Why everything we’ve been taught about quitting is wrong — Annie Duke

 All right. I'm excited for the conversation. Annie let's make it happen.

 so in, building venture back technology companies, I I'd say in general, we kind of prey at the altar of never quitting or giving up. and I think that there are good reasons for that, right? We've seen in companies that we've partnered with, like Roblox or notion, it often takes years and years and years of grinding to get to that inflection point.

 You also just recently saw it in Figma, a, a similar story of years and years of building and iterating and pivoting to get to a very, very large outcome. Um, and so I'm curious, what do you think is wrong with this line of thinking?

 oh, gosh. Well, well, first of all, let me just say. A little bit what's right about it. And then a little bit of a redefinition if I can, , so the first thing I wanna say is like, I know that my book is called quit, but in no way, do I wanna say that grit isn't an important trait.

It's, it's not a muscle that you should develop because the thing that grit does to your point is it gets you to stick to things that are worthwhile, but also very hard. And most of the things that we do that are worthwhile, that are gonna change the world, um, are gonna have periods, you know, which are really, really difficult.

And grit is what allows you to power through that, where we go wrong. In thinking that grit is just good period. Right? So it's true that in order to be successful at something, you will have had to stick to it, but that doesn't necessarily mean that sticking to something makes you successful. And so when we look at something like a notion or a Roblox or a Figma, there's a temptation to say sticking to it is just good. But the problem with that is that that, that, uh, grit that we can show that allows us to power through things will also get us to stick to things that aren't worthwhile. And there's value in understanding the difference. Because if you have a founder who's incredibly brilliant, right? Who is really gritty, who has the ability to change the world and they're butting up against a problem that is insoluble.

That's never gonna work. They can achieve product market fit, right? They can't get into the market that they want to and start to grow. And it's very obvious looking at it that it's never gonna work out. It doesn't make sense for them to keep pounding at it because while you might have the temptation to say, but Roblox or, but notion, we know that those situations are somewhat rare and we need to be able to tell the difference between when there's enough of a probability to be able to achieve that kind of outcome.

And when it really is someone just really honestly wasting their time when they could be switching to something that would absolutely change the world. And that's where it gets really hard. Cuz it's hard to tell the difference between those two, but even so you mentioned the word. And we know that. For example, with notion, there was a pivot involved.

Well, a pivot is a quit. And I think that we also need to remember that, that quitting isn't just about the whole thing we're doing, right? Like we're pursuing a company and trying to bring it to exit it's that within a company, when we start to think about developing products, or we start to think about projects that we might engage in, or we're thinking about strategic initiatives or what our strategy is, or, you know, are we gonna be B2C or, or B2B or whatever it is that when you butt up against those things within the larger goal that you have, that aren't working, we also have to be able to quit those and that when you see these situations where things look on the brink of.

and then they sort of rise from those ashes, um, into success. There's a lot of quitting that's usually happening, right? They're not just pounding their head against a product that isn't working. They go back and they hide in their room and they completely redesign the whole thing.

And then all of a sudden you have notion coming out of it.

 So building on that, when you've looked at these, what we would define as slow bake or long build companies, do you think more often than not, they were executing the way in which you think is optimal or no, they just, if you get a thousand people to try to do the thing, one out of a thousand it'll it'll work.

 Yeah. You know, I mean, it's the thing is it's hard to say. Um, but the base rates would tell us that it's the latter. right. And that's the thing that I think that we have to remember right now in, in general, there is gonna be a difference between people who are sort of pounding the same pavement. And it just takes a really long time before they hit versus people who are pounding the pavement, and then they figure that pavement isn't working and then they, what we would call a pivot, but which is really a quit while keeping the company sort of intact, right?

Like I'm still pursuing the same company, but now I've made some sort of switch within it. Um, I think that those tend to be the ones that do better. Right. Uh, the fact is that it's gonna be very rare when you get feedback from the market for that long, you know, for five years or 10 years, that you're sort of banging against the same, um, wall, and you're trying to develop the exact same product or the same community or whatever that, um, it's the case that.

