruders the same. Like, do 30 interviews for each product that you're thinking of launching. Like and iterate and then show them the mock ups, test the landing pages, etc, Then once you and we have our checklist, like, is the TAM large enough? Do you think you have a value prop for the supply? Do you think you have a product value prop for the demand? What does competition look like? How differentiated are you from them? What are the union economics look like, etc. And once you've done that, okay, now we feel comfortable, we can pull the trigger.

Yeah, no, I love it. And I include in there a lot of.

Call them a paper prototype. Right. You sort of a drawn version, not a software coded version of the product.

Yes.

And getting feedback on that because usually there's multiple ways to solve things and you know the chance that you're going to get it right the first time. And there's an art actually to doing that and getting feedback on it. People, people tend to tell you they like whatever you put in front of.

That's also true.

So you know, there's a good trick.

There'S a way to ask questions like, you know, you can't. It cannot be a leading question. Do you like this? Because the answer is going to be yes. They want to make you as the interviewer or potential entrepreneur happy. So.

Yeah, yeah. So what I tell my students to do is put two things in front of the interviewee and ask them which they like better. You'll get much more honest feedback.

I totally agree. And I can't remember if it's called the mom test or something like that, but the way you ask questions matters a lot.

Yeah, yeah, no. For everyone listening, it's a wonderful book. Go to Amazon and buy it. It's not expensive. It's more wisdom in like 100 pages than anything you will buy on Amazon. And the premise of the mom test is if you ask your mother whether she likes your entrepreneurial idea, of course she does. Because you're her child and she feels sorry for you and she wants you to feel good. So don't ask. You know, the point is, don't ask people who are going to tell you what they think you want to hear. And there's a lot of good guidance in the book on how to pose the questions and not ask leading questions.

Exactly. So number three, false positives. How do you know and how do you identify that you have a false positive? Because when things start working, the last thing you want is like questioning, oh, why is this working? Is it going to continue? Or did I just get lucky? So it's really, really hard. And also you do want to build things for your kind of early adopters and you want to please them and you think they're going to bring you in the right direction. So like again, the gold lilacs of figuring this out is really hard.

Yeah, I mean you need the early adopters. So the question has to be, is there anything about this situation I'm looking at that isn't replicable?

And so the anchor case for this chapter of the book is a company called Barou. I think it's a French word. Fabrice. Barou is when a dog tilts its head sort of listening for trying to figure out what humans really.

Never heard that word. I did not know what that was, but I it's a great name.

It's my favorite name for a startup ever. Unfortunately, the startup didn't work. Barou was a dog walking service like Wag or Rover. The wrinkle here was instead of however, Wagon Rover get customers, I assume through social digital marketing. They were going to go to apartment buildings and basically have a relationship with the building and the concierge, new tenant moves in and they're going to refer, you know, somebody who's coming from a different neighborhood or a different city need a dog walker. So boom. And that actually was a pretty good idea and that worked. The false positive for Beru. There's a building in in Boston called the Ink Block. It used to be the headquarters and printing facilities for the Boston Herald newspaper. Big building was all converted to condos like all on one day, flipped a switch and filled up 400 units. And all of those people came from different neighborhoods or different cities. A good chunk of them were actually a Hollywood movie crew that was there in Boston for months and months on per diem, you know, brought their pets working 15 hour days so needed a dog walking service. And to top it all off, this was the winter of 2015 when Mother Nature dumped.

3 meters 9ft of snow on Boston in 30 days. It's unbelievable. And so Beru just did amazing business in its first, with its first customers and all these other business, all these other buildings in Boston heard about this and said we must have this amazing concierge pet service too. So she signed up another half dozen buildings just like that and had thought she was going to slowly build the business, took some angel money, thought she was going to slowly build it to profitability before she moved to the next city. And now sort of the, the opportunity started to loom large. Like wow, it looks like I could raise venture capital and supercharged this thing. And so she let go of her original plan to sort of build slowly within a year, launched in Chicago. A year after that, launched in Washington D.C. never really raising enough money to pull it off and never having the operations under control to the point where she could actually build the business. There's a, there was good, strong demand, but.

