Alchemy is a book by the legendary marketer Rory Sutherland. To me, it read like an elaboration of the idea I wrote about in Cost of Rationalism earlier this year (although at that time, I didn’t know about this book). What I categorized at the intersection of “things that we do not fully understand” and “things that are good/useful for us” is what Sutherland termed as “alchemy”. In the book’s definition, alchemy (or magic, used interchangeably) is the stuff unexplained by logic and rationality, but that is nonetheless true. Near the beginning of the book, Sutherland spells out his own take on the cost of rationalism: when you demand logic, you pay a hidden price, you destroy magic. It was very interesting to see the slightly different way of expressing the same idea. But what the book spends most of its time explaining though, is why alchemy is important, why it’s under-utilized, and how to find it.

I attributed the over-reliance on rationalism in the modern world to rationalism’s elevated status as the supreme intellectual way of thinking. Sutherland, I feel, shares the same fundamental belief, but instead of getting philosophical, gives more practical reasons instead. He notes that it’s much harder to get fired when you employ logic, even if you get it wrong. After all, it is much easier to recognize and punish people for being illogical than for being unimaginative. When things go wrong, we fixate on anything that deviated from the norm as a potential cause, much more so than the “standard” choices that were made. Additionally, our natural tendency to post-rationalize can exaggerate the role that logic played in new discoveries, when in fact many of them are merely the result of accidental flukes.

But why is over-relying on rationalism bad, and why do you need alchemy? Sutherland expands on this in great length. Here are my summary of the key points.

• If you only use logic to solve problems, then the problems that are easily solvable by logic will quickly disappear, leaving you with only problems that are tough to solve with logic
• People are fundamentally irrational beings, so logic often yields ineffective, even counterproductive ways to deal with people. There is a saying about why market research is hard: people don’t think what they feel, they don’t say what they think, and they don’t do what they say
• It’s simply too easy to generate rational arguments for anything (e.g. by cherry-picking data, ignoring certain inconvenient facts), especially for smart people. Because of this, it’s easy to be mislead by data and logic. In the end, it’s actually better to be roughly right than precisely wrong
• In a world where most people are using logic to evaluate things, a bit of irrational thinking can yield you some comparative advantage. Looking for things (e.g. hiring candidates) with unorthodox criteria, for example, can get you things that are undervalued by everyone else

So how do you find alchemy? The book mostly illustrates via examples, but here are a few general principles that I have distilled from them.

• Avoid false dichotomies: the opposite of a good idea can be another good idea. For example, both raising and lowering a product’s price could increase sales, for entirely different reasons
• Ask dumb questions with an open mind. People don’t do this because dumb questions already have commonly accepted logical answers. But these answers may be wrong, or there might be other answers, that can only be uncovered by thinking through the questions from scratch
• Recognize the width and concreteness of the context of the problem, and adapt accordingly. Rationality is well suited for solving narrow-context problems, in situations where everything is easily quantifiable, such as the natural sciences. Intuition on the other hand, is much more effective in tackling wider-context, blurrier problems, since that’s what it evolved to do

It’s clear that Sutherland has a big focus on dealing with people’s psychology. He went as far as to coining the term “psycho-logical”, for the normally illogical things that become logical when taking how people’s minds work into account. One should not be surprised about this I suppose, since he is a marketer by trade. It would not be too much of a stretch to summarize the book’s thesis as “rationality is ineffective when dealing with people’s behavior”. But this scope feels a bit limiting. To me, the cost of rationalism goes beyond just things involving human behavior. We pay it to a certain extent in areas of science, mathematics, and technology, even in the realms of pure theory. It is much harder to find concrete examples of this though, so I can understand the book’s narrower focus.

One thing that I do find bothersome is the use of hand-wavy examples not backed by data. Sutherland uses a lot of real life stories to illustrate his points, based on true experiences and famous companies. However some of the examples are merely speculations, and are much less convincing. An especially memorable instance is in a section attempting to demonstrate that the real reason people do things is often not the obvious one, he uses the example of teeth brushing. He claims that if you asked people why they brush their teeth, they’ll probably say to avoid going to the dentist and avoid teeth problems. Yet toothpaste is always mint flavored, revealing the true reason that people do this: for fresh breath. This unsubstantiated example is not very believable, and actually hurt the point that he was trying to make somewhat.

Overall, despite coming from the same fundamental idea around the limits of rationality, Alchemy is not the book that I would’ve written had I chose to expand on the idea myself. Given the emphasis on human psychology and behavior in business contexts, it is clearly a marketing book, not a philosophy one. But it did get the point across, so to speak, in its own narrower domain of focus. A lot of the most interesting bits of the book are actually the examples used to support his arguments, which my generalized summary above does not include. So here at the end, I will list a few of them.

• Asking the dumb question: why do people hate standing on trains? The obvious answer is that people get tired standing, but actually, many people hate standing because it prevents them from being able to use their phones with both hands. Recognizing this, adding more edges for people to lean on (i.e. gently resting their bums) solves this problem without sacrificing as much space as seats do
• In the blind pursuit of efficiency, a hotel replaced all its doormen with automatic doors. But the obvious logical utility of a doorman, i.e. to open doors for guests, is far from the only value. They also help guests hail cabs, carry luggage, and signal the quality of service at the hotel
• It makes no logical sense for Five Guys to give everyone that extra scoop of fries, yet they do it. Intuitively it signals that they rely on repeat customers for business, and thus builds trust