In this final chapter I have three hopes. First, I want to convey my enthusiasm for what the Theory of Jobs can accomplish for innovators, because it answers one of the most important questions that has bedeviled managers for decades: Is innovation inherently a question of luck? Our answer is decidedly “no!” Second, I want to convey the boundaries of the theory—what it can explain, and what it cannot. This is critical. If the theory is applied beyond what the theory was designed to explain, it will lose its clarity and predictability. Third, I want to cultivate your curiosity—by showing you through illustration the depth and breadth of questions we can explore through the lenses of Jobs Theory.
For many who will read this book, the word “theory” connotes a series of equations or a formula that describes how the independent variables or factors affect the outcome of interest. More often than not, the architecture of the research that produces the theory was deductive in character. Research in this genre starts with a core proposition of causality, and then searches for data or phenomena that validates (or invalidates) the core proposition.
Other theories are built through inductive research. Scholars doing this work start without a proposition of causality. Instead, they simply and carefully examine the phenomena and data about the phenomena. Then step by step they develop a proposition about what causes things to occur, and why.
The Theory of Jobs to Be Done was built inductively. Because failure in innovation has been so common, I could not start with a core proposition of causality in successful innovation that I could test in a deductive way. So for two decades I have carefully and inductively observed what people who sold and bought things were trying to do, and tried to get answers to the question, “Why?”
A key goal in building a theory inductively is to develop one or more “constructs.” Constructs are rarely directly observable. Rather, a construct is an abstraction—quite often, a visualization that helps observers see how the phenomena interact with and change each other, over time. Whereas correlations reveal static relationships among the phenomena, a construct is a stepping-stone that helps us to see the dynamics of causality.
In chemistry, for example, Auguste Laurent’s (1807–1853) visualizations (constructs) of chemical compounds enabled him to explain how compounds arise and are transformed into other compounds. In economics, Adam Smith’s construct (1776) of an “invisible hand” helped explain how free markets work. Figuratively, the invisible hand allocates capital and labor to activities that bring prosperity, and it takes resources away from entities that waste them. It has helped billions of people understand how capitalism, properly structured, helps mankind. In the theory of disruption, the key to defining the essence of disruption was being able to visualize trajectories of how technological progress and market needs interact.
Why this diversion into the role of constructs in theories? The term “job” is a construct. It fits exactly the definition of what a construct is, and the role it plays in the Jobs to Be Done Theory. Understanding jobs as a construct involved carefully defining the terms that I needed to communicate what I had been seeing. The terms “hire” and “fire,” for example, are not simply cute words. Rather, they helped me visualize how the processes of buying and selling actually work.
Some who read this book might criticize it because real stories of real people in real companies are not data of the sort that can be manipulated in a spreadsheet. This concern is wrongly applied to the development of good theory. When you see numerical data, remember that it was created by people: individuals or groups of people who decide which elements of the phenomena they include in published data and which they overlook and destroy. Hence, data reflects bias. A wonderful book, Relevance Lost by H. Thomas Johnson and Robert S. Kaplan,1 shows that there is a complicated story behind every number. These stories are hidden when they are parsed and distilled into numbers. When the stories are told, they are rich in data. The insights from the right cases are deep. Numbers that were distilled from stories offer insights that are often shallow but broad.
For these reasons, we feel confident that the “Theory of Jobs” is well named.
A theory never pops out of a researcher’s mind complete and perfect. Rather, it evolves and improves as people use it. Good theories actually need anomalies—things that the theory cannot explain—in order to improve. The discovery of anomalies forces researchers to dive back into the phenomenal muck. They need to improve the theory so that it can account for the anomaly, or define a new boundary beyond which the theory ought not be used. Each time we discover and account for an anomaly, we learn something more about how the world works.
One of the silliest habits of many in academia is to orchestrate a finding that “disproves” a theory that a colleague has developed and published. The authors publish their paper in a journal of repute, and then smugly lie back on a beach somewhere because their paper is now “in the literature.” This helps no one. Anomalies do not disprove anything. Rather, they point to something that the theory cannot yet explain. Scholars who find anomalies need to roll up their shirtsleeves and work to try to improve a theory or replace it with a better one.