Yeah, but, but that's not a false positive. That's more a, that's more a nail it before you scale it like before you put pedal to the metal. Like make sure that like you're, you know how to make sure you're vetting the dog walkers correctly, you're managing your operations correctly. I mean she made a lot of mistakes. I think they, she wanted W2s opposed to contractors because she thought they were churn loss but they actually turned out to turn more or as much, yet costs a lot more because you'd pay them even when they weren't working. Yeah, I think there's clearly a false positive there as well because the false.

Positive here was, I think she was beguiled, seduced by the strong demand at the ink block and that gave her the confidence to scale before she sorted out these many, many problems you're pointing out. I think she'd have got there, she'd have just slowed down. I think she would have figured out that, you know, contract workers probably were the right way to go, et cetera, et cetera.

Yeah, but the false positive might have been like this building in a way maybe was unique because it was only new tenants who had a lot of money, who had dogs. And if you go to a building that has existing tenants who have existing dog walking relationships, this is an important emotional connection and someone's good at trusting someone who's going to take care of your dogs. Replacing them is kind of hard. Hard. Right, like so, yeah.

Switching costs are.

Yeah. So it works better for new people in a building than. So maybe there was A false positive like that building was rather unique because if I, if I read the rest of the case correctly, you know, some of the buildings that people weren't really pushing it all that much. Then the, then the. Meaning the, the building managers or whatever weren't really all that enthused. They signed it. But then. And then people's expectations of what the service was relative to what they delivered in the operational side was rather different as well. And then also Rover and Wag just came in and that was a. That was a pretty amazing. That was a pretty amazing service that worked really well. And I think originally there was a funnel. You know, the people were like, if I'm dog sitting, that's. It's a different business than dog walking. But actually they're the same business because the same dog owners at the same time, skill set, etc. So I think there are a lot of, like, things that got wrong.

Who do you let into your apartment?

Exactly.

Exactly.

Yep.

Interesting. My brother is watching the show and he says, olivia Grinda, I actually participated in the feedback for your book. Cool.

There he goes. And he got a copy of it.

Great.

Okay. We've talked about bad betfellows and false starts and false positives. Maybe you can go to the late stage speed trap. How do you avoid the speed trap? Because it's actually really hard. I mean, once you're growing really quickly, either like, go the. The transition of having generalists to hiring specialists and making sure they're loved and making sure your operations are scaling, hardly all that is really hard. And again, finding the right balance seems very difficult. How would you go about doing it?

Yeah, I mean, because the, because there's so many different problems, there's lots and lots of different solutions. But the, the real advice I'd give to a founder here is just basically, before you step on the gas and enter the speed trap, figure out what the speed limit is. I know this sounds corny, but I actually think the metaphor works. And the book proposes a test that I call the RAUI test. R A W I. I love it. Because basically, in Arabic, a rawi is a reciter of ancient poetry. So something poetic about using that word to sort of help an entrepreneur figure out should they keep going, should they accelerate. And so the RAWI stand for ready, able, willing and impelled. And the words are vague enough that you sort of have to define them by ready. Basically, do you have product market fit, Fit. But more importantly, can you sustain it as you grow? Because you're going to go be going after mainstream Customers and who may not be as interested as the early adopters. Is the total addressable market big enough for you to continue to grow and is your margin currently high enough to withstand the inevitable? You know, we can hope that the margins are going to expand because of scale economies, but usually it's more often the case that competitive pressures and going after marginal customers puts pressure on the. On the price, price, cost.

So on the ready by the way, before we go to the willing, able, etc. The to be honest, I've actually seen that as companies scale, the courts improve the for the most part for most companies, the business improves the because they get better at they improve, they keep multi iterating in their funnel and the funnel becomes more effective. They they figure out the correct customer segment for them, they target them. And so if you look at like the first cohorts from when they launched, six months later they're buying, they bought two times but then the core that started a year later, six months later they bought five times because actually they got the right people. And so many businesses clearly not Fab.com but many businesses actually I find that things do improve and the and that when you see these things you're like whoa, yeah, absolutely, we're ready.

