Here is the paradox: many companies know they have great strategies and great people, but their performance doesn’t meet the aspirations of their top managers or shareholders. The cause of this disappointment may even seem obvious. If the CEO assigns a small task force of senior managers or highfliers to find the answer, it will quickly pinpoint the immediate culprits—confused accountability within the leadership group, perhaps, or an unforeseen shift in the balance of power between business units and the corporate center.

Once the task force has identified the problems, the CEO may be tempted to fix them straightaway—for example, by spelling out the accountability of different executives or rebalancing the rights and responsibilities of the business units and the corporate center. But performance-sapping organizational problems like these may have complicated roots linking one problem to another. Unclear accountability at the top may be symptomatic of a serious power vacuum within the leadership group, and this may be spurring business units to become more autonomous. Trying to fix what is actually an effect of the problem rather than the cause won’t solve anything.

The kind of high-level, "outside-in" diagnosis that reveals the problems undermining performance is unlikely to afford insights into their root causes, how they are linked to one another, or how to sort them out. Getting more people to step inside the company’s workings and uncover the complex roots of underperformance can be the first step toward developing lasting solutions, as the case of ManufactCo (not its real name) illustrates.

Collective discovery

ManufactCo, a manufacturing organization with 8,000 employees, produces and markets its products in more than 20 countries. In the 1980s and ’90s, it went through acquisitions that brought in a heritage of varied business cultures and ways of getting things done. Three or four years before the period described here, ManufactCo’s return on capital employed (ROCE) had been in the industry’s top decile. Yet over those years, its performance gradually dipped into the second quartile—good but not leading edge. The organization wasn’t doing as well as almost anyone in it thought it could.

The sense of theater the quarterly reviews created gave managers a strong incentive to come prepared

Leaders at ManufactCo weren’t sure why it had lost its performance edge. Its products were strong. It had a good reputation and high-quality people. Managers and employees placed an extraordinary emphasis on delivering high performance. At the beginning of every year, the center and the business units negotiated stretch targets for each unit. Every month the units submitted to the center detailed financial results, with key performance indicators and updated annual forecasts. The CEO publicly challenged the heads of the business units in regular management meetings: he dug into the numbers to discover what had really happened in the previous quarter and tested the executives on the competitive environment—all to gauge what level of performance might be possible. Targets were thus often increased throughout the year. All unit managers attended these quarterly reviews, and the sense of theater they created was a powerful incentive for each manager to come prepared to give a convincing performance. Despite the strengths of this process, the company persistently fell short of its aspirations.

Starting to dig

One newly appointed business division head felt so strongly that ManufactCo had lost its performance edge that he was determined to do something. But he also felt sure that there was no simple fix and that ManufactCo would need to understand the problems fully before they could be solved. He therefore invested an unusual amount of time and resources in diagnosing ManufactCo’s performance issues, starting the process by assembling an unusually large team composed of a dozen future leaders—not yet business-unit heads but highfliers—whose role was to pinpoint them.

The team began by interviewing 60 members of the organization: the entire divisional leadership team and about 30 others who represented all key functions and businesses—again, a lot of people for a diagnostic exercise. The members of the interviewing team had no preconceived ideas about what they would discover. They built the interviews around ten open questions, crafted not to lead the interviewees, covering behavioral and more formal, structural issues. Among other things, the team asked, "What are the barriers to improving performance?" "Do processes facilitate or get in the way of delivering performance?" "How effective is teamwork?" "What strengths of the organization are worth preserving?"

After half a dozen interviews, it was clear that the leaders generally agreed on the company’s problems, if not the solutions. For one, the leaders found producing all the data for performance reviews a burden and the reviews themselves a trial rather than a catalyst for better results. Dialogue between corporate and business-unit leaders about performance no longer felt honest, exploratory, and inspiring but defensive and dispiriting. All of the leaders regarded the target-setting process as a convoluted game. Although the team now had a complete list of the problems, it interviewed all 60 key figures to ensure that they felt like part of the process. Word of the interviews had spread, and people wanted their points of view heard.

