I have three sons who, at least for the moment, all love baseball (as does their father, as evidenced by the frequent use of baseball examples throughout this book). I have the good fortune to help coach each of their baseball teams, although soon their skills and knowledge will surpass mine. Recently my eldest son came to the plate with the bases loaded and one out against a hard-throwing but wild pitcher. Most of the team was either striking out or walking. He ripped a pitch, but unfortunately it went straight to the shortstop, who fielded it on one hop and, given how hard it was hit, easily turned a double play from second base to first base.

My son’s response was not one of grudging acceptance that he had done everything right but gotten unlucky. Rather, it was “Dad, even a weakly hit ground ball would have scored a run.” Of course, no coach would send a player up for an at bat and tell him to mishit the ball in an attempt to get lucky. But after seeing what happened, that is exactly what my son was wishing for. All my sons, when evaluating their performance in a game, react this way. They tend to view how well they hit the ball as a function of whether they got on base (the outcome), not of how hard and where they hit it (more accurate measures of the process).

Unfortunately, their tendency is not uncommon. Most of the time, even though we know that learning requires evaluating the process we used to get to an outcome, we focus on the outcome instead. That’s not the approach Robert Booth takes, however, and we can learn from him how a process focus leads to learning.

A Process Focus in Orthopedic Surgery

When Robert Booth began his career as an orthopedic surgeon, in the 1970s, he looked much the same as others like him.¹ He performed procedures such as hip and knee replacements and arthroscopy and provided nonsurgical care for orthopedic conditions. As time passed, though, Booth began to appreciate that if he wanted to improve the quality of outcomes for his patients, he needed a full understanding of the entire process of care—from initial meetings to surgery and through recovery. He decided to focus first on knee and hip replacements and eventually on total knee replacements only, thus increasing the number of surgeries he performed of that type. As he grew more familiar with that procedure, he was able to identify new areas for improvement.

Booth’s model centered on completing the work quickly in order to achieve the best possible results for his patients. He wrote, “I once heard it said that there are three kinds of surgeons—fast/good, fast/bad, and slow/bad—but there are no slow/good surgeons. Clearly, the ability to operate quickly and efficiently is a priority. At some centers, the average operative time for a primary total knee arthroplasty (TKA) is <30 minutes. If you observe such a procedure, what you should focus on is the process, not the prosthesis… The more efficient we become, the more we study that process, the more skill we develop, and the better results we get [italics added].”² To build an efficient and effective model, Booth created and continually improved a process that managed care as a system.

Booth made a number of changes in comparison with other orthopedic practices. For preoperative care these ranged from the relatively straightforward (for example, he called patients personally the night before a procedure to calm fears and decrease the no-show rate) to the more complex: he arranged for the hospital he was working in to provide his practice with a dedicated admissions space so that his team could not only improve its own workings but also avoid being interrupted by other, less-process-focused doctors. He standardized numerous things, from each step taken during a procedure to using a standing rather than supine X-ray to get the exact view of a knee he wanted.

Booth and his team paid attention to aspects of the process that other surgeons often did not. Each year they would analyze the surgical tool sets they used and remove infrequently used tools. This saved money and time on sterilization and created space in the operating theater. The removed tools were kept in a sterile backup set in the operating room in case they were unexpectedly needed.

Booth even focused attention on the staff members in charge of sterilizing equipment. They were often among the lowest-paid people in the hospital, but he recognized that if they did not do their jobs, he wouldn’t be able to operate. He brought them into the OR to show them the importance of their work and created competitions in which the people who did their work best got tickets to local sporting events. Finally, to coordinate the entire process, Booth held weekly staff meetings that not only addressed the coming week’s logistics but also focused on improvement opportunities.

Booth carried his process focus into the operating room, too. Over time he decided that epidural anesthesia provided the best pain management during and after knee replacements. It took more time to deliver and was more complex, so he worked with both the anesthesiologists and the hospital to secure additional space close by.

