Back to Top

Download PDF

By Paul Adler and Charles Heckscher
Work is increasingly a matter of knowledgeable experts cooperating on projects in rapidly changing environments. Our research has attempted to identify the form of organization best equipped to support such work. We have reached two conclusions. First, this kind of work requires a strong sense of community that allows contributors to trust each other. The two other main “tools” of organization, financial incentives and bureaucratic authority and procedures, are useful but ineffectual without a backdrop of community. Second, the kind of community needed today is very different from the traditional community based on loyalty; it takes a new form we call “collaborative.”

Throughout human history people have cooperated with others who were like them and were part of shared long-term communities where personal reputations were well known. For many decades, many of our most effective organizations fostered this kind of community, clothing the skeleton provided by the formal bureaucratic structure in a tissue of strong loyalty. Loyalty similarly added robustness to market relations with key suppliers and customers.

Recent trends have seriously compromised the effectiveness of these arrangements. Within organizations, people are asked constantly to cross boundaries—to work with people they don’t know well and who are very different from themselves. Increasingly, work requires flexible cooperation across functions, divisions, and levels within organizations. Moreover, the network of inter-organizational ties is changing ever more rapidly and broadening to encompass new organizations often based in different national cultures.

Many business organizations in recent years have reacted to these challenges by ignoring or casting aside community. They have tried to meet their performance challenges by restructuring to strengthen bureaucratic controls and by sharpening financial incentives. In the process, they have destroyed trust. These approaches tend to generate fear and competition rather than cooperation and openness. When such organizations attempt to bring different kinds of knowledge and skill together to solve problems, the absence of trust undercuts the knowledge sharing that the work demands.

Our research suggests that successful businesses meet these new challenges not by abandoning community but by reconstructing it along more “collaborative” lines. This new form is more flexible and less insular than loyalty-based community, and trust is established more quickly. This new form is distinguished by its organization, which supports horizontal interdependence rather than relying on top-down control or autonomous self-interest guided by financial incentives; by its values, which emphasize interdependent contribution to a collective purpose rather than loyalty or reliability; and by the social character of its members, which is tolerant of ambiguity and conflict rather than comfortable with fixed roles and status.

IBM Creates a Collaborative Community

* note )

An interesting case in point is IBM’s effort to overcome its bureaucratic past in order to create a flexible services capability. By the late 1970s, it became apparent to many in the company that the long tradition of proprietary mainframe products— “big iron”—was losing ground. Many leaders tried to spur innovation by attacking the traditional IBM culture based on strong expectations of loyalty and conformity to company values (right down to the IBM dress code). The IBM PC was thus developed in an autonomous division set up in Florida, far from headquarters, and kept separate from the rest of the company. The PC was a great product, but the price of the division’s autonomy was that it never overcame the resistance of other divisions and power bases, and as a result IBM lost the opportunity to dominate a new market.

By the early 1990s insiders knew that a crisis was coming. John Akers, the first CEO brought in from outside the company, tried hard to break down traditional loyalties and shake up the company, relying mainly on stronger lines of authority and sharper financial incentives. He restructured, created new divisions, and instituted tough performance standards. These efforts hit a wall of incomprehension and resistance. When their failure became obvious, Akers went into a tirade at a senior management meeting: “Everyone is too damn comfortable. [We have] too many people standing around the water cooler waiting to be told what to do.” He was gone shortly afterwards, and the company nearly went bankrupt within a year.

Like many top managers in troubled companies, Akers put the blame on “the frozen middle” as the principal obstacle to change. This diagnosis hid the inability of top leadership to create a sense of shared community around a new direction, as it does in many other organizations we have studied. In pushing for change, Akers created uncertainty and fear rather than unified commitment. For people to go beyond their narrow jobs and embrace the new challenges and opportunities, they need to have a vision of a positive future, to believe that others will respond positively, and to be convinced that by contributing to a larger whole they will be part of something meaningful. IBM in this first phase destroyed the stifling bonds of traditional loyalties but did little to build a new, more effective community.

The solution for IBM—and, we believe, for many other organizations—was not simply to return to a familiar community, to re-knit broken bonds of loyalty. The old community of twentieth-century corporations was secure and stable, but it was also hierarchical, conformist, and conflict- averse. Trust relied on the assumption—now obsolete—that employees felt they shared a common identity and that their roles were basically stable. The traditional form of community had many strengths but one big flaw: it was insular. The new context requires a new form of trust that fosters open discussion and debate, that not only tolerates but also encourages diversity of views and capabilities.