I hope that you and the other readers of this book will find things that the Jobs Theory cannot yet explain. If you will communicate these problems to me, it will help me improve our collective understanding. At the time of writing, I am developing ways for these theories to be improved collaboratively online and I welcome your thoughts. My deepest thanks, in advance, for joining the quest to learn more and more about the theory, and helping all of us to understand how to manage innovation more successfully.
Two decades ago, I used the term “disruptive innovation” to describe the phenomena by which entrant companies can topple powerful incumbent companies. The Theory of Disruption has guided tens of thousands of companies toward prosperity. But because, like most words in the English language, “disruption” has many meanings, the Theory of Disruption has also been misapplied. It’s been used to describe many phenomena and situations where it actually does not apply. I’ve puzzled about whether I could have used a better term or descriptor, but have yet to find a better alternative.
For this reason, I would like to try to put boundaries around the word “jobs” as we are using it. It is easy to slip into using “jobs” to describe our attempts to understand a wide range of human motivations. But not everything that motivates us is a Job to Be Done. Jobs, as we’ve defined them here, take work to uncover and understand properly, thus dubbing something a job shouldn’t roll off the tongue with minimal thought. My definition of a job in this book is intentionally precise. I see two problems that you must avoid as you study and apply Jobs Theory.
First, if you or a colleague describes a Job to Be Done in adjectives and adverbs, it is not a valid job. It might describe an experience that a customer needs to have in order to do the job, but it is not a job, as we have defined it here. For example, “convenience” is not a Job to Be Done. It might be an experience that might cause a customer to choose your product rather than a competitor’s product, but it is not a job. A well-defined Job to Be Done is expressed in verbs and nouns—such as, “I need to ‘write’ books verbally, obviating the need to type or edit by hand.” In contrast, the sentence “We should aspire to be more honest” is a noble goal, but it’s not a job.
Second, defining a job at the right level of abstraction is critical to ensuring that the theory is useful. This can be more art than science, but there is a good rule of thumb: if the architecture of the system or product can only be met by products within the same product class, the concept of the Job to Be Done does not apply. If only products in the same class can solve the problem, you’re not uncovering a job.
A couple of examples: “I need to have a chocolate milk shake that is in a twelve-ounce disposable container” is not a job. The possible candidates that I could hire to do this are all in the milk shake product category. I could call this a need or a preference—but it isn’t a job. We need to go up another level of abstraction in order to discover the job. “I need something that will keep me occupied with what’s happening on the road while I drive. And also, I’d like this to fill me up so that I’m not hungry during a 10:00 a.m. meeting. I could hire a banana, doughnuts, bagels, Snickers, or a coffee to do this job.” The candidates to do the job are all from different product categories; and our rule of thumb is that this is the right level of abstraction.
Another illustration: “I need a thin sheet of material that we can wrap around a house just before we apply shingles, siding, or bricks. It needs to have a high coefficient of friction; a low coefficient of thermal conduction; and a high coefficient of toughness—so that it won’t rip as we wrap the house. Oh—and it also must be impervious to moisture.” This isn’t a job, it’s a technical specification. It gives me the choice of buying Tyvek by DuPont, or to be cheap, ignore the spec, and use nothing instead.
I need to go up to a higher level of abstraction in order to discover a job. This is what we might find as we explore for it:
“We’re building a new house here in Boston, where the cold, damp air of winter and the hot, humid air of summer both easily penetrate walls. I want my family to feel warm and cozy in my home in the winter—and cool and dry in the summer. I need to insulate the outside walls of this house so I can minimize the costs of heating and air-conditioning.”
I could hire wood (paper) pulp and blow it into the space in my walls to do this job. I could also hire rolls of fiberglass insulation and staple it to the studs in the wall. Or I could hire Tyvek by Dupont. And to nail things down even tighter, I could hire Tyvek and rolls of fiberglass together. Or I could plan on compensating with extra sweaters in the winter and throwing open more windows in the summer. Maybe I should buy a couple of dehumidifiers and fans. Or maybe I could just hire Santa Barbara or San Francisco, where Mother Nature has obviated the problem of insulation—and I could move there.
We can see that this is a Job to Be Done and not a technical specification or requirement. We know this because the alternatives of things to hire to get the job done come from very different categories of products and services.