To improve efficiency during a procedure, Booth always worked with the same team of surgical nurses; they knew what he wanted when he wanted it. He also used only one prosthesis supplier for his replacement joints. Although that meant that a device might not be perfectly tailored for an individual, differences in device-patient fit were typically quite small. This focus helped him gain additional attention from the supplier and learn the intricacies of the device. He was also able to suggest novel innovations within the prosthetics that resulted in improved quality for patients. Booth did not try to be on the cutting edge of technology. He recognized that learning within the process was most important for delivering efficient and effective care, so it was better to stick with a tried-and-true, improved approach than to jump from one new idea to the next. Finally, after finishing a procedure, Booth gave patients his phone number to comfort them and so that he would hear about any issues that arose.

Booth’s focus on the process served him and his patients well. Over fifteen years he conducted more total knee replacements than any other surgeon in the United States. He was recognized four times by the Knee Society with its research award and served as its president.³

The case of Robert Booth illustrates the difficulty of and the opportunity from taking a process focus to learning. Each part of a system is given careful study in order to build deeper understanding. With practice, the parts improve, but so do the connections between them. In this approach, the focus isn’t on the outcome—although that, too, improves, at least eventually. Process-focused learners recognize that they aren’t fixed in their ability to learn. With effort and study, they can achieve significant change.

Why Does a Process Focus Lead to Learning?

Process-focused learners can be found in many places. Taichi Ohno, the creator of the Toyota Production System, which transformed Toyota Motor Company from a post–World War II afterthought into one of the world’s largest and most consistently profitable automakers, is an example. Ohno knew that the system inside any organization was flawed, not because people didn’t work hard but because perfect understanding of a complex situation rarely exists—and even if it does, the outside environment will change and introduce new challenges. He once said, “Having no problems is the biggest problem of all.” Because all systems have flaws, seeking out those flaws and eliminating them is the only way to learn and improve. Ohno’s approach to production, today known as “lean,”⁴ is now used not only in manufacturing but also in industries from health care to software.⁵ In one of many examples, Ohno, inspired by the cord a rider could pull to stop a trolley, installed “andon” cords on his production lines and instructed workers to pull them whenever anything went wrong in the process. Why? Because it would allow them to learn immediately what had gone amiss and how to fix it.

A process focus is also often a winning strategy in sports. After becoming the general manager and performance director for Great Britain’s Team Sky (professional cycling) in 2010, Dave Brailsford was determined to make everything 1 percent better. He examined all aspects of the process—from obvious choices, such as how riders trained and ate, to less obvious ones, such as effective hand washing to avoid infection, the best pillow to take to hotels for sleeping, and the most effective massage gel.⁶ Brailsford’s focus paid off when his team won not only back-to-back Tours de France—the most prestigious race in cycling—but also 70 percent of the available gold medals at the 2012 Olympics. (Brailsford was the coach for the British Olympic cycling team as well.) As Nick Saban, the head football coach at Alabama, who has multiple national championships to his name, says, “When you have a system, you kind of get in a routine of what’s important… and then you spend a lot more time on thinking of things that would make it better.”⁷

Why is a process focus so central for learning? At its core, learning involves understanding what (and how) inputs affect important outputs—building a model of the way things work. Usually you need to accomplish some task—replace a knee with a prosthetic device, build a car, win a bicycle race—but to accomplish it, you need to understand the many pieces that contribute to the task and how they interact with one another. A process focus provides value on both fronts.

When you take time to learn the process, you recognize that it often involves more inputs than you first imagined. Booth realized that his patients’ outcomes depended on him, of course, but they also depended on many other factors, including nurses, anesthesiologists, central supply workers, administrative workers, the prosthetic device chosen, and so on. The choices of the people involved were also inputs to the process. Focusing on the output rather than the process shrouds the details, and your model of the process will be incomplete.

