A few hundred years ago, Renè Descartes, the fella who brought us the Cartesian coordinate system 📉, isolated himself for a while to figure out (among other things) whether we can know anything for certain. The result of his retreat was his Meditations on First Philosophy, one of the most important philosophical texts of the modern era and one of my personal favorites.
Early on in his Meditations, he decides that it’d be useful to engage in a thought experiment. He says,
…I shall suppose that some malicious, powerful, cunning demon 😈 has done all he can to deceive me…I shall think that the sky, the air, the earth, colours, shapes, sounds and all external things are merely dreams that the demon has contrived as traps for my judgment. I shall consider myself as having no hands or eyes, or flesh, or blood or senses, but as having falsely believed that I had all these things.
Descartes invokes the demon to figure out if there’s anything that can be known with absolute certainty. He finds that even if an all-powerful demon attempts to deceive him about everything he believes, there’s one thing he can know for sure: he knows that he exists. Even if the demon deceives him about everything else, he can still be certain that he exists since deception presupposes someone who is being deceived. This thought eventually leads him to the famous “I think, therefore I am” conclusion. 🤔
Hold that ☝️️ thought for a second.
Lately, I’ve been questioning a lot of the lessons I think I’ve learned about business over the past couple years. Between my second reading of Thinking Fast and Slow, a book about how our minds trick us into believing silly things, and my recent first reading of Leadership BS, a book about how much of the leadership/business focused content we consume is little more than bull-shit, feel-good stories devoid of any real scientific backing, I’ve started to worry that my attempts at learning about business are misguided — that I’m a fool looking for knowledge where there is none to be had. 🤷♂️
I see my worry as similar to the one that led Descartes’ to write his Meditations. They are both worries about what can be known given imperfections in our sources of information. The only difference is that my worry is scoped to business-related knowledge instead of knowledge in general.
Since my worry is similar to Descartes’, I figured it’d be helpful and fun to use the same sort of thought experiment to address it. The demon is really a sort of colorful abstraction over the various imperfections of our sources of information.
To address his worries about what can be known in general, Descartes imagined an all-powerful demon and asked if there was any knowledge that would be safe from his power. Similarly, to address my worries about what can be known about business, in the first section of this essay, I’ll sketch a demon who has the power to trick us in two ways we are often mislead. We’ll look briefly at some of the ways we’re mislead by feel-good stories and imagined causal forces.
Next, I’ll explicitly point out the types of business content that can’t withstand the demon’s powers. In other words, I’ll point out that some of the things we thought we knew are actually BS 💩. I’ll list out a bunch of books (e.g., How Google Works, Exponential Organizations, etc.) that I’ve read over the past few years that I’ve probably taken too seriously as guides to business.
Our Cartesian Demon
What powers shall we give to our demon? Again, we’re giving him powers that correspond to imperfections in our information about business and imperfections in how we process that information. Let’s start with the over-representation of information that makes us feel good and how that over-representation translates into errors of judgement.
The Demon’s Temptations of Feel-good Stories
One of the most useful points of Leadership BS is Pfeffer’s argument that because people who create leadership content are often trying to make money 💰, we should be suspicious of the content itself.1 In particular, we should expect information that makes us feel good to be over-represented since feel-good stories sell better.
For example, consider Simon Sinek’s inspiring story during his “Leaders eat Last” talk — a talk that get’s explicitly called out in Pfeffer’s book. In this story, a submarine captain makes several small changes that supposedly results in his crew going from mediocre to the highest rated crew in naval history. ⚓️
Sinek, after telling this story, then makes an extremely strong claim. He says, “It’s not the people. It’s the environment. It’s always the environment. The people are fine. It’s always the environment.” Inspiring! As leaders, we can change the environment and get great results! If we’re clever and patient, we will be heros of our own similar stories!
Its a nice story, but what if, pace Sinek, sometimes it’s the people.2 🤯 And what if we find ourselves in a situation where it’s not obvious whether we need to make some personnel moves. In that case, we’ll need to assess the probability that our environmental tweaks and general efforts will lead to a sufficient improvement in our people’s performance. If we don’t think environmental tweaks will do the job, some folks may be out of a job.
