“The trouble with market research is that people don’t think what they feel, they don’t say what they think, and they don’t do what they say.”

— David Ogilvy

Beautiful equations are seductive. Mathematics brings logic and structure to a complicated world and allows us to make sense of things that once seemed inexplicable. In hard sciences such as physics, mathematics represents verifiable truths in the physical world such as the rules of planetary motion. But we need to know the limits of mathematics to avoid being seduced into false precision and delusional conclusions in areas where equations fail to model reality.

Given the success of mathematics as a tool to model many aspects of the physical world, it is not surprising that academics in other fields would attempt to bring mathematical rigor and credibility to their areas of expertise. However, when it comes to social sciences such as economics, we enter into the realm of human behavior which has yet to be reduced to a set of beautiful equations. Human decision making is amazingly complicated, messy, unpredictable, and often seems highly irrational. It is simply not possible to represent human decision making as precisely as we can represent planetary motion or nuclear reactions, but that has not stopped economists from trying.

Rory Sutherland’s book, Alchemy, is a highly entertaining look at the pitfalls that await those who seek to reduce human behavior to a set of equations. Sutherland’s background as Vice Chairman of Ogilvy has put him in a great position to observe the seemingly irrational aspects of consumer behavior. However, his observations extend well beyond tips for marketers and touches on important aspects of human psychology and decision making. As Nassim Nicholas Taleb has pointed out, any book worth reading cannot be “summarized”, and that holds true in the case of Alchemy. The book reads like a series of short essays arranged into various themes, but all center on Sutherland’s opening quip that “the human mind does not run on logic any more than a horse runs on petrol.” However, it would be a mistake to think that this means there is no rhyme or reason behind human behavior. We need to focus on what people are really trying to accomplish which often is very different from what they appear to be trying to accomplish.

Before discussing a few of Sutherland’s ideas in more detail, it is worth taking a brief look at his “rules of alchemy” presented at the start of the book:

Reading through these rules provides a flavor of Sutherland’s overall philosophy of human decision making which he refers to as psycho-logic. Far from appearing “rational” or “logical”, Sutherland’s notion of psycho-logical decision making has evolved over time to be useful rather than optimal. Psycho-logical decision making is roughly equivalent to the concept of using heuristics, or coming up with problem-solving approaches that are “good enough” for our purposes, even if they do not always yield the best results. As Sutherland points out, “there are two separate forms of scientific inquiry: the discovery of what works and the explanation and understanding of why it works.” The use of aspirin to reduce pain was known for decades before scientists found out how it actually works. We should be open to utilizing things that we know work even if the rationale behind it remains a bit murky.

Don’t Listen to Your Customers

Sutherland makes a great point early in the book regarding free markets: In contrast to planned economies, free markets are able to generate innovative things even when their popularity does not make any rational sense at the time. By defying conventional wisdom and perceived logic, free markets can accomplish advances that central planning cannot.

Sutherland takes this a step further:

“For a business to be truly customer-focused, it needs to ignore what people say. Instead it needs to concentrate on what people feel.”

Alchemy, p. 45

I am sure that when many people read this quote they will immediately think about Steve Jobs, the legendary founder of Apple. Jobs was famous for not seeking input from customers regarding product design. Jobs observed that “people don’t know what they want until you show it to them.” Readers of Walter Isaacson’s biography of Jobs know that Jobs constantly strove to understand the psychology of customers and had a unique talent allowing him to combine aesthetics with functionality in a way that met needs that customers never knew they had.

In a planned economy (or, indeed, at companies like Nokia or Research In Motion), a market researcher in charge of consumer technology in 2005 might have interviewed users of cell phones to find out what features they would like to see in the future. Most people would have wanted to “do email” on their phones. So the central planner might have responded by making physical keyboards more capable. Jobs was able to envision what consumers really wanted which was a device capable of an unlimited number of functions, one of which was email. By introducing a virtual keyboard for the first iPhone, Jobs created a product capable of morphing itself into a device with capabilities limited only by human imagination.

The point here is not that customer feedback is never important and should be ignored in all cases. Clearly, there are cases where a business would want such feedback in order to make incremental changes to a product or service. However, major breakthroughs are unlikely to be made by seeking responses to obvious questions. Thinking about the underlying reasons for human decision making, which are very often subconscious, is a necessity for those who aspire to achieve real advances.

Perception is Reality

What would create a greater increase in customer satisfaction: Reducing the average speed of a subway ride between suburban Virginia and downtown Washington D.C. by ten minutes or installing electronic signs at each station showing commuters when the next train will arrive?

It is likely that a ten minute reduction in a thirty minute train ride would increase satisfaction to some degree, but the “new normal” would soon set in and familiar gripes would resurface. The problem is that implementing a ten minute reduction would be incredibly expensive. The Washington metro rail system is near capacity already and the two-track system cannot facilitate express trains. For years, long range planners have considered adding an additional tunnel under the Potomac River to alleviate congestion. Clearly, this type of improvement is not possible for the foreseeable future.

On the other hand, installing electronic signs at all stations is a more modest capital outlay and was implemented for the Washington metro rail system around the turn of the century. These signs increased the satisfaction of commuters but saved them no time at all. What the signs did was reduce the sense of uncertainty regarding when the next train would arrive.