Even when your view of the inputs is accurate, you still have to discover how they interact to produce an outcome. In some learning scenarios, this process is straightforward. For example, in the game of blackjack, the objective is to get closer to twenty-one than the dealer does without going over. (Each card is worth the number it shows except face cards, which are worth ten, and aces, which are worth one or eleven.) To begin, the dealer gives each player two cards faceup and takes one card faceup and one facedown. Each player then decides whether to “hit” (take another card) or “stay.” A player who goes over twenty-one “busts” and loses. If he stays below twenty-one, the dealer “hits” up to seventeen or more points.

Blackjack can be confusing. Players make a number of choices (and some additional moves include splitting cards or doubling down). But with careful study, it is possible to completely characterize blackjack—that is, to come up with the optimal strategy for every situation in the game. After simulating the possible outcomes again and again with a computer, given what the player has and what the dealer is showing, you can decide whether hitting or staying has the best probability of winning. For example, if the dealer is showing seven and a player has sixteen (say, a six and a jack), the player should hit every time. That risks busting, but the probability of losing otherwise is too high not to hit.⁸

Although blackjack has many moving pieces, it is still possible to simulate and come up with the exact choice to make in any situation, because all the inputs (the cards) are known, as is how the inputs interact (the rules for when a dealer must hit or stay). With this complete information, one can “learn” blackjack.

If only real life were so neat and complete.

When I teach operations to my MBA students, I start with a process focus—something we call process analysis. The same was true when I was an MBA student at Harvard Business School. On the first day of class, Professor Frances Frei began unpacking the wonders that were Benihana—the Japanese-style hibachi restaurant. Admittedly, I was on edge from the day I met her. We had just been given our section assignments and the seats we would take for the entire semester. As I recall, after being introduced, Frei stepped to the middle of the room and said, “Sky deck [the last row in the classroom], I’m coming after you.” She may have said something else first, but because I was sitting in the sky deck, nothing else seemed to matter. I survived that first day in part because we spent the class helping her draw a diagram of every step that Benihana followed, from when customers walked into the restaurant until they left (called a “process-flow diagram”).

By understanding the process, we could see that Benihana had learned enough to create an entirely new type of dining: each step in the process built on the previous one, from the bar that served as a holding pen for batching customers at tables of eight, thus improving capacity utilization, to the order in which food was served—beginning with cheap vegetables and rice and giving very small portions of the expensive meat (although customers remembered being fed an enormous amount)—to eventually cleaning the grill with ammonia so that customers wouldn’t linger and the next group could come in. Frei brought alive for us the fact that a deep process understanding had led to a better and continually improving model.

As learners focus on the process, they can see through the noise that surrounds the valuable signal. UC San Diego professor Roger Bohn studied learning in semiconductor manufacturing plants. He found that in plants that limited the noise through well-run processes, learning was more rapid.⁹

Not only does a process focus help identify relationships but it can reveal causal ones. We frequently hear “Correlation does not imply causation.” It captures the point that although a and b may be related, a doesn’t necessarily cause b. Tyler Vigen, of the Spurious Correlations project, has gone to great lengths to document absurd relationships that are correlational but clearly not causal (see http://www.tylervigen.com/spurious-correlations). For example, per capita cheese consumption in the United States showed a 94.7 percent correlation with the number of people who died by becoming entangled in their bedsheets from 2000 to 2009, while the marriage rate in Kentucky showed a 95.2 percent correlation with the number of people who drowned after falling out of a fishing boat from 1999 to 2010.

Fortunately, neither people who enjoy eating cheese nor Kentuckians getting married need fear for their lives. As one might imagine, these are spurious relationships. If we focus solely on outcomes, we may believe that items are related when in reality, the relationship is due to random variation or even a third factor. By carefully studying the process you wish to learn, you increase your knowledge of the causal relationships.