You're supposed to necessarily keep going, because I do think that that will happen sometimes, but the question is, is it gonna happen enough of the time? And I think the answer to that is probably not

 How do you think about the individual relative to the system? Meaning, and we should talk about re-explain, what base rates are. Cuz it's such a powerful idea that you've taught me over the years, that I think a lot of that, that tends to be underappreciated, but you know, meaning that the base rate could be very low.

 so that means that the average individual stepping into the system doesn't make sense. But that the one in the thousand is so transformative that, that broader society where it works for the whole. Do you think about that at all?

so yeah, so let's, let's just start with just reminding people what base rates are. Uh it's how often does something happen in the situation that you're considering? So there's all sorts of examples of base rates. Um, an example would be, uh, how often. Year over year, um, in any given fiscal year is the stock market up from where it started at the beginning of the year.

 Right? So that that's a base rate. Um, I think depending on how you calculate it, it's, it's, I think somewhere around 70%, 75% of the time the stock market is up. And what that means is that, um, it doesn't matter if over the last five years it was up all five of those years, um, the next year, unless something significant has changed, but the next year, uh, it should be up about three quarters of the.

As an example, um, we can think about base rates, which, which we've talked about, um, in terms of valuations, right? So if historically, uh, you know, the top SaaS companies are valued in about 16 to 20 times, revenue then, um, if all of a sudden there's one year where they're being valued at 40 to 60, um, times revenue that would.

Odd compared to the base rate. Now that doesn't mean that that valuation might not be true. It means that something has to be really substantially different than other companies or something has to have substantially changed in, in the environment that would make you think that the base rate that you would calculated before it didn't hold anymore.

So I I'll give you two examples of that. One would be, um, if you look at the valuation for retail companies, uh, it was very different in the nineties than in the early two thousands in the odds. And the reason was that retail went from brick and mortar to on online and that's a paradigm shift.

So at that point you wouldn't use, you know, how has Walmart or Kmart, or target, um, or whatever been traditionally valued in terms of trying to figure out what the valuation for Amazon is because there's a paradigm shift. Another example of that would be, it doesn't make any sense to look. You know, historically like say over the last a hundred years at either the strength of hurricanes or the frequency of hurricanes, because we know that there are big shifts in average temperature and how warm the ocean is, uh, that are pretty recent, um, that are making hurricanes stronger and more frequent now.

And so those base rates wouldn't hold because something significant has changed about the world. So when, when you're asking me about, you know, if it's one in a thousand that are actually gonna be able to break out in that way for those slow bake companies, does that make sense?

We're really talking about a base rate there. And your question is, um, in that case, should we expect that there would be something different than the reference class that you're thinking about, which is sort of, of all companies that are being started? So I, I hope that's, I hope that's a clear definition of base rate.

I'll kind of throw that back to you to see if it's clear.

 It is, I guess B before we, we loop back to the original question. Can you explain maybe why this idea is so useful and maybe why it, it, it is often not used.

 Sure. So, um, the way that people actually make predictions and forecasts tend to be something like this, uh, you get some information, um, you're sort of thinking about it in relation to ship to your own experience. Uh, and then you make a forecast. well, I mean, here's a simple example, like you're sort of making predictions for yourself about how long it's gonna take you to finish a project.

So you're thinking about what's the project, uh, you know, what, what I sort of have to accomplish in order to complete the project. And then, um, you sort of make a determination for yourself, how long do I think it's gonna take? The base rate asks you to do an extra step, which is to say, don't just sort of look at the information that you have and then think about it in, in relationship to your own experience.

Uh, take one step in between, which is look at the information you have, think about the problem you're trying to solve, and then sort of try to figure out how that generally goes before you start to think about your own experience. And the reason that you have to do that is because the way that we look at the world, the, our perspective on the world is, is actually quite riddled with cognitive bias.

So there are things like availability bias, for example, which is just that things that are easier for us to recall, we think of as more frequent. So we wanna go and look at a base rate because that's gonna help us sort of solve that problem of our own availability. So, as an example, if I were to ask you to estimate, um, the probability of a terrorist attack on us soil, uh, your estimate would be.

Much higher than what the actual base rate is and the reason it just is that it it's easier for you to recall cuz those things tend to be on the news. Um, and so we think about those as more frequent than they actually are because, uh, ease of recall becomes sort of a proxy for our judgements of frequency.