Yeah, that's a healthy business. And.

I mean it forces you to sort of look into the future and try to be intellectually honest with what actually is going to happen. But you're right, I mean scale economy should kick in. You should be leveraging your overheads over a bigger, bigger volume. You should optimize the funnel. So if it's really working well, look.

At your cohorts or the cohorts improving and if so you're probably going in the right direction.

Yeah, Abel is because willing.

I think everyone's willing, right? I've never met a founder who's not willing to go with big.

I think it's still worth asking.

Because a couple of things are going to happen when you grow. You're going to take a lot more capital, you're going to dilute your equity stake. You're going to bring new board members on board. This is the point at which a founder goes from having control of the board dominant, usually a majority of board seats to a minority and now they can get fired if they're no longer doing the job as founder CEO, they can get replaced as CEO. And there are probably some founders there who are more attached to the leadership role than they are necessarily to the opportunity to ring build personal.

Founders or.

No longer CEOs after round whatever. I mean you mentioned that number in the book, sir.

This again comes from my colleague Noam Wasserman. The Founder's Dilemma book, his research. 60% of Series D companies are no longer led by a founder. CEO.

6060 at Series D. Okay.

Yeah, yeah. And so it's quite common to replace. And it makes sense, right, because the skills and attitudes and capabilities that let you get the thing off the ground are not necessarily, you know, we can, we can name a lot of founders who, Zuckerberg or, or, or, or, or Gates. I mean there are plenty who can manage the transition but you know, more, more often it's the case that somebody can.

Yeah, no, I actually had a call with the founder yesterday and he's like, look, I, and I was like, are you being pushed out or. And he's like, no, I'm like not. I don't, I don't. My COO is extraordinary and my new CFO is extraordinary and they're all in their 40s and I'm 29 and I love the 0 to 10. He loved the 0 to 1, but he actually liked the 1 to 10 as well. But like the 10 to 1000. Yeah, it's not that exciting and I think what needs to be done is clear and these guys can do it. I'm, I'm going to chairman, I'm moving myself and like completely voluntarily. And that's a Series B company. And, and I'm like, wow, that's very mature and made a lot of sense and like a lot of introspection and self knowledge.

Yeah. And, and that's, that's the willing stage. Sort of. Are you willing to. Personal sacrifices? You know, you're just gonna work your ass off for another five years, you know, and, and the, all of the problems you've seen in the early stage are just going to get amplified when the thing is bigger. There's more things that can go wrong. You know, there's the possibility of getting fired and, and so it's, it is worth it. But I'd agree with that. And by the way, it's not, it's not always your decision as the founder. Right. At a certain point it becomes the board's decision about how fast you're going to go. And that's important to keep on perspective. Able is all about do you have the team at all levels, frontline workers and then managers, special specialist managers who can manage the frontline workers and the systems in place. And you got to remember, an early stage company has nothing in the way of Systems, you know, managing operations, customer service, you know, managing the inventory.

No processes, no playbook. You make it off of the fly. But eventually you need, those have to.

Be put in place. And the ones that make sense for an early stage company are not the ones that make, you know, so they not only have to be put in place but they have to be migrated. And companies can get this wrong. So do you have the people and the systems and the capital or can you continue to acquire? So that's able. That's a whole bunch of questions. And then impelled is a tricky one. By impelled I mean you're in a market context, a market situation where the competitive dynamics and the structural attributes of your business and your business model force you to go fast. And these would be the kinds of network effects business. So many of your portfolio companies are marketplaces that have to harness network effects and, or switching. If, when switching costs are really high, that will cause companies to race to get first time customers. Right? Because switching costs are really high and it's the new tenants in the building have never been in the building before, switching costs are high, changing dog walkers. So if I can sign everybody up in that building, I'm probably going to keep them. And that will lead competitors to race to acquire. So are you in a place where people are going to race for customers either because of network effects or switching costs? Sometimes scale economies drive this and sometimes just irrational behavior. Markets get overheated and without those attributes you still have craziness. And then as a entrepreneur, you got to figure out, do I stay in pace with the crazy? Do I do it too? Or do I hang back, conserve my capital, assume these people are going to all of these bike sharing services across Asia, you know, that they're going to eventually crash and you know, and then, then I'll be in a good position to pick up.