Here’s where the over-representation of inspirational stories becomes problematic: It turns out that we often subconsciously make probabilistic judgments about X occurring based on whether we can easily recall instances of X occurring. Here’s a nice example from Thinking Fast and Slow that shows both the heuristic and the types of mistaken judgments it can lead to:
…I recently came to doubt my long-held impression that adultery is more common among politicians than among physicians 👩⚕️ or lawyers 👩💼. I had even come up with explanations for that “fact,” including the aphrodisiac effect of power and the temptations of life away from home. I eventually realized that the transgressions of politicians are much more likely to be reported than the transgressions of lawyers and doctors. My intuitive impression could be due entirely to journalists’ choices of topics and to my reliance on the availability heuristic.3
If Pfeffer is right that feel-good stories are over-represented because they’re easier to sell and if Kahneman is right that we often mis-estimate the probability of an occurrence because we use an “availability heuristic” when judging that probability, then we’re likely to overestimate the chances that a leader can make a change to her environment that will significantly impact her people’s performance.
We’ll recall all the inspiring, best-selling stories we’ve heard about leaders improving their teams with clever tricks and misjudge the likelihood that we’ll be able to do this same. This will lead to lots of wasted time trying to adjust an environment that won’t actually result in significant improvements in people’s performance. 🤬
Hopefully, I’ve said enough here to convince you that mistakes from feel-good stories can happen and that they can have non-trivial negative consequences on an organization. Don’t get hung up on the debate about people vs. their environment. If you agree with Sinek and think leaders can always dramatically change the way their people behave, great. 👍 Just pick another example. Remember, I’m just interested in what can be known about business in spite of the fact that these types of mistakes are likely, given our imperfect information and imperfect minds.
In any case, with this type of mistaken judgment, the demon metaphor/abstraction is especially apt. We can imagine the demon 😈 tempting us with beliefs that make us feel good. He says to us, “Yes. That’s it. Doesn’t that feel nice? Don’t think critically here. Think of all those inspiring stories you’ve heard. You can be the hero in your version of those stories.”
The Demon’s Causation Mirage
At one point in Thinking Fast and Slow, Kahneman asks us:
A study of the incidence of kidney cancer in the 3,141 counties of the United States reveals a remarkable pattern. The counties in which the incidence of kidney cancer is lowest are mostly rural, sparsely populated, and located in traditionally Republican states in the Midwest, the South, and the West. What do you make of this?4
Go ahead. Think about this question yourself for a second.
If you’re like me, you speculated as to why this pattern occurred. The speculation likely involves some sort of causal story about healthy, stress-free living in rural areas. Kahneman predicts 🔮 this error in his readers and then goes on to explain why it’s an error:
Now consider the counties in which the incidence of kidney cancer is highest. These ailing counties tend to be mostly rural, sparsely populated, and located in traditionally Republican 🐘 states in the Midwest, the South, and the West. Tongue-in-cheek, Wainer and Zwerling [authors of the study Kahneman is discussing] comment: “It is easy to infer that their high cancer rates might be directly due to the poverty of the rural lifestyle—no access to good medical care, a high-fat diet, and too much alcohol 🍻, too much tobacco.” Something is wrong, of course. The rural lifestyle cannot explain both very high and very low incidence of kidney cancer. The key factor is not that the counties were rural or predominantly Republican. It is that rural counties have small populations. And the main lesson to be learned is not about epidemiology, it is about the difficult relationship between our mind and statistics.5
The correct answer to this question, it turns out, is that there is no plausible causal explanation for these two patterns. These patterns occur by mere chance 🎲 because the populations of the counties are small. Extreme outcomes like very high or very low incidence of kidney cancer are much more likely with small samples.6
It’s easy to imagine how this same sort of mistake could play out in published thoughts about business. A business writer (or even an experienced employee) could say, “I’ve seen this solution twice before at these other two companies and it had X positive effect. I’m confident we’ll see the same thing if you implement the solution at your company.” After the kidney cancer example, we know better than to take this argument seriously. The causal story according to which their solution reliably leads to some positive outcome is suspect because 2 is a small sample size and because we know that we’re likely to “see” causation where there is only chance.7
So, to sum up this type of error in our judgment, we’ll say that the demon creates “causal mirages.” Descartes’ demon could make us think there’s “the sky, the air, the earth, colours, shapes, sounds and all external things” when in fact they were “merely dreams 😴…contrived as traps” for judgment. Our demon makes us see causation where there is only chance.