Professional baseball provides an excellent example of the progression in process-focused thinking. The eventual goal is to win a championship, which requires winning games. Successful teams aim to score more runs (offense) while giving up fewer runs (pitching and defense). Traditionally, baseball managers and scouts looked for hitters who could hit for average (hits divided by official at bats) and power (home runs and runs batted in, or RBIs). When Bill James and other so-called sabermetricians looked at the data, they recognized that the model was incomplete in some places and wrong in others. Yes, hitting for average is a good thing. But a hit is just one way to get on base. If a batter gets four balls, he walks and is awarded first base. A walk is almost the same as a hit, but it isn’t captured in the batting average. Similarly, hitting a home run is a valuable skill, but so is hitting doubles or triples. Finally, knocking batters in to score runs is largely outside a hitter’s control, because unless he hits a home run, the hitter must have someone on base to get an RBI.

Thus the core metrics used to evaluate offensive production were flawed. The sabermetricians realized that it was necessary to look at what factors truly led to runs. By focusing on the process, they could learn more precisely how to value various aspects of hitting and eventually identify where the market had mispriced players.¹⁰

Finally, a process focus helps build discipline in your learning objectives, even when you encounter numerous other demands on your time. John Steinbeck kept a diary while he wrote The Grapes of Wrath; reflecting on the writing process, he said, “In writing, habit seems to be a much stronger force than either willpower or inspiration. Consequently, there must be some little quality of fierceness until the habit pattern of a certain number of words is established. There is no possibility, in me at least, of saying, ‘I’ll do it if I feel like it.’ ”¹¹ A focus on the process—particularly when combined with a specific learning goal—will help you build productive habits for learning.

Challenges That Prevent a Process Focus

Given the importance to learning of a process focus, why do we often devote attention to the outcome instead? First, we incorrectly assume that the outcome provides more meaning than it actually does—that’s called outcome bias. Second, we believe that the outcome is a reflection of our finite ability and thus we judge it as an evaluation of ourselves—so we focus on performance goals rather than learning goals, to our detriment. Let’s consider each of these factors in turn.

Outcome Bias

Randomness is a general challenge in learning. Building a model, mental or otherwise, is important because it helps us distinguish signal from noise. In the absence of such a model, outcomes change how we interpret the events that occurred: if the outcome is positive, we evaluate the process as good, and vice versa. Of course, outcomes do matter. When we try new ideas and they don’t work, we may want to change our perspective on the process that produced those results. Alternatively, it may be that the process was right, but we were unlucky. When we succeed, the process is likely to experience a halo effect. When we fail, we assume that the process was wrong, even if randomness was the underlying cause. As Cade Massey, a Wharton professor and a friend of mine, says, “There’d be a whole lot less drama in the world if people understood variance. Even a little bit.”¹²

Incorrectly interpreting the process because of the outcome is not confined to kids playing baseball. BYU Professors Lars Lefgren, Brennan Platt, and Joseph Price examined how professional basketball coaches adjusted their strategy after close wins as opposed to close losses.¹³ With a close game that finishes in a win or a loss, the coach has similar information: the team was roughly equal to its opponents.¹⁴ But the researchers found that coaches were much more likely after a loss than after a win to change the starting lineup in the next game. This was true even when the team was expected to lose a game or when the loss resulted from factors outside the team’s control.¹⁵

The challenge of outcome bias has been shown in a controlled laboratory environment as well. Rebecca Ratner and Kenneth Herbst presented subjects with a simple question: Would you rather invest a hypothetical $5,000 with a broker that has a 54 percent chance of increasing your investment by 15 percent over the next twelve months or a broker with a 43 percent chance of the same success? (No other information on risk or failure was provided.)¹⁶ Given this limited information, the choice is obvious, but at the same time, there is a significant ex-ante chance that the broker will fail. After the participants had spent twenty minutes on other, unrelated tasks, the researchers randomly varied the investment outcome they were given—either a 15 percent increase or a 15 percent decrease—and then asked each group about the quality of their earlier decision. Those in the group that experienced a negative outcome (15 percent decrease) rated their decisions lower (4.57 out of 7) than those in the group that received positive feedback (6.33 out of 7). The two groups had the same prior information, and nothing meaningful was learned from the outcome; choosing a 54 percent chance versus a 43 percent chance at the same gain is undoubtedly right. Moreover, through multiple studies, the researchers found that the outcome bias was driven by negative emotional reactions. Unfortunately, after an outcome, we struggle to consider what actually happened, considering a good outcome to reflect a good process and a bad outcome to reflect a bad one.¹⁷