Yeah, but, but some of the, some of the biggest mistakes, I think people are impelled. They're like, they think it's a land grab when it's not. When and if you go for the land grab and you burn every, every capital you have, and then because everyone's burned too much, the economics are underwater. You've subsidized either side of the marketplace. You're, that's a good way to die. And by the way, I think that's one of the reasons BP died is like we had good unit economics in three cities and we're like, you know, let's keep growing, et cetera. But they're like, no, there's shift there's carvana, there's room. We need to go everywhere and pedal to the metal. And that increased burn because of course in the new cities you don't have density, you don't have, you have completely underwater unit economics. I mean these network effect businesses, density matters a lot. And it ultimately burned too much relative to traction and couldn't raise the next round and we died.

Yeah, I think there's a profound level of confusion about when network effects should push you to sort of go all out for growth and a lack of understanding. A lot of people think markets are winner take all when they're not. And ride sharing. When Asia heated up, how many calls I got from venture capital is asking me, what do you think? Is it winner take all? Should I back one of these players? You know? No, I don't think so. You know, if, if, if Uber can get me the car in three minutes and Lyft is going to take four minutes because they don't have quite as many drivers, you know, I think they probably both can survive.

Yeah, no, I think what happens is there's some marketplaces where you can only sell one, right. A used good on ebay, that's unique. You can, you cannot put it in two marketplaces, you can only sell in one because otherwise you're going to disappoint the buyer in the other one. The, the issue.

The. So some marketplaces and like Craigslist may be winner takes all like these classified sites, but like Uber where you have, it doesn't take that much supply in any one region to give you three minute wait or four minute wait. The, the, the network effects are not exponential, they're logarithmic. Right. Like there's a massive increase in value from going from 10 minute wait time to 4 minutes by the marginal increase value 4 minutes to 3 minutes, not that great. And then it keeps going down. And so that, that creates an issue where you maybe you get scale effects and, and, but you don't get that many network effects at some point.

Yeah, 100% agree.

So what do we go to the next one and you call it the talents or what was it.

The next big area that large sales startups make, the ones where they don't. Or did we not finish the ready, willing, able and impelled?

No, I think we got the speed trap done. Yeah, that ready, able, willing, impelled is the test. And for the late stage cascading miracles, I wish for the audience I had a solution to this one. I mean so, so much of it comes down to the, to Keeping a charismatic monomaniacal founder CEO under control. And just by their nature, they're hard to control. We kind of don't want to control them.

Right. I think that one I think we can leave on the side because most ideas, very few ideas are like that and very few founders are like that. Going after the complete moonshot where everything needs to align, I think that's less relevant, I think for frankly our audience and 99% of entrepreneurs. But the next one, I think before the, the, the, the everything, everything needs to align was something about talents.

Oh, help. Help wanted. Yeah.

Oh, help wanted. Yeah.

Missing manage.

Getting. Getting your ops guy or whatever. So, so how do you. Yeah, even that one is really hard because it's, you know, like they, the organizations of generalists started rejecting the specialists. The specialists don't feel loved. When, when do you bring them on and why or. And how do you make it happen?

Yeah, there's, there's that issue of sort of when do you want them and, and if you move too early, what you get is the specialist tries to shoehorn whatever they were doing in some other company into yours. Yeah. And, and will probably prematurely build systems that don't make sense for your company and the cultural rejection will be all the stronger. The earlier they come, the less felt need there is for a specialist. The other problem here, you know, and Dot and Bow is an example of this. So the founder there, he was a product guy with a lot with deep insight on marketing.