Having identified the problems limiting performance—many concerning the performance-management system—the diagnostic team was tempted to jump immediately to solutions. But it resisted the impulse because of previous disappointments with problem-solving initiatives that had made no discernible impact on performance. Instead, the team pressed on to discover the root causes.

Even deeper

This work took place in two one-day workshops, again involving the diagnostic team and all 60 interviewees. To uncover the causes, they took each surface problem revealed by the interviews as a starting point and continually asked, "Why is this so?" until they reached a fundamental constraint or insight. For example, from the starting problem—that "dialogue between corporate and business-unit leaders about performance no longer feels honest"—the process identified some fairly obvious causes, such as a lack of relevant data and high-level performance reviews that were insufficient as forums for debate. But the process also yielded a range of nonintuitive factors related to the philosophy of ManufactCo, its structure, the way its people behaved, and even its human-resources practices. Managers found it hard to have open, hard-hitting debates about performance, for example, partly because they hadn’t been sufficiently trained or coached in handling meetings. This process showed that the absence of honest dialogue about performance didn’t arise from a lack of integrity or honesty by individuals; in one-to-one conversations, they were disarmingly frank, as their openness in the diagnostic interviews demonstrated. However, this honesty was lost in the group meetings of the formal performance-management processes.

As the workshops went deeper, the participants uncovered complex, systemic connections between the problems and their causes. They even found that apparently useful beliefs and behavior were having malign effects. The value placed on delivering performance, for instance, might seem to be an asset, but it meant that managers who failed to meet targets were stigmatized, losing standing and advancement prospects though rarely getting fired. Business-unit heads thus went into the annual target-setting negotiations aiming to get sufficient slack in the initial targets to accommodate the extra ones that would be added during the year. Sensing that unit heads were not telling the whole story, ManufactCo’s leaders made them produce more and more data for performance reviews so that the leaders could challenge the proposed targets effectively. Naturally, the unit heads resisted, so the process still yielded too little real information about actual performance and what could be done to improve it. Consequently, the game of sandbagging targets and requesting more data continued, thereby creating a self-reinforcing downward spiral of related problems.

Putting such a premium on better performance had other implications, too. On the plus side, managers were continually looking for opportunities to improve performance and were good at finding them. The company had thus kicked off a plethora of initiatives, but it lacked an effective mechanism for prioritizing them or stopping those that weren’t delivering, leaving far too many to manage. Resources were stretched too thin to complete and track the implementation of all the initiatives, so few delivered as promised.

Designing solutions

The diagnosis had now grown from a set of confidential interviews into a journey of collective discovery. Feeding back to the 60 interviewees a summary of their own revelations quickly opened up powerful, honest, and frank debates among them, thus creating a sense of uncomfortable liberation. There was liberation because for the first time they could discuss the organization’s disappointing performance openly, with the prospect of being able to do something about it, but the uncomfortable aspect was admitting that their own behavior was part of what had gone wrong. Even so, the group had gained a deep and broad understanding of the problems as well as their root causes and interconnections. Its members also shared a view of the organization’s strengths, a high level of engagement, and a desire to make positive changes. When the interviewees relayed their experiences to their own work teams, this sense of uncomfortable liberation spread through the organization, making the diagnosis a collective achievement.

Now that the organization understood the systemic relationships between its problems and their causes, the time was right for the team to design a program for solving them holistically. It understood the links between the many problems the interviews had revealed, so instead of finding individual, unrelated solutions, the team aimed to fix the system as a whole. First, it developed a detailed list of strengths to be preserved and underlying issues to be addressed—in effect, creating the design criteria for the solution. Then it planned interventions to combat every root of each problem, down to the final "why."

You can see how this process worked by looking again at one of the problems: the lack of honesty in discussions of performance. Since the roots of the problem and its links with the others were known, the design team could develop a suite of interventions to eliminate it for good. While recognizing the essential differences among ManufactCo’s businesses, for example, the team drew up measures to simplify the data required for performance reviews. It also created more forums for performance dialogues in order to involve everyone in the organization, from the front line to the board. It tackled the less obvious causes, too—for example, by training people throughout the company in skills such as conducting challenging dialogues and debriefing meetings to ensure that their objectives had been reached.