Such biased evaluation poses a fundamental challenge to learning. It’s not that we should ignore the outcome, but we focus too much on it. We should learn to weight it appropriately. In fact, in settings that are overly focused on outcomes, the efforts of individuals may decrease.¹⁸ The old saying “You are never as good as they say when you win or as bad as they say when you lose” isn’t just an excuse for losers; it contains an important element of truth. Rather than judging the process through the lens of the outcome, seek to learn about the process that got you there.

Performance Mindset

The second reason we focus on outcomes is rooted in how we view intelligence. Although The Little Engine That Could first appeared in the early 1900s, the engine’s mantra—“I think I can”—connects to research on how we learn. Stanford psychologist Carol Dweck has showed that individuals tend to hold one of two views about intelligence: that it can always be improved, known as a growth mindset, or that individuals are endowed with a certain fixed amount of intelligence—a performance mindset.¹⁹ These views have important implications for a process focus and learning.

Your perspective on intelligence affects how you think about process and outcome. To a performance mindset, outcomes have an evaluative aspect. In other words, success or failure is a direct result of the individual’s intelligence. But to a growth mindset, the outcome is simply one input about the state of the process and the individual’s general learning.²⁰

In a fascinating study on this topic, Claudia Mueller and Carol Dweck examined how praising intelligence rather than effort affected subjects’ views of intelligence and subsequent learning behavior.²¹ In multiple experiments, children received randomly assigned praise for one (“You must be smart at these problems”) or the other (“You must have worked hard at these problems”). Those praised for intelligence subsequently viewed intelligence as more fixed than did those praised for effort; they also subsequently performed worse on various tasks, kept at the tasks for less time, and expressed less enjoyment in doing them.

These findings are relevant to the workplace as well as to learners in the classroom. A study of seven Fortune 1,000 companies showed that a growth mindset was related to happier employees; a more innovative, risk-taking culture; and better performance.²² In ongoing work I’ve done with Dan Cable, Francesca Gino, Julia Lee, and Alison Wood Brooks, we examined consultants’ growth mindset and subsequent work performance. We found a relationship to supervisory-rated performance and learning and also that individuals with a growth mindset were more likely to help coworkers be successful.

The second implication of a growth mindset is that our brains actually respond differently to challenges. Michigan State University professor Jason Moser and his collaborators looked at the cognitive responses of individuals according to whether or not they had a growth mindset.²³ As the participants completed tasks, the researchers measured their brain activity with electrodes attached to their heads. Those with a growth mindset actually showed greater brain activity after incorrect answers than did other subjects. In particular, they showed enhanced Pe amplitude—a brain signal that captures the conscious attention given to mistakes.²⁴ When we’re focused on learning and the process, we don’t have false confidence; rather, our brains are actively trying to learn and improve. It’s as if the Little Engine was given a new power source.²⁵

Successfully Learning from the Process

Fortunately, we can build a process focus in several ways. The first is focus itself. Martha Graham was named the “dancer of the century” by Time magazine. She was both a remarkable dancer and a renowned choreographer who invented numerous techniques. Describing process, Graham said, “Freedom to a dancer means discipline. That is what technique is for—liberation.” Success, even in a novel situation such as dance, depends on understanding the building blocks. Once we understand the basics, we can deviate productively to innovate and learn.