Collaborative Community: Structure

The new collaborative form of community requires a new structure built around horizontal processes—supplementing, though not replacing, traditional vertical controls. People need to orient their actions not just to what the boss wants and thinks but also to process mechanisms that cross divisional boundaries and are constantly updated and adapted to changing circumstances. We call this structural element of collaboration “interdependent process management.”

IBM has worked hard to create effective processes. Current CEO Sam Palmisano has defined his goal as “an enterprise whose business processes—integrated end-to-end across the company and with key partners, suppliers, and customers—can respond with speed to any customer demand, market opportunity, or external threat.” As a result, we see within the company fewer traditional organization charts of hierarchical boxes and more process maps that define roles and responsibilities in horizontal flows. Whereas the traditional organization chart focuses on upward responsibility and downward authority, these process- focused representations are tools that help people coordinate horizontally with their peers.

IBM has gradually built processes that link parts of the company horizontally toward the customer, crossing from technology development and manufacturing to sales and marketing. Employees have had to change their orientation: instead of looking for a rule to follow and settling into a stable routine taken from their boss or a handbook, they must continually redefine roles and responsibilities through discussion with their project colleagues.

A traditional bureaucratic structure creates clarity by ensuring that each person has a defined realm of authority that matches accountability. The new IBM embodies the collaborative approach in breaking this traditional rule: it is expected that people will take responsibility for things they can’t fully control and that they will move outside the zone of their formal accountability. Moreover, in a traditional bureaucracy, power and influence flow downward. Procedures that specify horizontal processes are typically defined by a centralized staff and imposed on the rest of the organization. The new IBM embodies the collaborative approach: it still relies on standardized, formalized processes, but this centralization is combined with high levels of participation, because the processes are defined and refined over time by input from all levels and units.

Collaborative Community: Values

The second pillar of collaborative community is a set of shared values that give priority to interdependent contribution. A traditional bureaucracy emphasizes doing a good job; it values conscientiousness, reliability, loyalty, and devotion to duty. In a collaborative community, this orientation is no longer enough: people must look beyond their jobs and take larger initiatives. The main question becomes not, “Did you do a good job?” but “Did you contribute to the mission?” We call this value dimension of collaborative community “the ethic of interdependent contribution.”

Lou Gerstner’s first actions when he became CEO in 1993 advanced just such an ethic. As one of our interviewees noted, he immediately took dramatic action to focus everyone on the customer:

“Lou turned the thing way upside down—[he said] the most important thing a CEO does, or any executive in this company does, is meet with customers. That was like a rocket through the company.”

The move was revolutionary because it shifted attention from internal tasks and relationships to “customer solutions” as the common objective that united everyone. Contribution to that shared, external objective could now serve as the common yardstick. IBM has engaged in a major effort to define this concept of contribution to make it real for managers and employees. A significant portion of variable pay is now based on assessments of the individual’s “contribution to IBM,” which attempts to account for hard-to-measure dimensions of performance like teamwork and helping out a colleague.

This is a profound shift in orientation. People focused on contribution are scornful of bureaucrats, and vice versa. The former see the latter as narrow and timid; the latter see the former as undisciplined and overly aggressive. Building a true ethic of contribution requires a great deal of trial and error, during which people learn what kinds of challenges and risks are really constructive and which are destructive; which kinds of conflicts create stronger results and which divide and weaken. Above all, it requires that everyone gain a richly detailed picture of the shared purpose toward which they are all working, a shared understanding of the strategic challenges and opportunities that they are addressing.

Collaborative Community: Character
Some people are just content with just managing their piece of turf and don’t want anyone mucking around in it. [here you need] someone who would be open to criticism or suggestions as opposed to someone who would get defensive, be scared by people meddling around.


Finally, there is the dimension of character. People who are dependent on authority and seek security and clarity are not well adapted to collaborative systems. What is increasingly required is the ability to tolerate ambiguity and manage multiple conflicting commitments. In a unit of Citibank we have studied, collaborative community had become the norm, and one of our interviewees described the difference between his new organization and the older, more bureaucratic Citibank in these terms:

“Some people are just content with just managing their piece of turf and don’t want anyone mucking around in it. [Here you need] someone who would be open to criticism or suggestions as opposed to someone who would get defensive, be scared by people meddling around.”