The clever wrinkle for Dot and Bow was your room as a TV episode he'd come out of. He'd come from CBS Interactive, CBS Digital, and he sort of had an insight there that like, wow, people were actually wanted to buy the couch that they saw on the, on Friends. On the TV show Friends, they were actually sort of calling CBS constantly saying, like, how can I get the furniture? So he's going to turn all of the items in your room into characters and you know, he'll name them and they actually did a really clever job. And then people see the room and how well it hangs together, just like a professional decorator would do. And they want to buy everything. So, so it's actually good for generating repeat purchases and big tickets and this worked for them. But the guy didn't know anything about integrated logistics. And so as a CEO in that situation, which is a very common situation. Right. To not know anything about the crucial function. You know, a product person who knows nothing about performance marketing, how are you going to find you don't have a network that's rich with these specialists you have no personal basis for. Actually, if you interview somebody for figuring out are they good or not.

And by the way, that was also the early stage problem of Quincy. It's like they didn't have anyone who, like, knew how to make dresses and deal with factories, et cetera. So that problem can happen in both cases.

This is the Quincy problem carried to Series C, essentially.

Exactly.

And it gets more serious when you've got tens of millions of dollars at stake of inventory and sales.

And so the solution here, if there is a good one, is basically to have a board who can help with this. You know, what you want is somebody who can, with a. With a network who can bring you candidates and do the specialty part of the interview. You, the CEO, can still figure out, does the person have the right attitude, you know, will this specialist fit into the culture that I've been building? So you can do that part of the interview, but you gotta rely on other people.

You do need it. Yeah, because I think they hired someone who, on paper, looked good. Good, but actually was completely the wrong person. And so they've got the wrong ERP and like, AI destroyed them.

They actually made. They made two mistakes. The first person they hired was a classic COO type. And I think Anthony Suhu, who was the entrepreneur, correctly saw he was going to need a COO type. And so they brought in an operations generalist who had never done anything like E commerce fulfillment. And the second person they brought in was from Netflix, who in theory had, like, learned how to do operations and logistics. But these, if you remember the little red envelopes that they used to send the DVDs around it, you know, shipping out hundreds of thousands of those is not like shipping out 30 couches.

Yeah, exactly. Very different.

Yeah. And the guy was actually a bad cultural fit. You know, he brought in a whole bunch of big company, you know, sort of massaging the number kind of things that just sort of set the founders on edge. So, you know, he accomplished a lot, but, you know, he didn't quite nail the problem. Yeah. So it took three tries. You saw this at Jibo too. You know, it took a long time to find the person. This is the social robot that you'd have read about early in the book. Took them a long time to find the person who could actually architect something as complicated as a robot that actually will listen to you and generate a conversation.

I think in their case, building that was really hard. So maybe it's not surprising it took that long. I mean, if I look at like on Indigo Indiegogo and all those things like end up being way late because they people underestimated the engineering. But I think their problem was less that because it was more Alexa. You know, for 200 bucks you get essentially all the things, same thing without the robot. And that was enough. And that's what people wanted.

Yeah, yeah. No, it killed him. And it does raise the question sort of how often could you possibly have seen that coming? Most people did not.

I mean. Correct.

We thought Siri and yeah, we expected it to be in our smartphone, but nobody had conceived of a thing called a smart speaker. And when it happened and when it cost $200 instead of $900, they were finished.

So I know you need to go.

For your open house or session or whatever, any last minute words of wisdom or anything else we didn't cover or anything. People should. Yeah. Parting words for how to not fail.

So we've spent a lot of time on the failure patterns and how to avoid them. There's actually a third part of the book which we didn't get into at all, which I'd encouraged. So for the entrepreneurs and the aspiring entrepreneurs who are listening, the third part of the book is actually as a founder, how to fail. Well, so if you follow all the advice and you sort of spot the patterns and you take the avoidance actions and you still fail, you probably will, by the way. I mean, just statistically.