Underlying the process approaches of Booth, Brailsford, Ohno, and Graham is the idea of deliberate practice. None of them attempted to improve the entire process at once. That would have improved nothing. Instead, they focused on small, digestible pieces. Whether you’re attempting to learn how to negotiate a deal, structure a financial transaction, or hit a baseball, starting small around one piece that can be mastered in a reasonable amount of time makes it possible to focus productively on the process.

That means developing a keen eye for where value is created. Building on a process-flow diagram, Taichi Ohno tweaked the model to create a “value-stream diagram.” He recognized that waste—non-value-added steps in a process—was largely socially constructed. If one didn’t focus on the process, one wouldn’t learn what could be removed. Toyota used process-flow diagrams to map inputs and the steps used to create outputs, but it also categorized each step as value-added or waste. A step that didn’t provide value to the customer could be removed. A value-stream focus demonstrates that one cannot or should not fix everything; one should fix those areas with the highest value return.

This approach is applicable for learning in many contexts. For example, I conducted research at Wipro Technologies on its application of a Toyota-inspired approach to developing software.²⁶ I saw numerous teams using value-stream diagrams to understand and improve their processes. One team that was writing code for the software that runs printers realized that it was taking too many steps—literally. The test printers were on a different floor, and team members had to run back and forth—and when they made a mistake, they had to hope no one would change the settings before they returned. Another team recognized that its process for customer feedback included multiple, redundant meetings that could be consolidated. Still another realized that its basic approach to writing code was too linear instead of being iterative, so it wasn’t learning rapidly enough from the mistakes it made.

Often the argument against a process focus is that it turns people into automatons. Interestingly, Frederick Taylor pointed out in his masterpiece, Principles of Scientific Management, that with process a worker can “use his own originality and ingenuity to make real additions to the world’s knowledge, instead of reinventing things which are old.”²⁷ The key is why we do focus on process. If it’s for learning, treating the steps as building blocks that can and will change, then process is a powerful tool that can lead to improvement.

To improve a process focus, you need measures for capturing what is going on. Returning to baseball, one of the most significant shifts in sabermetrics was from discussing wins and losses to discussing the determinants of those outcomes. Early on, runs and other limited measures were considered. But measures have advanced along with technology. In 2015 Major League Baseball finished outfitting every park with the Statcast system—a combination of radar technology and high-definition cameras that permits talent evaluators to capture process metrics such as how hard a ball is hit off a bat (“exit velocity”), where in the stadium each ball goes, and the routes that defenders take to the ball. With this level of data, it becomes possible to judge the success of the process. For example, one team, the Tampa Bay Rays, is said to focus its evaluation of players not on batting average but on exit velocity.²⁸

Companies are attempting to incorporate process measures in their evaluations as well. After collecting data showing that more than half of executives found little value in their performance management techniques (which are typically outcome focused), Deloitte Consulting made its performance reviews more frequent and used more process-focused questions to move away from outcome challenges and improve learning.²⁹ In the same vein, General Electric has been moving away from traditional performance evaluations (including Jack Welch’s famous outcome-driven approach of culling workers at the bottom each year). This change involves focus and new measures, but the process is slow; Janice Semper, an HR executive at GE, says, “Absolutely the challenge is how do you unlearn some things and how do you create new habits. We are trying to build the muscle.”³⁰

Early research focused on growth and fixed mindsets as enduring, individual characteristics. However, additional work revealed that a growth mindset can be taught.³¹ Professors Peter Heslin, Gary Latham, and Don Vandewalle studied how managers’ mindsets shaped the way they worked with their employees.³² In multiple studies the researchers found that managers with a fixed mindset were less likely to see their employees’ talents and more likely to remain anchored in their initial views. With education, their mindsets changed, with resulting performance improvement. People who learn about a growth mindset can identify opportunities to improve themselves and those who work with them. They are increasingly likely to take on challenges and to stick with them—all choices vital to learning.

The factors that push us toward a focus on outcomes are strong, but we must overcome them if we want to learn. To paraphrase Martha Graham, a process focus is the learner’s freedom.