So we'd wanna actually go look up the statistics on that, um, in order to discipline that. And that's true in every way, like there's something called the planning fallacy. So if I'm trying to figure out how long a project is gonna take, I'm gonna tend to underestimate how long it's gonna take. And I ought to look at the base rate for that.

There's also something called illusion of control, which is that we think we have more control over outcomes than we do. Uh, this is also included, um, sort of similar to this is the gambler's fallacy that we think we have a higher likelihood of things happening, um, than they actually will. And if we wanna actually discipline those cognitive biases that make it so that our forecasts, that our predictions so that, uh, our decisions are inaccurate.

One of the best ways to do that is to go, um, look out and see what's going on in the world in general. Right? And if we go back to say venture valuations, I heard a lot of people in the summer of 2021 talking about, you know, the world is different and really convincing themselves that the valuations in part, because they were investing at those valuations, right.

In part, because they were seeing other founders raise at those valuations. It's very easy to convince yourself that the world has somehow changed and that this is the way it is. And that you're seeing something in these companies or the companies have fundamentally changed in terms of the, the way they're executing or, or, uh, what part of the market that they can capture.

Um, but again, unless there's some sort of paradigm shift like that, Walmart to Amazon shift, going and looking up the base rate and saying, well, what, what is the historical base rate for, for valuations of companies of this sort helps to discipline that desire that we have to think that something special is going on.

And again, I wanna be clear, it's not that something special isn't ever going on. It's that it gets you to sort of think about the other side and at least have to disprove that the base rate would hold in that situation.

 So going, I, I think that was a great explainer. So going back to what we were talking about a second ago, have you sort of contemplated, you know, in, in the type of outlier systems, like venture the difference between something being bad for the individual and good for the broader ecosystem?

 So I absolutely have, I mean, let's think about it from an evolutionary standpoint, right? Uh, a bunch of, um, humans ventured out to far away places because, uh, to seek resources because human beings in general are curious whatever. And obviously for the individual that did that, it may have not been.

Positive expected value, right? For any given individual who went and ventured off to go to Europe or whatever from Africa, it, it may have been, uh, not a great choice for them individually. In other words, the probability of death may have been higher than you would expect someone to be willing to tolerate.

But that one person who bucks the odds right. Uh, and gets through, it's really good for humanity as a whole. And that's true in anything where you're gonna see power law. So you can think about like, these migrations are kind of power law, right? Like the chances of success, the chance actually getting to the destination that you wanna get to safely is gonna be in, you know, it's incredibly rare, it's small, but if you have enough people trying.

Then somebody's gonna break through and that's gonna be really good for the system as a whole. And that's true with innovation as well. Like any individual founder who might be working on a problem, the probability of failure is really, really high. And while there are certainly lots and lots of founders where the payoff is positive, expected value, right?

Uh, even in the case of, of the chances of succeeding being really small, there are certainly lots and lots of people who are working on problems, where the payoffs don't actually justify the low probability of them doing it. But it's good for humanity that people do do that because you know, the lowest probability is gonna happen sometimes.

And when it does happen and it's successful, you know, it's a big innovation and it's amazing. So, you know, I think this is something that, that it, you know, is kind of a paradox and we need to think about it, that you can have a lot of individuals where personally. They're losing to the decision long run, but humanity or, uh, is, is winning to it because humanity is kind of holding a portfolio of people who are trying to innovate and make the world better.

And the fact that people are willing to do that, even though they individually might not be winning to that decision is really good for humanity as a whole.

 So switching gears slightly, one of the things that I've noticed in working closely with. Early stage companies over a very long period of time is for whatever reason it feels like when things aren't working more often than not a founder makes incremental changes

 you know, an example would be, you know, our go to, market's not working.

I just, once I hired the right salesperson, everything's gonna work. Have you sort of observed this and, and sort of, what do you make of, of incremental change versus drastic change when things aren't working?

