What to Read Next
Not long ago, it became fashionable to embrace failure as a sign of a company’s willingness to take risks. This trend lost favor as executives recognized that what they wanted was learning, not necessarily failure. Every failure can be attributed to a raft of missteps, and many failures do not automatically contribute to future success.
Certainly, if companies want to aggressively pursue learning, they must accept that failures will happen. But the practice of simply setting goals and then being nonchalant if they fail is inadequate.
Instead, companies should focus organizational energy on hypothesis generation and testing. Hypotheses force individuals to articulate in advance why they believe a given course of action will succeed. A failure then exposes an incorrect hypothesis — which can more reliably convert into organizational learning.
What Exactly Is a Hypothesis?
When my son was in second grade, his teacher regularly introduced topics by asking students to state some initial assumptions. For example, she introduced a unit on whales by asking: How big is a blue whale? The students all knew blue whales were big, but how big? Guesses ranged from the size of the classroom to the size of two elephants to the length of all the students in class lined up in a row. Students then set out to measure the classroom and the length of the row they formed, and they looked up the size of an elephant. They compared their results with the measurements of the whale and learned how close their estimates were.
Note that in this example, there is much more going on than just learning the size of a whale. Students were learning to recognize assumptions, make intelligent guesses based on those assumptions, determine how to test the accuracy of their guesses, and then assess the results.
This is the essence of hypothesis generation. A hypothesis emerges from a set of underlying assumptions. It is an articulation of how those assumptions are expected to play out in a given context. In short, a hypothesis is an intelligent, articulated guess that is the basis for taking action and assessing outcomes.
Research Updates From MIT SMR
Get weekly updates on how global companies are managing in a changing world.
Please enter a valid email address
Thank you for signing up
Hypothesis generation in companies becomes powerful if people are forced to articulate and justify their assumptions. It makes the path from hypothesis to expected outcomes clear enough that, should the anticipated outcomes fail to materialize, people will agree that the hypothesis was faulty.
Building a culture of effective hypothesizing can lead to more thoughtful actions and a better understanding of outcomes. Not only will failures be more likely to lead to future successes, but successes will foster future successes.
Why Is Hypothesis Generation Important?
Digital technologies are creating new business opportunities, but as I’ve noted in earlier columns, companies must experiment to learn both what is possible and what customers want. Most companies are relying on empowered, agile teams to conduct these experiments. That’s because teams can rapidly hypothesize, test, and learn.
Hypothesis generation contrasts starkly with more traditional management approaches designed for process optimization. Process optimization involves telling employees both what to do and how to do it. Process optimization is fine for stable business processes that have been standardized for consistency. (Standardized processes can usually be automated, specifically because they are stable.) Increasingly, however, companies need their people to steer efforts that involve uncertainty and change. That’s when organizational learning and hypothesis generation are particularly important.
Shifting to a culture that encourages empowered teams to hypothesize isn’t easy. Established hierarchies have developed managers accustomed to directing employees on how to accomplish their objectives. Those managers invariably rose to power by being the smartest person in the room. Such managers can struggle with the requirements for leading empowered teams. They may recognize the need to hold teams accountable for outcomes rather than specific tasks, but they may not be clear about how to guide team efforts.
Some newer companies have baked this concept into their organizational structure. Leaders at the Swedish digital music service Spotify note that it is essential to provide clear missions to teams. A clear mission sets up a team to articulate measurable goals. Teams can then hypothesize how they can best accomplish those goals. The role of leaders is to quiz teams about their hypotheses and challenge their logic if those hypotheses appear to lack support.
A leader at another company told me that accountability for outcomes starts with hypotheses. If a team cannot articulate what it intends to do and what outcomes it anticipates, it is unlikely that team will deliver on its mission. In short, the success of empowered teams depends upon management shifting from directing employees to guiding their development of hypotheses. This is how leaders hold their teams accountable for outcomes.
Members of empowered teams are not the only people who need to hone their ability to hypothesize. Leaders in companies that want to seize digital opportunities are learning through their experiments which strategies hold real promise for future success. They must, in effect, hypothesize about what will make the company successful in a digital economy. If they take the next step and articulate those hypotheses and establish metrics for assessing the outcomes of their actions, they will facilitate learning about the company’s long-term success. Hypothesis generation can become a critical competency throughout a company.
How Does a Company Become Proficient at Hypothesizing?
Most business leaders have embraced the importance of evidence-based decision-making. But developing a culture of evidence-based decision-making by promoting hypothesis generation is a new challenge.
For one thing, many hypotheses are sloppy. While many people naturally hypothesize and take actions based on their hypotheses, their underlying assumptions may go unexamined. Often, they don’t clearly articulate the premise itself. The better hypotheses are straightforward and succinctly written. They’re pointed about the suppositions they’re based on. And they’re shared, allowing an audience to examine the assumptions (are they accurate?) and the postulate itself (is it an intelligent, articulated guess that is the basis for taking action and assessing outcomes?).
Seven-Eleven Japan offers a case in how do to hypotheses right.
For over 30 years, Seven-Eleven Japan was the most profitable retailer in Japan. It achieved that stature by relying on each store’s salesclerks to decide what items to stock on that store’s shelves. Many of the salesclerks were part-time, but they were each responsible for maximizing turnover for one part of the store’s inventory, and they received detailed reports so they could monitor their own performance.
The language of hypothesis formulation was part of their process. Each week, Seven-Eleven Japan counselors visited the stores and asked salesclerks three questions:
- What did you hypothesize this week? (That is, what did you order?)
- How did you do? (That is, did you sell what you ordered?)
- How will you do better next week? (That is, how will you incorporate the learning?)
By repeatedly asking these questions and checking the data for results, counselors helped people throughout the company hypothesize, test, and learn. The result was consistently strong inventory turnover and profitability.
How can other companies get started on this path? Evidence-based decision-making requires data — good data, as the Seven-Eleven Japan example shows. But rather than get bogged down with the limits of a company’s data, I would argue that companies can start to change their culture by constantly exposing individual hypotheses. Those hypotheses will highlight what data matters most — and the need of teams to test hypotheses will help generate enthusiasm for cleaning up bad data. A sense of accountability for generating and testing hypotheses then fosters a culture of evidence-based decision-making.
The uncertainties and speed of change in the current business environment render traditional management approaches ineffective. To create the agile, evidence-based, learning culture your business needs to succeed in a digital economy, I suggest that instead of asking What is your goal? you make it a habit to ask What is your hypothesis?