The Theory of Jobs has evolved a lot over the last two decades. Without intending to do it, but in the course of trying to help many different people with many different problems, I have been stunned by how broadly and deeply the Theory of Jobs can be used. Almost every day I become aware of an interesting new example of Jobs Theory in action. My daughter Katie recently regaled me with details of Drybar, a salon in which you can get only one service: a perfect “blowout” for your hair—along with the attendant experiences that help you prepare for a special night out and make you feel good about yourself while you’re there. (I had no idea such things were coveted, but I stand duly corrected.) In just a few years, Drybar has become a visible success in cities around the country.
And at the other end of the spectrum, I recently had the pleasure of discussing Jobs Theory with a four-star Air Force general grappling with challenges of motivating and retaining top personnel in an era of government budget constraints. When he left my office, it was as if he saw his dilemma in an entirely new light—one that provided some hope. “Never in my wildest imagination would I have guessed that a story about milk shakes would change the way I thought about recruiting in the military,” he declared as we parted company. It’s a challenging problem he faces, but I hope Jobs Theory offers him the perspective to make a difference. These are just a few examples that come quickly to mind from recent days. But I’ve been thinking about using insights from Jobs Theory for some of the bigger problems in our families and our society for a long time, such as in our personal lives, education, and health care.
In 2012 Karen Dillon and I joined with James Allworth, one of my most thought-provoking former students, to write How Will You Measure Your Life?2 In its sixth chapter, we put on the Jobs to Be Done Theory, like a set of lenses, and looked at what goes on in our personal lives. We see things that have been hiding in plain sight in our personal lives and families for many years. We pose questions such as, “What is the job that our children hire parents to do for them?” and “What is the job(s) that my wife needs to get done, for which she might hire a husband?” These questions are posed at the right level of abstraction. For example, when something breaks in the house a wife might hire her husband to do the job. She could also hire a tradesman to repair it. She might simply do it herself. Or she might just live with it, never fixing the problem. Another job is that she needs to feel loved. She could hire a husband to get this job done. But all too often the husband doesn’t do the job very well. So she could hire friends and family to do the job, or her profession to do it. Or she could live her life without ever getting this job done well. We hope you will read this and think about the jobs you are being hired for in your life—and if you are performing them well. It might be a sobering exercise.
In 2010 Michael Horn, one of my brightest former students and now a leading voice in the national discussion of the future of education, and I published Disrupting Class3—an inquiry into why our public schools struggle to improve. Improving schools is a very complicated problem, of course. As we mentioned earlier in this book, one of the most important insights we conveyed in Disrupting Class came when we put on the Theory of Jobs lenses and explored what the job is that students are trying to do. We concluded that school is not a job that children are trying to do. School is one of the things that children might hire to do the job. But the job is that children need to feel successful—every day. And they need friends—every day. Sure, I could hire school to do these jobs. But I could drop out of school and hire a gang to feel successful and have friends. Or I could drop out of school, get a minimum wage job to earn some money, and buy a car—and cruise around the neighborhood with my friends.
Most schools don’t do this job well at all. Instead, most children feel failure when they go to class. They could also hire athletics to do the job. For a few, sports do the job well. But for the less gifted, athletics makes students feel failure, too. So they hire electronic games to feel successful. And yet for many, even such games yield failure. So they hire friends who have feelings of failure, too—and engage in drugs and other things to feel successful.
I was heartened to learn recently that Corning CEO Wendell Weeks and his wife, Kim Frock, have set up an alternative school, the Alternative School for Math & Science, in Corning, New York, with the explicit goal of helping children feel successful at school. That’s what the Khan Academy is focusing on, too. It gives me enormous hope to know that great people are working on getting the job of students right. We’ve learned that these important Jobs to Be Done in our children’s lives have been hiding in plain sight.4
In 2009 I teamed with another of my terrific former students, Jason Hwang (now cofounder and chief medical officer of Icebreaker Health), to write The Innovator’s Prescription5—a book to explore why the cost of our health care system increases at an unsustainable rate, even as accessibility declines. Again, a key for unlocking this dilemma has been the Jobs to Be Done Theory. For example, the job of most people is that they want to be so healthy that they don’t even have to think about health. Yet, in systems where the providers of care are reimbursed for services they provide, they actually make money when the members of their system get sick—it’s effectively “sick care” rather than “health care.” When the members are healthy, the providers make little. In other words, the Jobs to Be Done of members and providers are not aligned in the US health care system.