Throughout the design process, the team used the interviewees as a sounding board for its proposals by conducting a series of miniworkshops and focus groups, which helped ensure that every element of each downward-spiraling loop was tackled satisfactorily.

The wider applications

This collective-discovery approach can work on the whole range of organizational systems and processes. For example, a global industrial company used collective discovery to fix problems in its planning system. This company had one underachieving region comprising six business units and 12 sites. Collective discovery revealed that local managers lacked a common understanding of the company’s regional strategy. They understood the future challenges broadly described in the high-level strategy—for example, their region was expected to deliver earnings and capital growth. But these challenges had not been translated into concrete objectives and actions for the units; it wasn’t clear, for example, whether they should focus on ROCE or cash improvement to deliver immediate growth or on capital investment for future growth. Nor was it clear what skills the units would need to meet these aims. Deeper analysis of the issues and the links among them resulted in a new planning system. Together, unit and regional leaders filled in the missing details of the regional strategy, generating a plan that reflected the high-level strategy but was much more specific about what would be expected of each unit and the capabilities and processes it would need.

In fact, collective discovery usually discloses more than one organizational system that needs fixing, as it did at both ManufactCo and the industrial company. But when program designers develop systemic solutions, they would do well to work on a single system at a time, starting with the one that has the most far-reaching ramifications.

Outcome

It is still early to declare a resounding victory for collective discovery at ManufactCo, but a number of unusual advantages are emerging there. For example, the level of buy-in from the people involved is far greater than it was during previous major change programs. Each meeting between the team and the interviewees helped create a groundswell of energetic support for change. The scope of the transformation and the speed of implementation have also been far greater than they were in the past. And the process itself has brought about important changes in the way managers behave. For example, the initial interviews revealed that one of the organization’s weaknesses was a failure to celebrate its successes. The interviewees and the design team therefore immediately worked out the objectives of the collective-discovery process they had embarked on and made sure that they celebrated when these objectives were met.

For organizations interested in the collective-discovery approach, the experiences of companies that have used it offer some useful lessons.

1. Don’t forget the organization’s strengths. During collective discovery, it is easy to lose sight of them; ManufactCo, for example, wasn’t a poor performer—it had just lost its edge. It needed only to consolidate its strengths and to address its problems.
2. Choose the team with care. If a fairly large group of respected managers leads the diagnosis, they cannot be dismissed as just a few individuals airing private grievances.
3. Be inclusive. If people’s behavior and interactions are part of the problem—and they often are—everyone involved must gain collective ownership of the diagnosis and the solution. One person complaining about the way managers behave is a whining misfit; a broad cross section of managers complaining about poor behavior, including their own, cannot be overlooked. Collective ownership of the solutions gives them much more sticking power than top-down initiatives or fixes that result from turning up the heat under particular individuals could ever have.
4. Create a common language. It isn’t easy to talk meaningfully about mental states, behavior, and interaction styles—partly because there is no simple, rigorous language to describe social interactions and partly because discussing emotional topics can make people defensive. An organization must therefore take the time to develop and disseminate a single, neutral language to discuss its behavioral characteristics. At ManufactCo, for example, the word "assistance" became imbued with a meaning that reflected the value of acting for the company’s greater good. The choice of words is not as important as the shared meaning created within the team. A shared language also lets individuals in the company communicate its values and approach to new members, thus helping them integrate quickly.
5. Be holistic and systemic in problem solving. Many companies have a "default" lever they pull in times of poor performance, be it reorganizing, tightening the performance culture, or developing a new strategy. It is important to think beyond this single lever both during the diagnosis and when designing interventions and to use a broader range, including levers that will help replace a negative performance cycle with a positive one.

The collective-discovery approach proved to be a remarkably powerful way of helping ManufactCo’s people to understand the performance challenge they faced and to translate that understanding into consistent, complementary, and far-reaching actions. The paradox of a company with a great strategy and great people but lackluster performance can be resolved by delving inside the organization to unearth the root causes of its problematic behavior and mind-sets—and by engaging the organization as a whole in the struggle to eliminate them.