Is Yours a Learning Organization?

by David A. Garvin, Amy C. Edmondson, and Francesca Gino

LEADERS MAY THINK THAT GETTING their organizations to learn is only a matter of articulating a clear vision, giving employees the right incentives, and providing lots of training. This assumption is not merely flawed—it’s risky in the face of intensifying competition, advances in technology, and shifts in customer preferences.

Organizations need to learn more than ever as they confront these mounting forces. Each company must become a learning organization. The concept is not a new one. It flourished in the 1990s, stimulated by Peter M. Senge’s The Fifth Discipline and countless other publications, workshops, and websites. The result was a compelling vision of an organization made up of employees skilled at creating, acquiring, and transferring knowledge. These people could help their firms cultivate tolerance, foster open discussion, and think holistically and systemically. Such learning organizations would be able to adapt to the unpredictable more quickly than their competitors could.

Unpredictability is very much still with us. However, the ideal of the learning organization has not yet been realized. Three factors have impeded progress. First, many of the early discussions about learning organizations were paeans to a better world rather than concrete prescriptions. They overemphasized the forest and paid little attention to the trees. As a result, the associated recommendations proved difficult to implement—managers could not identify the sequence of steps necessary for moving forward. Second, the concept was aimed at CEOs and senior executives rather than at managers of smaller departments and units where critical organizational work is done. Those managers had no way of assessing how their teams’ learning was contributing to the organization as a whole. Third, standards and tools for assessment were lacking. Without these, companies could declare victory prematurely or claim progress without delving into the particulars or comparing themselves accurately with others.

In this article, we address these deficiencies by presenting a comprehensive, concrete survey instrument for assessing learning within an organization. Built from the ground up, our tool measures the learning that occurs in a department, office, project, or division—an organizational unit of any size that has meaningful shared or overlapping work activities. Our instrument enables your company to compare itself against benchmark scores gathered from other firms; to make assessments across areas within the organization (how, for, example, do different groups learn relative to one another?); and to look deeply within individual units. In each case, the power is in the comparisons, not in the absolute scores. You may find that an area your organization thought was a strength is actually less robust than at other organizations. In effect, the tool gives you a broader, more grounded view of how well your company learns and how adeptly it refines its strategies and processes. Each organization, and each unit within it, needs that breadth of perspective to accurately measure its learning against that of its peers.

Building Blocks of the Learning Organization

Organizational research over the past two decades has revealed three broad factors that are essential for organizational learning and adaptability: a supportive learning environment, concrete learning processes and practices, and leadership behavior that provides reinforcement. We refer to these as the building blocks of the learning organization. Each block and its discrete subcomponents, though vital to the whole, are independent and can be measured separately. This degree of granular analysis has not been previously available.

Our tool is structured around the three building blocks and allows companies to measure their learning proficiencies in great detail. As you shall see, organizations do not perform consistently across the three blocks, nor across the various subcategories and subcomponents. That fact suggests that different mechanisms are at work in each building-block area and that improving performance in each is likely to require distinct supporting activities. Companies, and units within them, will need to address their particular strengths and weaknesses to equip themselves for long-term learning. Because all three building blocks are generic enough for managers and firms of all types to assess, our tool permits organizations and units to slice and dice the data in ways that are uniquely useful to them. They can develop profiles of their distinctive approaches to learning and then compare themselves with a benchmark group of respondents. To reveal the value of all these comparisons, let’s look in depth at each of the building blocks of a learning organization.

Building block 1: A supportive learning environment

An environment that supports learning has four distinguishing characteristics.

Psychological safety. To learn, employees cannot fear being belittled or marginalized when they disagree with peers or authority figures, ask naive questions, own up to mistakes, or present a minority viewpoint. Instead, they must be comfortable expressing their thoughts about the work at hand.

Appreciation of differences. Learning occurs when people become aware of opposing ideas. Recognizing the value of competing functional outlooks and alternative worldviews increases energy and motivation, sparks fresh thinking, and prevents lethargy and drift.