“Knowledge” the demon takes
So what can be known about business given that we’re prone to see causation where this only chance and given that we’re prone to hear feel-good stories over information that actually helps us lead our teams and businesses? By pointing out mistakes in judgment, we’ve already hinted at some answers to this in the last section. So, let’s make these hints more explicit.
The demon definitely gets to take away what we think we know from the main-lessons of anecdote-based theories about how to lead businesses to success. There are plenty of examples of these, but some of the books that I’ve read or heard about recently that fall into this category are
The Science of Growth 🔬, in which Sean Ammirati, a business professor and genuinely impressive (I’ve heard him speak) serial entrepreneur from Carnegie Mellon, tries to derive a theory of what makes companies grow by looking at pairs of similar companies and seeing the difference between the successful and failed companies in those pairs.
Good to Great, in which Jim Collins basically tries to do the same thing Sean Ammirati does. Good to Great is actually highly recommended by John Doer, partner at Kleiner Perkins and early investor in Google.8 9
Here’s another set of books that also fall prey to demon’s powers. These are “fortune-telling” books that attempt to predict how the future will unfold and how that future will affect how business is done:
- Exponential Organizations, about how businesses of the future will operate.
- The Inevitable, a book by Kevin Kelly, the former editor of Wired that says it’ll help us understand “the twelve technological forces that will shape the next thirty years.”
- Thank You for Being Late, ⏱ a book by Thomas Friedman, the NYT writer, promises to help us thrive in “an age of accelerations” and attempts to identify important trends and predict how those trends will play out
Here’s one more set of books that I’ve read that I’ve probably taken too seriously. These books are stories about the success of big companies and individuals. The big worry10 here is that these stories likely ignore the role of blind chance in explaining success and invite us to imagine causal relationships between the company culture and the leadership at these companies and their ultimate success in the marketplace.11
- How Google Works (Google’s story)
- Into the Cloud (Salesforce’s story)
- The Everything Store (Amazon’s story)
- Elon Musk (Tesla and Space X’s story)
I’m not saying that there’s no relationship between the actions of a leader a company’s culture and that company’s ultimate success or failure. I’m just agreeing with Kahneman that because of our tendency to see causation where this is none, the relationship between these things is likely more complicated than what we’re being sold in many business books.
“🙄 you were just being naive”: An Objection
At this point you might be thinking, “Fine, but this is really a trivial point. No one really thinks that the authors are expressing certain knowledge in these books.” Fair, but we do think that these authors have a privileged perspective on these subjects and that’s why we pay attention to them.
This privileged perspective doesn’t appear to survive the demon’s shenanigans, and that shouldn’t surprise us. Garbage in, garbage out holds even if “experts” have more garbage coming in than we do. At one point, Kahneman relays the result of a landmark 20-year study that’s so shocking its worth quoting nearly in-full:
Tetlock interviewed 284 people who made their living “commenting or offering advice on political and economic trends…Tetlock gathered more than 80,000 predictions…Respondents were asked to rate the probabilities of three alternative outcomes in every case: the persistence of the status quo, more of something such as political freedom or economic growth, or less of that thing.
