Book Review of Tim Harford's "Adapt"
If you want to succeed, you need to get some failures behind you.
That’s the overall paraphrased message of Tim Harford’s book “Adapt.” This message is not new to some of us, but it’s important to hear again and again because so many of us who believe this principle fall back much too quickly into the perfectionist trap. We theoretically know we need to try, fail, and fail again before we ultimately succeed, but aren’t as willing in practice to walk through the inevitable gauntlet of mistakes, errors, fumbles, falls, corrections, rejections, critiques, loses, and disappointments before rising victoriously with a new skill, a better job, a higher quality of life, or a successful relationship.
We also seem to rarely see, or at least notice, the struggles of others and thus are biased into thinking that the only way to success is jumping straight to perfection without a morsel of failure. After all, the observable evidence in the public domain seems to support this. Successes are magnified on social media where we often post only the end state, the final product, the travel picture (after months of saving), and so on, while the failures, work, and efforts are diminished, brushed aside, or not shared at all, on our social networks. Many of us, myself included, believe we don’t look as impressive if anyone else sees us in some state before success has been reached. Perhaps we’re right, but that doesn’t change the fact that every single person must be in, and pass through, a state clearly defined as short of success. Every great musician and author you and I listen to and read, respectively, must have had a flawed melody or sentence many times before we finally consume their finished product.
Like many others, my favorite sculptor is Michelangelo. He was known to destroy many of his earlier iterations or preparatory sketches and drawings of major projects. He was, perhaps, worried about others stealing his ideas - what we call intellectual property today - but I like the explanation that he wanted his masterpieces to be viewed as miracles, emanating magically from his genius mind without effort. This makes a set of unfinished sculptures, called the “Prisoners” or “Slaves,” in the Galleria dell’Accademia all the more intriguing,. These works reveal genius - but genius in process. They are amazing in their own right but add to the crescendo of experience at the end of the Tribune where one meets the David, a statue considered one of the greatest art pieces of all time.
I love these “Prisoners” because it suggests the obstacles and real-world challenges that Michelangelo and all of us face on the path to creating our own Davids. These challenges include: Changing requests by a customer, friend, or boss; Budget cuts, inflation, or downsizing; Distractions, boredom, and even competition (Raphael and Michelangelo were competitors in the same field), to name just a few.
Michelangelo faced some or all of these in various ways. Although we live in a different millennium, we aren’t any different and face similar challenges on the road to success. However, he kept going, made adjustments, learned, and adapted.
Tim Harford’s book “Adapt” brings these same ideas to the 21st century with poignant and relevant examples to hopefully motivate us beyond the first, second or even fiftieth failure that will surely come our way before we see success in our endeavors. Box.
He spends considerable time with the metaphor of natural selection in evolutionary biology. In particular, he discusses both variation and selection as crucial steps for us to reach success in our own lives, vocations, or projects.
In the chapter on Variation, Harford describes the need for a plurality of ideas in almost every domain. Since most ideas will be bad or insufficient in some way, the development and exploration of these ideas should be as cheap and quick as possible. Unlike the relatively slower process of crossing or mixing DNA material in organisms to generate potentially different offspring, a number of strategies and incentives have been realized to accelerate and expand the set of possible solutions considered in other domains. The internet provides a platform where companies, with enough users, can try hundreds or even thousands of small tweaks to websites, programs, and other interfaces for next to nothing in cost with answers returning in days, hours, or even minutes. Harford agrees with Clay Shirky suggesting that this technique amounts to “failure for free.” There is little to lose and so much more to gain from these lessons of failures, which in turn inch the organization that much closer to success.
For complex problems, larger incentives are needed for people to generate potential solutions but can still be accelerated and cost effective. Cash prizes are one of the most obvious strategies to motivate variation within interested parties. The Netflix Prize, Innocentive, and the X Prize Foundation, have awarded substantial sums of money to individuals or groups who solve real-world, complex problems. With a larger number of people competing for a significant sum of money, the number of potential solutions increases along with the probability that one of those will succeed. The cash prizes might be large but the impact and possible return on investment of these cash prizes are even larger.
Harford continues in the chapter on Selection, that variation is necessary but insufficient for success. One still needs to select from among the possible concepts, which can be just as challenging. Why? Because evaluating a concept might take weeks or even years before one will know if it was the best option, or, on the other hand, they should select something else. Comparatively, if evaluating and then selecting an ideal concept for a website can be done in days or hours, selecting the best curriculum and structure for teaching children in a developing country could take years before the idea can be fully evaluated. To help with this selection process, RCTs or random control trials can help ascertain if an idea is better, but this has issues as well. For example, if an experimental drug for cancer is given to one group of patients and a placebo to another group, and the drug is showing clear signs of efficacy in treatment before the experiment is complete, is it ethical to deny the placebo group access to the drug so the true performance of the drug can be quantified?
Regardless of the ongoing discussion and arguments along these lines, uncontrolled experiments are clearly not desirable at any time. If little or no feedback can be gleaned from an experiment, then the only real effect is lost time and effort. Carefully controlled experiments should be the default in any selection process, from the distribution strategies of foreign aid to domestic medical insurance programs to the types of features on a consumer product. All experiments should answer important questions, increase knowledge, and help in the selection process of the best option. Thus, before implementation of an experiment of any type, one should set it up so that after completion, they can avoid the sentiment of: “the results are inconclusive.” Harford makes a plea, therefore, that feedback loops about the performance of a test need to be both shorter and better, in order to answer the question of “what works?” more effectively.
In the final case study chapters, Harford applies his model to the two areas of climate change and the fragility of financial systems. While he succeeds in the former, he loses his pacing in the latter getting bogged down in the description of the operation of banks and the definition of financial instruments that are covered in more detail (and perhaps better) in many other books since the credit crunch of 2008. Despite this one weakness or “failure,” the book is still highly recommended.
With everything taken into consideration, Adapt was a great read. The chapters are broken into useful subsections with lengthy headings or quotes making the book an accessible resource to skim and track down a particular anecdote without too much effort. The notes at the end are abundant and provide additional starting points for the interested reader. For anyone interested in how to fail better, fail faster, and fail successfully, Adapt is a great place to start. Four out of five stars.