Openness to new ideas. Learning is not simply about correcting mistakes and solving problems. It is also about crafting novel approaches. Employees should be encouraged to take risks and explore the untested and unknown.

Time for reflection. All too many managers are judged by the sheer number of hours they work and the tasks they accomplish. When people are too busy or overstressed by deadlines and scheduling pressures, however, their ability to think analytically and creatively is compromised. They become less able to diagnose problems and learn from their experiences. Supportive learning environments allow time for a pause in the action and encourage thoughtful review of the organization’s processes.

To change a culture of blame and silence about errors at Children’s Hospitals and Clinics of Minnesota, COO Julie Morath instituted a new policy of “blameless reporting” that encouraged replacing threatening terms such as “errors” and “investigations” with less emotionally laden terms such as “accidents” and “analysis.” For Morath, the culture of hospitals must be, as she told us, “one of everyone working together to understand safety, identify risks, and report them without fear of blame.” The result was that people started to collaborate throughout the organization to talk about and change behaviors, policies, and systems that put patients at risk. Over time, these learning activities yielded measurable reductions in preventable deaths and illnesses at the institution.

Building block 2: Concrete learning processes and practices

A learning organization is not cultivated effortlessly. It arises from a series of concrete steps and widely distributed activities, not unlike the workings of business processes such as logistics, billing, order fulfillment, and product development. Learning processes involve the generation, collection, interpretation, and dissemination of information. They include experimentation to develop and test new products and services; intelligence gathering to keep track of competitive, customer, and technological trends; disciplined analysis and interpretation to identify and solve problems; and education and training to develop both new and established employees.

For maximum impact, knowledge must be shared in systematic and clearly defined ways. Sharing can take place among individuals, groups, or whole organizations. Knowledge can move laterally or vertically within a firm. The knowledge-sharing process can, for instance, be internally focused, with an eye toward taking corrective action. Right after a project is completed, the process might call for post-audits or reviews that are then shared with others engaged in similar tasks. Alternatively, knowledge sharing can be externally oriented—for instance, it might include regularly scheduled forums with customers or subject-matter experts to gain their perspectives on the company’s activities or challenges. Together, these concrete processes ensure that essential information moves quickly and efficiently into the hands and heads of those who need it.

Perhaps the best known example of this approach is the U.S.Army’s After Action Review (AAR) process, now widely used by many companies, which involves a systematic debriefing after every mission, project, or critical activity. This process is framed by four simple questions: What did we set out to do? What actually happened? Why did it happen? What do we do next time? (Which activities do we sustain, and which do we improve?) In the army, lessons move quickly up and down the chain of command, and laterally through sanctioned websites. Then the results are codified by the Center for Army Lessons Learned, or CALL. Such dissemination and codification of learning is vital for any organization.

Building block 3: Leadership that reinforces learning

Organizational learning is strongly influenced by the behavior of leaders. When leaders actively question and listen to employees—and thereby prompt dialogue and debate—people in the institution feel encouraged to learn. If leaders signal the importance of spending time on problem identification, knowledge transfer, and reflective post-audits, these activities are likely to flourish. When people in power demonstrate through their own behavior a willingness to entertain alternative points of view, employees feel emboldened to offer new ideas and options.

Harvey Golub, former chief executive of American Express, was renowned for his ability to teach employees and managers. He pushed hard for active reasoning and forced managers to think creatively and in unexpected ways. A subordinate observed that he often “came at things from a different angle” to ensure that conventional approaches were not accepted without first being scrutinized. “I am far less interested in people having the right answer than in their thinking about issues the right way,” Golub told us. “What criteria do they use? Why do they think the way they do? What alternatives have they considered? What premises do they have? What rocks are they standing on?” His questions were not designed to yield particular answers, but rather to generate truly open-minded discussion.