The results were devastating…people who spend their time, and earn their living, studying a particular topic produce poorer predictions than dart-throwing monkeys 🙉 who would have distributed their choices evenly over the options. Even in the region they knew best, experts were not significantly better than nonspecialists…
“We reach the point of diminishing marginal predictive returns for knowledge disconcertingly quickly,” Tetlock writes. “…there is no reason for supposing that contributors to top journals—distinguished political scientists, area study specialists, economists, and so on—are any better than journalists or attentive readers of The New York Times 📰 in reading our emerging situations.”12
There’s little reason to think that folks who comment on emerging business trends or “leadership best practice” would fare any different in a task like this, especially since — as Pfeffer has pointed out — the predictions and recommendations of folks in this field are not often measured against reality.13
My claim here is not that reading these books or consuming similar non-written content was a complete waste of time. Rather, the claim is that given, among other things14, our tendency to imagine causation where there is none and to accept and encounter what feels good rather than what is important, I don’t think we have the foundation needed to make robust and reliable predictions about business trends or to make worthwhile claims about the relationship between quirky company cultures and leaders, on the one hand, and those companies’ ultimate success or failure in their market, on the other.
There’s probably value in studying a lot of the business-related content floating around, but whatever value there is in this study, it is not going to be found by uncritical acceptance of the feel-good, causal stories we’re sold in run-of-the-mill business-related content. That’s not how to study business, and if we fail realize this, the demon wins. 😈
- As someone who’s flirted with content creation as a means of income myself, this point really resonated with me. There have definitely been times where I felt tempted to cover something that I didn’t feel was particularly important simply because that’s what would “sell.” When Google announced that it’d be supporting Kotlin a few years ago at Google I/O, I felt that I should be writing a bunch of articles on Kotlin. The problem is that I didn’t particularly feel that these articles were important or interesting. I thought it’d be more important and useful for me to write about testing, object oriented programming, and architecture, but alas, those topics weren’t “hot” anymore. [return]
- Tom DeMarco and Timothy Lister, the authors of Peopleware actually have a nearly opposite view of Sinek. They say, “For most efforts, success or failure is in the cards from the moment the team is formed and the initial directions set out…managers are unlikely to change their people in any meaningful way. People don’t usually stay put long enough, and the manager just doesn’t have enough leverage to make a difference in their nature.” Tom DeMarco and Timothy Lister, Peopleware, 107-110. [return]
- Daniel Kahneman, Thinking Fast and Slow, under the section called “Origins.” [return]
- Daniel Kahneman, Thinking Fast and Slow, under the section called “The Law of Small Numbers.” [return]
- Ibid. [return]
- Ibid. Kahneman explains this wonderfully by pointing out that it is much more likely that you’ll draw all red or all white marbles from an urn containing half red and half white marbles if you’re drawing 4 marbles rather than 8. [return]
- The reader might object that there’s a difference between the kidney cancer example and an example in which a person is constructing a causal story based on direct experience. However — and this shouldn’t really be a surprise — it turns out that we can make the same kind of mistake even if we’re directly experiencing the phenomenon we wish to explain. Kahneman’s example of Israeli airforce commanders shows this. Because fighter pilots often performed worse after receiving praise and better after being scolded, they believed that negative feedback was more effective than positive feedback. Again, the pattern can be explained by mere chance and regression to the mean. You can find this under the section called “Regression to the Mean.” We also see this same sort of thing play out in the debate between Malcom Gladwell and Steven Levitt over the “broken window theory” of crime. [return]
- Doerr makes the Collins recommendation in his recent book Measure what Matters. [return]
- Built to Last, another one of Collins’ books that takes the same pair-based approach actually gets called out explicitly by Kahneman as a book with a misguided approach. Ibid. under “Recipes for Success” [return]
- Another problem with these narratives is that they likely contain distorted memories. Pfeffer calls attention to this when he writes, “Even in the absence of motivational reasons to misremember and misreport, people invariably recall past events with considerable error…for the leaders who talk or write or blog about their leadership experience, the problem becomes even more pernicious. In telling their stories, leaders create and re-create their own reality so often that soon it becomes almost impossible to distinguish the actual truth from what they recall as being true, even if they wanted to do so.” pg 22. [return]
- Ibid., under “The Illusion of Understanding.” [return]
- Ibid. Under “The Illusions of Pundits” [return]
- Jeffery Pfeffer, Leadership Bs, pg. 13-14. [return]
- I’m thinking of the biases that fall under what Kahneman calls the “what you see is all there is” category and the issues caused by spotty memory I mentioned in note 13 here. [return]