How do you figure out first when and whether to pull the plug, which turns out to be a really difficult decision for an entrepreneur and many will defer it for too long. If you sort of ask failed founders, do you wish it shut the thing down earlier? The answer is very often yes. You know, I could have moved on to something else. My team could have moved on to another project. I could have given some money back to the investors, you know, and there's just a whole bunch of pressures on the entrepreneur to.

Keep going. Right. Self image entrepreneurs are persistent. People are counting on you. See, your employees are counting on you for their livelihood and their medical benefits. There's a whole bunch of moves you need to try out before you pull the plug. You're going to try a pivot, which takes time to figure out if it's working, going to try to raise more money, going to try to sell the thing. That's a particularly.

Slippery event because basically you get a lot of enthusiastic early response. Everybody wants to see your company, right? You announce you're up for sale. Every competitor wants to see, who have you got on your team, how much Are you paying them? What's your product really look like?

Yep. I know, like people will take meetings, but then they won't give you an offer or they'll be, or will be pennies on the dollar, you know, it'll be an acqui hire. I think. The way I think about it, by the way, is if you're, if you're unit economics, you're underwater and you're like 10 to 1 away from where you need to be, you know, you don't have product market fit, it's time to shut down, you know, or if it worked for the first, you know, 100k of spending per month, but like you can't find acquisition channels where it keep, it keeps going, you know, and you're 10 and one away. You can shut down if you're 50% away, frankly, most likely through iteration. You're going to get there. Yeah, you're going to figure it out. Yeah, I know. If you, if you don't need the magic bullet and in terms of how to fit, just close down, I would say do it, you know, tell everyone. Why do you try and, and now you take the responsibility because actually it's a pain in the neck. Right? Like you have to go and like close the llc, pay all the outstanding invoices, go through bankruptcy if need be. But you have to do it yourself and treat your investors well. Because if you just like run away, you're going to be blacklisted. Like, no one will ever want to work with you ever again. So, yeah, fail.

Well, I've heard that from a bunch of VCs. The worst thing in the world a founder can do is sort of throw the keys on the table and say, you know, good luck, I'm out of here. There's no equity upside for me. I've got better things to do. You guys figure out how to wind it down. Yeah, like terrible, terrible move.

Yeah.

Yeah. And just in closing, you know, a failed founder to the point of sort of owning it the way you just said. The mistake that I see a lot is some people is very human nature. To blame other people goes back to the fundamental attribution error. My co founder screwed up. My investors didn't sort of hold their end up. And those founders probably haven't learned much from the failure. There's another extreme, which is actually people who beat themselves up too hard for what went wrong. I am ill suited for this. Maybe they are. Some people really just were never suited to be an entrepreneur. But more likely they did a lot of things right. And A lot of things wrong. And if you can just learn from it and figure out whether you should do it again. Half of the failed founders I looked at go off and found again, you know, hopefully. Hopefully with some learning.

I think the success rate of second time founders were failures. The first time is actually higher than first time founders, which actually is good. Means on average they're learning something. And yeah, I mean, if you approach it with humility and like, okay, these are the things I did wrong and I'm not going to do those mistakes in the future, people will give you credit for it. I mean, less true. Maybe in other gos or countries where people like, there's a stigma and failure, but the US is extraordinarily forgiving. You know, you're only. And it goes in both directions. You're only as good as your latest success, but you know, your last failure is forgiven. As long as what you're doing today is compelling, that's fine.

Yeah, exactly.

Perfect. Well, thank you for joining. I'll tell the entire audience. This is the book again, why startups fail. Go buy it.

It's amazing.

You'll learn more insights of what we discuss here. And yeah, be founders and try to learn from other people's mistakes as opposed to doing them on your own. It's very hard. No one actually ever does that, but try it.

Fabrice, thank you. That was a lot of fun.

Thank you, Tom, for joining. Thank you.

Bye bye, bye bye.

Okay, so you're.