The three building blocks of organizational learning reinforce one another and, to some degree, overlap. Just as leadership behaviors help create and sustain supportive learning environments, such environments make it easier for managers and employees to execute concrete learning processes and practices smoothly and efficiently. Continuing the virtuous circle, concrete processes provide opportunities for leaders to behave in ways that foster learning and to cultivate that behavior in others.

Uses for the Organizational Learning Tool

Our diagnostic tool is designed to help you answer two questions about the organizational unit that you lead or in which you work: “To what extent is your unit functioning as a learning organization?” and “What are the relationships among the factors that affect learning in your unit?” People who complete the survey rate how accurately a series of brief, descriptive sentences in each of the three building blocks of learning describe their organization and its learning culture. For the list of statements in the complete survey and details about how it works, see the sidebar “Assess the Depth of Learning in Your Organization.”

There are two primary ways to use the survey. First, an individual can take it to get a quick sense of her work unit or project team. Second, several members of a unit can each complete the survey and average their scores. Either way, the next step is to compare individual or group self-evaluations with overall benchmark scores from our baseline group of organizations. The benchmark data are stratified into quartiles—that is, the bottom 25%, the next 25%, and so on—for each attribute, arrayed around a median (see the sidebar “Benchmark Scores for the Learning Organization Survey”). Once you have calculated your own scores, you can identify the quartile in which your scores fall and reflect on how they match your prior expectations about where you stand.

Having compared individual or unit scores with the benchmarks, it’s possible to identify areas of excellence and opportunities for improvement. If employees in multiple units wish to take the survey, you can also make the comparisons unit-by-unit or companywide. Even if just two people from different parts of a firm compare scores, they can pinpoint cultural differences, commonalities, and things to learn from one another. They may also discover that their unit—or even the company—lags behind in many areas. By pooling individual and unit scores, organizations as a whole can begin to address specific problems.

Holding Up the Mirror at Eutilize

Consider how managers from a major European public utility, which we will call Eutilize, used the survey to assess their company’s readiness for and progress in becoming a learning organization. In the summer of 2006, 19 midlevel managers took the survey. Before learning their scores, participants were asked to estimate where they thought Eutilize would stand in relation to the benchmark results from other firms.

Virtually all the participants predicted average or better scores, in keeping with the company’s espoused goal of using knowledge and best-practice transfers as a source of competitive advantage. But the results did not validate those predictions. To their great surprise, Eutilize’s managers rated themselves below the median baseline scores in almost all categories. For example, out of a possible scaled score of 100, they had 68 on leadership, compared with the median benchmark score of 76. Similarly, they scored 58 on concrete learning processes (versus the median benchmark of 74) and 62 on supportive learning environment (versus the median of 71). These results revealed to the Eutilize managers that integrating systematic learning practices into their organization would take considerable work. However, the poorest-scoring measures, such as experimentation and time for reflection, were common to both Eutilize and the baseline organizations. So Eutilize was not unusual in where it needed to improve, just in how much.

The portrait that emerged was not unexpected for a public utility that had long enjoyed monopolies in a small number of markets and that only recently had established units in other geographic areas. Eutilize’s scores in the bottom quartile on openness to new ideas, experimentation, conflict and debate, and information transfer were evidence that changing the company’s established culture would be a long haul.

Eutilize’s managers also discovered the degree to which their mental models about their own ways of working were inaccurate. For example, they learned that many people in their firm believed that “analysis” was an area of strength for Eutilize, but they interpreted analysis to be merely number crunching. The survey results helped them to understand the term analysis more broadly—to think about the degree to which people test assumptions, engage in productive debate, and seek out dissenting views. Each of those areas was actually a weakness in the firm. This revelation led Eutilize’s managers to understand that without a more open environment buttressed by the right processes and leadership, the company would have difficulty implementing a new strategy it had just adopted.