At some health care providers, such as Intermountain Healthcare, Kaiser Permanente, and Geisinger Health System, managers are aggressively working to align the jobs of providers and the consumers in their care. One of the most important ways they do this is by assuming responsibility for the cost of care—by insuring consumers, for instance. The organizations’ financial sustainability therefore depends on keeping consumers as healthy as possible over the course of their relationship with providers; and this dependency enables jobs-focused innovation around disease prevention, and care efficiency and effectiveness, to flourish. It enables providers to focus on keeping consumers healthy, instead of waiting until they get sick to step in; and on helping them get well as soon as possible, or effectively managing chronic conditions, when they do get sick. The result? The jobs of the provider and consumer are aligned well.
These plans are in stark contrast to traditional plans, in which providers are paid only for specific services delivered to consumers. In this context, providers have no financial incentive to keep consumers healthy over the long term or to manage the cost of care delivery—and they have every incentive to increase the volume of care delivered. Their jobs and consumers’ jobs are painfully misaligned.
What job do we elect a political leader to do for us and our nation? What job do they think we are hiring them to do and how does that compare with the job we think we’re hiring them to do when we go into the voting booth? Are they aligned? Are we hiring people to lead us? Or to give voice to our fears? They’re not the same thing. As I mentioned in chapter 8, Peter Drucker famously cautioned us: “The customer rarely buys what the company thinks it is selling him.” I suspect there’s a profound disconnect between voters and politicians, too, and that’s why we’re continually dissatisfied with the people we elect to serve us.
Think about this disconnect the next time you go to church, synagogue, or another house of worship—or deliberately don’t go. The Jobs to Be Done Theory explains why so many churches are struggling to keep their members. They have lost a sense for the jobs that arise in their members’ lives, for which they might hire a church.
I could go on for hours about how the Theory of Jobs helps us see the world in unique and insightful ways. Good theories are not meant to teach us what to think. Rather, they teach us how to think. I encourage you to continue the conversation from here in your home or your office after you put this book down.
A few years ago, in the middle of the class I teach at Harvard—“Building and Sustaining a Successful Enterprise” (BSSE)—one of my students raised her hand to ask a question. We were about halfway through the semester and, as usual, we had spent our time learning various theories that I believe are the most important tools I can give my students before they venture out in the world. I’ve gotten a lot of questions over the years, and I’m usually prepared for anything. But this one threw me a bit. “Excuse me, professor. I don’t mean to be rude, but I wanted to know what’s the purpose of this course?” I was surprised because I thought it was clear that we were preparing them to accomplish great things in their careers and personal lives and to navigate the difficult decisions that would inevitably come their way. But I asked her if I could think about it overnight. The next day I had an answer that not only satisfied her, but satisfied me: “In this class we learn theories that explain what causes what to happen. Isn’t it great to know how things work?”
That has been the aim of this book, too. If you know how innovation works—what truly causes innovation to succeed—your efforts don’t have to be left to fate. We’ve allowed ourselves to believe that luck is essential for far too long. There are whole industries, such as venture capital, that are currently organized around the belief that innovation is essentially a game of playing the odds. But it’s time to topple that tired paradigm. I’ve spent twenty years gathering evidence so that you can put your time, energy, and resources into creating products and services that you can predict, in advance, customers will be eager to hire. Leave relying on luck to the other guys.
1. Johnson, H. Thomas, and Robert S. Kaplan. Relevance Lost: The Rise and Fall of Management Accounting. Boston: Harvard Business School Press, 1987.
2. Christensen, Clayton M., James Allworth, and Karen Dillon. How Will You Measure Your Life? New York: HarperCollins, 2012.
3. Christensen, Clayton M., Michael B. Horn, and Curtis W. Johnson. Disrupting Class: How Disruptive Innovation Will Change the Way the World Learns. New York: McGraw-Hill, 2008.
4. We know of at least five authors who have published books named Hidden in Plain Sight. They are Jan Chipchase and Simon Steinhardt, Erich Joachimsthaler, Andrew H. Thomas, and Peter J. Wallison. We thank them, and undoubtedly others, with this marvelous phrase that we shamelessly borrow in this chapter.
5. Christensen, Clayton M., Jerome H. Grossman, and Jason Hwang. The Innovator’s Prescription: A Disruptive Solution for Health Care. New York: McGraw-Hill Education, 2009.