 Yeah, so, oh gosh, there's so many different ways to go with this. I mean, you know, it seems like such a simple question. Right. But it's actually incredibly deep. Like there's so many layers here. so let me start with the first layer, which is that, you know, you are viewing from the outside, right? As a viewer from the outside, as someone who's observing, you can see that. No, it's not that you just need to hire the right salesperson. It's like, you need to change your, go to market motion.

 drastic. Right or the product or whatever that something really needs to drastically change. You need to go hide in a hotel room and completely start over. but, but what's interesting. And, and, and I imagine Brett, that you've had that feeling when you're looking at, which is why can't they see that, you know, why don't they see what I see?

And the fact is that we all have this intuition, this, this is the intuition. They have this very strong, which is when I see signals that the world is giving me that the thing that I'm doing, isn't working, obviously I'm gonna change gears. And this is a really important intuition because that intuition is kind of what allows us to make decisions under uncertainty in the first place.

Like when you first think about. You know, what product am I gonna develop? Or, uh, you have a product and what's my, you know, go to market motion, gonna look like, or whatever you're doing that under conditions of uncertainty, right? Like there's just a whole bunch of stuff you don't know. And there's gonna be a lot of information discovery, post decision, right.

Post having started something. So when we start things, there's a lot of stuff we don't know. And then also the world can just intervene, which is called luck. And so after we've actually started, uh, we're gonna learn a whole bunch of stuff. Um, so how do we then make a decision when we're making a decision under so much uncertainty?

It's because when we learn that new stuff, we can change our minds. we can do something different. We can walk away, we can change course we can quit. So, and we have that intuition that when we see those things, when we discover that information, that obviously we will quit. So let's just start with the first thing, cuz this is really like important for people to understand is that that intuition is wrong.

When we get signals from the world that are negative, that the thing we are doing, isn't working, we actually escalate our commitment to whatever it is that we're doing. We don't wanna abandon it. And there's a variety of reasons for why that is one of them, which I think is pretty familiar to people is called the sunk cost effect.

Um, which is when we've put a lot of effort into something. We don't want to switch because we think that then we'll have wasted all of the time or the effort that we put into the thing. So like the simple example of that is if you buy a stock at 50 and it's now trading at 40, um, you don't wanna lose the $10 

And at that moment that you might sell the stock, that's when you feel like, but then I, I will have lost that $10, but of course that $10 is already lost. It's already sunk. And what should matter is, is, is it worth it to go forward? And that's true in the case that you're talking about, right. It's broken.

 if I just, you know, hire another salesperson, get the right person, it's gonna work out that part of what's going on. I, I don't, if I switch then all the work that I put into this motion isn't is gonna have been wasted, but really what really matters is, does it make sense to spend resources, time a, a great salesperson's energy executing, the strategy that you have going forward?

 So that's kind of like piece number one is, is just this reaction that we have to bad news, where we tend to double down on what we're doing. And then piece number two, which I think is really important in terms of this kind of issue is that we really have a preference for the status quo. And once you've established to status quo, which in the example, you're giving me, there is one, right, which is the, the strategy that you're executing today.

Um, it's very hard for us to switch. And part of the reason that it's hard for us to switch is that we are much more tolerant. Of bad outcomes that come from the status quo. Then we are tolerant of bad outcomes that come from something new. So when we switch and get a bad outcome, so you can imagine thinking this, but what if we throw this, you know, throw this strategy away?

What if we just completely say, Nope, that go to market motion, wasn't working or that product wasn't working, I'm gonna dump it and go try to develop something else. We have this fear of, but what if that doesn't work? And when it doesn't work, we feel regret there's an as asymmetry to the way that we process that regretted.

It does add it not working. And we feel it much more keenly when we started something new. Then when we lose kind of like on the individual path that we're on. So now we have sort of two forces working against us. We don't wanna abandon the work. And so we'd prefer to make incremental changes, hoping that we can turn it around.

and if it doesn't work out, if it continues not to work well, we're just much more tolerant of that than we are in terms of toleration of the bad outcomes that come from switching to something new.

 So, so sort of given that framing, if this fictitious founder that we're talking about called you and said, Annie, what should I be doing? What, what is the advice and or tools that you would give him or her.