Eutilize’s experience illustrates how our organizational learning tool prompts reflective discussion among managers about their leadership and organizational practices. Without concrete data, such reflection can become abstract and susceptible to idiosyncratic assessments and often emotional disagreements about the current state of affairs. With the survey data in hand, managers had a starting point for discussion, and participants were able to point to specific behaviors, practices, or events that might explain both high and low scores. The results also helped Eutilize’s managers to identify the areas where their firm needed special attention.

Given that the survey-based scores derive from perceptions, the best use of the data at Eutilize was, as it would be at any company, to initiate conversation and self-reflection, not to be the sole basis for decision making. Discussions had to be conducted with a healthy balance of what scholars call “advocacy and inquiry.” The communication allowed people the latitude to assert their personal observations and preferred suggestions for action, but it also ensured that everyone took the time to carefully consider viewpoints that were not their own. In addition, managers learned the importance of using concrete examples to illustrate interpretations, to refer to specific practices or processes, and to clarify observations. Finally, the participants from Eutilize identified specific actions to be taken. Had they not done so, the discussions could have deteriorated into unproductive complaint sessions.

Moving Forward: Four Principles

Our experiences developing, testing, and using this survey have provided us with several additional insights for managers who seek to cultivate learning organizations.

Leadership alone is insufficient

By modeling desired behaviors—open-minded questioning, thoughtful listening, consideration of multiple options, and acceptance of opposing points of view—leaders are indeed likely to foster greater learning. However, learning-oriented leadership behaviors alone are not enough. The cultural and process dimensions of learning appear to require more explicit, targeted interventions. We studied dozens of organizations in depth when developing our survey questions and then used the instrument with four firms that had diverse sizes, locations, and missions. All four had higher scores in learning leadership than in concrete learning processes or supportive learning environment. Performance often varies from category to category. This suggests that installing formal learning processes and cultivating a supportive learning climate requires steps beyond simply modifying leadership behavior.

Organizations are not monolithic

Managers must be sensitive to differences among departmental processes and behaviors as they strive to build learning organizations. Groups may vary in their focus or learning maturity. Managers need to be especially sensitive to local cultures of learning, which can vary widely across units. For example, an early study of medical errors documented significant differences in rates of reported mistakes among nursing units at the same hospital, reflecting variations in norms and behaviors established by unit managers. In most settings, a one-size-fits-all strategy for building a learning organization is unlikely to be successful.

Comparative performance is the critical scorecard

Simply because an organization scores itself highly in a certain area of learning behavior or processes does not make that area a source of competitive advantage. Surprisingly, most of the organizations we surveyed identified the very same domains as their areas of strength. “Openness to new ideas” and “education and training” almost universally scored higher than other attributes or categories, probably because of their obvious links to organizational improvement and personal development. A high score therefore conveys limited information about performance. The most important scores on critical learning attributes are relative—how your organization compares with competitors or benchmark data.

Learning is multidimensional

All too often, companies’ efforts to improve learning are concentrated in a single area—more time for reflection, perhaps, or greater use of post-audits and after-action reviews. Our analysis suggests, however, that each of the building blocks of a learning organization (environment, processes, and leadership behaviors) is itself multidimensional and that those elements respond to different forces. You can enhance learning in an organization in various ways, depending on which subcomponent you emphasize—for example, when it comes to improving the learning environment, one company might want to focus on psychological safety and another on time for reflection. Managers need to be thoughtful when selecting the levers of change and should think broadly about the available options. Our survey opens up the menu of possibilities.

The goal of our organizational learning tool is to promote dialogue, not critique. All the organizations we studied found that reviewing their survey scores was a chance to look into a mirror. The most productive discussions were those where managers wrestled with the implications of their scores, especially the comparative dimensions (differences by level, subunit, and so forth), instead of simply assessing performance harshly or favorably. These managers sought to understand their organizations’ strengths and weaknesses and to paint an honest picture of their cultures and leadership. Not surprisingly, we believe that the learning organization survey is best used not merely as a report card or bottom-line score but rather as a diagnostic instrument—in other words, as a tool to foster learning.

Originally published in March 2008. Reprint R0803H