Sutherland points to similar innovations that have massively improved the perception of customers without actually improving the service itself. The Uber app could have been designed to simply allow users to summon a car and let them know when the car arrives. However, the inclusion of a map showing where the car is located dramatically improves the perception of the service because customers are able to monitor how far away the car is while waiting. Your car will not arrive a second faster, but you will be much less frustrated with the overall experience.

Sutherland refers to these types of improvements as “psychological moonshots”. An actual moonshot is a very ambitious innovation that is typically not only expensive but risky. Psychological moonshots are a way to “achieve massive improvements in perception at a fraction of the cost of equivalent improvements in reality.” This might sound nefarious — are we trying to trick people? Sutherland would likely respond by asking what difference it makes if we are? If you can achieve an increase in human happiness with psycho-logic, why not pick the low hanging fruit?

Signaling Must Be Expensive

Sutherland’s observations regarding signaling behavior closely tie into the necessity of trust in a well-functioning society. Building a reputation requires years of efforts and can be lost very quickly which implies an asymmetry well captured by Sutherland’s reference to a Caribbean proverb: “Trust grows at the speed of a coconut tree and falls at the speed of a coconut.”

The need to signal seriousness and intent explains a number of actions that are taken by marketers that, on a surface level, makes little sense. Let’s say that you have some information to communicate to clients or potential clients. To an accountant, the rational method of communication would be one that minimizes cost such as using bulk email. However, a bulk email lacks the gravitas associated with a letter on quality paper stock delivered by Fed Ex.

Sutherland recounts his experience working on a marketing campaign where the key to success involved sending a solicitation to try a product via Fed Ex rather than email. He also points out that there is a reason wedding invitations are not sent out via email but on heavy paper or card stock.

Branding Facilitates Satisficing

At a surface level, the willingness to pay a premium price for branded goods strikes many economists as irrational consumer behavior but this attitude stems from not fully understanding what consumers are trying to do:

“Economists tend to dislike the idea of branding and are inclined to see it as an inefficiency, but then they might view a flower as an inefficient form of weed. The reason they might not understand the flower’s extravagance in squandering its resources on producing scent and color is that they don’t fully understand what it is trying to do or the decision-making and information-transfer context in which it is trying to do it. It is not any more irrational for human consumers to pay a premium for heavily advertised products than it is for bees to prefer to visit heavily ‘advertised’ flowers.”

Alchemy, p. 193

Sutherland’s thesis on advertising is that companies are investing heavily in signaling behavior which adds credibility that the product will at least not be bad. After all, heavily advertising a bad product would be self-defeating since it would only spread the word more quickly especially at a time when online reviews can make or break a product.

There are many scenarios where it makes rational sense to seek the product or service that is “good enough” because the risk of picking something unknown and bad is far too high. This is why one might see American tourists flocking to a Starbucks or McDonalds location while traveling abroad. They know exactly what they are going to receive and are probably under no illusions regarding the idea that they are choosing the best possible option. But they know it will be good enough. Those of us who have taken long distance road trips can relate to this as well. Choosing to eat breakfast at IHOP rather than at a no-name diner is likely to be satisfactory whereas the unknown greasy spoon might result in digestive problems on the road.

The power of branding appears to have diminished to some degree in recent years for a number of reasons. Store brands have gained market share as retail establishments themselves become the “brand”. Consumers have come to trust that store brands such as the Whole Foods “365” line and Trader Joe’s branded products have quality very similar to their branded equivalents. Online reviews available on smartphone apps make it relatively easy to pick local restaurants while avoiding the chains. Nevertheless, Sutherland’s views on branding and “satisficing” will continue to justify advertising spending.

Square Pegs and Round Holes

Sutherland urges readers to “never call a behavior irrational until you really know what the person is trying to do.” This is excellent advice given the fact that human behavior simply cannot be studied in the same way as the physical properties of the natural world. Despite great advances in psychology in recent years, we do not fully understand how our minds work. But even if we do not fully understand the decisions people make, we need to recognize that the seemingly irrational could make perfect sense in light of the fact that people are using heuristics to make “good enough” choices under conditions of uncertainty.

Stephon Alexander, a professor of physics, saxophonist, and author of The Jazz of Physics makes the following observations regarding the rational beauty of mathematics:

“An elegant equation is refined, slimmed down to the bare essentials, simple and concise. An elegant equation is tastefully written in the mathematical language of numbers, letters, and symbols. An elegant equation is superior in its ability to house within it other equations that can be derived from it. An elegant equation is a beautiful thing.”

The Jazz of Physics, p. 55

Social scientists, especially in the field of economics, have longed for the ability to reduce human behavior to the types of beautiful equations that help us describe the cosmos, but by insisting on applying pseudo-scientific “rigor” inappropriately, they risk muddying the waters rather than bringing about enlightenment.

If you are looking for hard-and-fast rules governing human behavior, you will not find any in Alchemy or anywhere else, but of course that is precisely Sutherland’s point. Human behavior is often shrouded in mystery and we benefit from being open to considering counterintuitive and seemingly illogical ideas that, for some reason, appear to actually work.