 So I think one of the things that that's really important to realize is that it's, it's really hard to do it in the moment. Like when you're actually facing, like, am I gonna just throw this product out or am I gonna change my strategy? Um, it's hard to do it right then. So generally if you're trying to coach somebody, well, the first thing I would say is it's really good that they came to somebody else because one of the best strategies you can employ in order to get better at making these decisions to quit is actually to get somebody from the outside to help you.

 in the book, I talk about Ron Conway who really prides himself actually on coaching founders to quit. Um, it's kind of like people would be surprised because, you know, obviously Ron, Ron Conway has helped people really grit it out, uh, to great success. But he, I think he sort of thinks the thing he's proudest of is when he sees that a company is not gonna work, that he feels an obligation to the founder to free that founder up so they can go work on something amazing because the thing they're working on, isn't so amazing.

 And he, he sort of steps into that role because again, we can't see those things for ourselves. It's hard for us to get to those decisions for ourselves, but we can generally see them pretty well from the outside. So getting outside help in these things is really, I. But then the second thing that you need to realize is someone who's in that coaching position, is that the person's always gonna say, it's that what you said?

No, I know I can turn around. We just need to get the right salesperson in here. We need to make some tweaks to, you know, the way that we're messaging, you know, whatever it is, right. Whatever the things are that they say they need to do, that they need to make these small incremental changes and that, they're gonna turn it around generally.

It's a good idea, actually, not to disagree with them in that. And this is actually the strategy that Ron Conway uses. Um, and just say, okay, you're a brilliant person. I have no doubt. You can turn it around. But then what you have to do is follow that with, but what does turning it around look like within what time?

So in the case of this fictitious founder, that comes to me, I would say, okay, um, you think you can turn it around and you're gonna be able to start actually hitting your marks in terms of generating net new ARR. how long would we have to wait to see that change occur?

So in this case, they might say, well, you know, it's a long sales cycle. Uh, it might take six months and then I would encourage them to say, well, it might take six months to close new business to close the business, but you should be seeing changes at the top of the funnel. If it's working. Can we agree to.

And they'll usually say yes. And you say, okay, so let's say that this new salesperson really does what you say they're gonna do. Like how long would it be until you start to see the top of the funnel start to populate? They'll say, you know, within two weeks to a month and I'll say, okay, great.

So let's call it a month. What does it look like if we look at the quantitative data in terms of what's happening at the top of the funnel, what are the numbers that we're gonna see there? So now we, I sit down and then I would partner with them and get them on a pretty tight timeline and try to get to what are the leading indicators that things have actually turned around.

So I wanna get to the leading indicators, cuz I don't wanna wait until it's six months from now and they, again, haven't generated. New ARR. And they're gonna tell me again that they just need to tweak things. So I need to get to the leading indicators and then we actually write down what are the benchmarks?

What are the things that we're gonna see that are gonna tell us that things have turned around, uh, within the next month? So we work on that and then I would say, okay, so let's agree. We're gonna revisit in a month. And if you're not doing that, then something drastic needs to change. Then we really do need to think about quitting or pivoting.

Um, uh, I prefer the term quit, cuz I don't think we should hide behind the term pivoting. But um, so now what you've done is you've gotten the founder out of the moment so that they can sort of think about some version of themselves in the future.

You've agreed with them that you think they can turn it around, but you've gotten them to tell you what does turn it around mean? And now you have benchmarks and those benchmarks now become what I call kill criteria. If you miss them, we're gonna kill. We're gonna quit. And if you hit them, then we're gonna go along our Merry way and be really happy.

Now, one of the things that you might be saying is, but, but Annie, if you see it today that they need to change, isn't it better to get them to change it today? And my answer to that is of course, in a perfect world, if you can see very clearly that hiring a new salesperson is not gonna help the new seller, isn't gonna do it, right.

This is an incremental change and they actually need to do something more drastic then. Yes, of course, like in an ideal world, they would just stop what they're doing today and, you know, quit and go do something else. But it's not an ideal world and we need to realize we're dealing with human psychology and it just turns out that, okay, now they might not quit for a month, but.

That's better than them continuing down this road for six months to see that they haven't closed the new business, that they missed their number two more quarters in a row. And now maybe they've run through their capital and they can't even go out and raise, which is death. So if I can get them there in a month, instead of six months, that's a huge win, right?