Nor is collaborative community a hospitable environment for the “cowboy” types who value autonomy above all. What is increasingly required is a more “interdependent” sense of one’s self: a habit of caring and heedfulness. In a unit of a large software services company that had also made considerable headway in forging a collaborative community, one of our interviewees described the contrast between the old world of cowboy hackers and the new one in these terms:

“It’s a bit like streetball versus NBA basketball. Streetball is roughhousing, showing off. You play for yourself rather than the team, and you do it for the love of the game. In professional basketball, you’re part of a team, and you practice a lot together, doing drills and playing practice games. You aren’t doing it just for yourself or even just for your team: there are other people involved—managers, lawyers, agents, advertisers. It’s a business, not just a game. You have to take responsibility for other people—your teammates—and for mentoring other players coming up.”

Collaborative systems need people with interactive character and identities that embrace the complexity of interdependence rather than seeking refuge in either dependence or autonomy. This capacity requires considerable ego-strength and has roots in early socialization and education. We have found that in the shift to collaborative community, many people rise to the challenge and embrace this new self-concept. However, there also seems to be a minority who are incapable of dealing with the inevitable lack of clarity or who are frightened by the questioning of all-powerful authority.

Collaborative Community: The Process of Change

So how has IBM navigated this change from ponderous bureaucracy to relatively nimble, solutions-oriented company? There has certainly been a great deal of internal dissent and conflict, but for the most part they have been successfully overcome because the company has done some crucial things right. Not only have they oriented themselves toward promising, distinctively collaborative types of structure, values, and character, but IBM’s leadership has shown itself committed to a collaborative process for getting from here to there.

All too often, leaders propound a promising model but then pursue that goal with a process that dooms it to failure. For example, the last couple of decades have seen numerous CEOs propound “empowerment” goals, which they tried to implement by autocratic command. By contrast, IBM’s leadership has been courageous in adopting a more collaborative path toward collaborative community.

Sam Palmisano, who took over from Gerstner, felt that IBM’s key problem at that time was a lack of clarity on the core values of the company: the old expectations set by the founders were clearly inappropriate to the fast-moving, customer-centric business model being developed. Palmisano took the extraordinary step of starting a companywide, open conversation—a “values jam,” starting with an intranet-based free-for-all in 2003 and continuing through more focused dialogues in subsequent years. The jam deliberately broke the rules of hierarchical deference: top executives participated as equals in heated debates with lower-level managers, shop floor workers, and clerical staff. At the start, the discussion was dominated by expressions of anger and mistrust at the changes in the company, so strongly voiced that some top managers felt the discussion was getting out of hand and should be cut off. But Palmisano persisted, and the tone soon shifted to a positive collective effort to build a shared orientation to the new environment.

These moves have laid the foundation for a kind of trust that is more flexible than in the old IBM community. People relate successfully not only within the old IBM boundaries but also to the PricewaterhouseCoopers consultants acquired in 2002 and increasingly to a web of allies and contractors. They work together across these boundaries to bring integrated and flexible solutions to their customers. There is now wide agreement that traditional values that gave such prominence to loyalty are untenable, but this has not meant a turn toward individualistic cynicism. Commitment to the mission holds people together and provides the foundation for internal innovation that would have been inconceivable in the past.

The Challenges Ahead

IBM represents just one case that shows a way out of the dilemmas facing many organizations today. But we cannot afford naiveté. In the harsh world of business today, there are many forces that undermine any form of community and that frustrate the emergence of this new, collaborative form. Indeed, the dominant tendencies in industry over the past couple of decades have been toward reliance on command rather than collaboration, and on transient market relations rather than partnership.

Nevertheless, we remain optimistic. The old, cozy world of loyalty-based bonds seems lost for good, but the harsh world of market individualism and hierarchical command cannot build the constructive collaboration among different specialties that is increasingly essential to business. The only way out is to forge a new type of community that encourages people to work together toward a shared sense of purpose. Organizations that master this new form will have a huge advantage in our increasingly knowledge-intensive world.


* The following information on IBM is based in large part on Lynda M. Applegate, Charles Heckscher, Boniface Michael, and Elizabeth L Collins, “IBM: Uniting Vision and Values,” Harvard Business School case # N9-805-116 (2006); see also Lynda M. Applegate, Rob Austin, and Elizabeth Collins, “IBM’s Decade of Transformation: A Vision for On Demand,” Harvard Business School case # N9-805-018 (July 21, 2004). ( back to article )


About the Author

Paul Adler Paul Adler is a Professor of Management and Organization at the University of Southern California Marshall School of Business.


 Charles Heckscher Charles Heckscher is a professor in the School of Management and Labor Relations at Rutgers University and Director of the Center for Workplace Transformation.

About the Author

Share With Your Colleagues