If I can get there where maybe they're gonna burn a, a few extra, you know, maybe they're gonna burn a hundred thousand more, but it's gonna save them from burning 600,000 more. Then that's a huge win because I'm getting them there faster in a way where they can process it and actually execute on it. The other thing that comes from that is that when you really sit down and define what does success mean?

What comes with that is a question about what do the inputs need to be in order to create that success. And now you start to actually get to the real heart of the matter is, are the inputs correct? And that brings clarity to the whole conversation anyway, just as sort of like a side benefit of it.

 do you think about there's a couple jumping off points that I wanted to go to. One is how do you think about pivoting versus actually quitting and shutting down?

 Yeah. So a pivot is quitting. I just wanna be really clear about that. we think about quitting as leaving the court. Um, you know, pivoting is still quitting, right? If, if you were developing, um, game never ending, and then you pivot to slack, that's quitting, you quit developing game, never ending. And you started developing slack.

Um, you know, notion is a good example of what people sort of call a pivot, but it was really a quit. It's just like I'm throwing out all this code and I'm just gonna build in a new, so, um, I think that, that the, the distinction is in one case you have this sort of broader goal of, uh, really creating a successful startup.

And that's the thing that you're abandoning when you shut the company down. Now there's no question that psychologically that is much harder to do, um, because obviously that's the moment. That in, in the largest sense of the were toward the goal that you were ultimately trying to achieve, which is say, you know, create a venture scale business that actually exits, um, say over a billion dollars, um, or has a great exit.

Um, that's the moment that you have to say it was failing before, but now I have failed. That is what it feels like to us. Um, and that's really hard because that's, when you can't turn it around anymore, you don't have any chances. So pivot what we think of as pivots, which are really quit are, are easy for us to handle cognitively because we still have a chance to make the company work.

Even if we have to abandon the product that we were initially developing, for example, but I wanna make it clear that. Quitting and saying, this isn't the right product we need to switch to a different product that we're developing is also very difficult. And we tend to get those to those decisions too late as well.

They just tend to be a little bit easier because they're within the context of a, of a broader goal that we, we don't have to give up. And that's the thing that's really hard for us is giving up those broader goals,

 So when should someone leave the arena versus pivot?

 you know, isn't that the whole thing right there. So I can give you the objective answer. Um, the objective answer is when your expected value goes negative in comparison to other things that you could be doing. Um, so what do I mean by that? , uh, when we start anything, there's some sort of expected value, um, that you can calculate, which is just, uh, essentially you can think of it as like, does the upside, um, uh, outweigh the downside.

so, uh, when there's more upside, uh, you know, when the upside is, is great enough to, um, outweigh the downside, uh, then we would be, um, we would say we're winning to the decision is another way you can put it, but it means we're positive, expected value. So positive, expected value. I wanna be clear. It doesn't mean that you're not gonna realize the downside, right?

It doesn't mean, uh, just because, uh, you're winning to the decision to have started a company. Um, it doesn't mean that the company's not ever gonna fail. It means that if you were to sort of do that over and over and over again, that would be a positive, expected value decision, um, that there's enough upside to, to compensate for the downside that's associated with it.

Okay. So we can think about that just in terms. The road that we're on when the expected value is negative, we ought to walk away, but then we can broaden it to say, I can be in situations where I am gaining ground to my goal, right. Where, uh, there is enough upside to compensate for the downside, but there's other things that I could be doing, which are associated with even more upside.

624ba4fe4e53ef71c9b992c4

Annie Duke is First Round’s Special Partner of Decision Science. On the heels of releasing her new book, “Quit: The Power of Knowing When to Walk Away,” she dives into all of our misconceptions about quitting and makes the case that getting better at this skill lays the groundwork for greater success. 

Why everything we’ve been taught about quitting is wrong

On the heels of releasing her new book, “Quit: The Power of Knowing When to Walk Away,” Annie Duke dives into all of our misconceptions about quitting and makes the case that getting better at this skill lays the groundwork for greater success. 

Why everything we’ve been taught about quitting is wrong — Annie Duke

Read the article inspired by this episode

Grit or Quit? Advice for Tackling Tough Startup Decisions