點擊查看中文全文

Wharton management professor Ethan Mollickhas a message for knowledge-based companies: Pay closer attention to your middle managers. They may have a greater impact on companyperformance than almost any other part of the organization.

In other words, says Mollick, "the often overlooked and sometimes-maligned middle managers matter. They are not interchangeable parts in an organization." His view upends the long-held belief that performance differences between firms are due mainly to organizational factors – such as business strategy, management systems and HR practices -- rather than to differences among employees.

The importance of individual skills and characteristics can be especially significant when measuring firm performance in industries and fields that value innovation, like computer games, software, consulting, biotech and marketing, according to Mollick, who recently completed a paper on this topic titled, "People and Process: Suits and Innovators: Individuals and Firm Performance." It is in these knowledge-intensive industries where variation in the abilities of middle managers – the "suits" he refers to in his paper -- has a "particularly large impact on firm performance, much larger than that of individuals who are assigned innovative roles," he says.

The influence these suits exert, he suggests, stems from their key role in project management, including such tasks as resource allocation and supervision of deadlines – responsibilities often perceived as the bureaucratic, more routine and less glamorous side of the business. Middle managers also can play a key role in fostering innovative and creative environments.

A Look at the Gaming Industry

One challenge Mollick faced in his research was a lack of studies that measured the relative contribution of middle managers vs. innovators. He addressed that gap by analyzing the computer game industry, which not only is typical of many knowledge-driven industries, but also "represents a case where the tension between the firm and the individual should be at its most visible." The industry, he notes, is populated by companies that are relatively established, have clear product strategies and yet depend to a great extent on the "innovative output of key individuals." In addition, he writes in his paper, "success in the game industry relies not just on managers in charge of innovation, but also on project managers capable of organizing dozens of programmers and coordinating budgets that often reach into the tens of millions of dollars."

He does not include top managers in his analysis although his paper cites an earlier study showing that the impact of CEOs, CFOs and other top-level executives on large firms is limited. Indeed, these top positions explain less than 5% of the variation in firm performance among Fortune 800 companies – a finding that Mollick says is in line with other research in this area. In large, established organizations, "the top managers, at least, account for relatively little of why some companies perform better than others."

Mollick acknowledges that top management plays a significant role in setting the overall direction of the company. "But they don't have a big part in deciding which individual projects are selected and how they are run. At least for the computer game industry – and no doubt for a lot of knowledge-based industries – it is all about the middle managers."

His research differentiates knowledge-based companies from traditional industries where "economies of scale are critical, such as manufacturing, and where there seems to be little need to take individuals into account to explain performance." Toyota is an example. "With a six-layered bureaucracy, cross-trained workers and clearly delineated departments, Toyota built a manufacturing powerhouse that integrates workers in a complex mechanism to produce cars efficiently," Mollick writes. "Individual workers are ultimately replaceable and interchangeable with others who have received the same extensive training." The process "does not rely on any individual worker's skills but rather firm-level processes to hire and train the appropriate individuals for the appropriate roles."

Emerging industries, however, rely more on knowledge and innovation than on process and assembly lines. Given that model, what can be said about the effect of individual middle managers on firm performance? A number of studies suggest that "their success is heavily dependent on the structure of the organizations … and their impact on performance is determined by firm structure and culture, rather than individual differences."

Mollick's analysis of the game industry comes up with a different conclusion.

Good Skills Are Portable

In describing the focus of his research, Mollick notes that within the game industry, each game has a team of creators -- including designers, programmers and artists -- who work for firms of varying sizes. "Because accurate credits at both the individual and firm level are available for many games developed within the industry, it is possible to trace precisely both the individuals and firms responsible for innovation and entrepreneurship within the industry," Mollick writes.

The game industry can be broken down into producers – similar to project managers in the software industry – and designers. A producer (middle manager) has to ensure that the project meets its deadline, gets the right resources and conforms to industry standards. He or she must communicate effectively with the rest of the company and with outside vendors such as promoters and public relations firms, among other responsibilities. The designer (innovator), by contrast, comes up with ideas and helps the development team turn the idea into a game, paying attention to story lines and characters, but also to logic, sequence and interaction. The producers fill the managerial roles, and the designers fill the innovative roles, according to Mollick, who focused his research on PC games as opposed to console games like those that run on the Nintendo, Xbox or Sony systems.

Using a Multiple Membership Cross-Classified Multilevel Model (MMCC) analysis of 854 games across multiple companies, he measured, over 12 years, "how performance changes as you combine different people in different companies in different ways." He uses the revenue of each company, controlling for costs, to measure firm performance.

The games he analyzed accounted for about $4 billion of revenue and included 537 individual producers, 739 individual designers and 395 companies. With the MMCC model, he was able to determine which project success was due to individual designers as opposed to producers or the firms.

Mollick found that it was middle managers, rather than innovators or company strategy, who best explained the differences in firm performance. Managers accounted for 22.3% of the variation in revenue among projects, as opposed to just over 7% explained by innovators and 21.3% explained by the organization itself – including firm strategy, leadership and practices. "Far from being interchangeable," Mollick writes, "individuals uniquely contribute to the success or failure of a firm…. Additionally, even in a young industry that rewards creative and innovative products, innovative roles explain far less variation in firm performance than do managers."

Or, as Mollick later writes: "High-performing innovators alone are not enough to generate performance variation; rather, it is the role of individual managers to integrate and coordinate the innovative work of others." So while innovators may come up with new games and new concepts, managers play the more crucial role of deciding which ideas are actually given resources. It is this "selection ability" that Mollick measures.

The best managers are able to work closely with the innovators to turn their ideas into realistic project plans, he adds, and they are effective at motivating the team and facilitating "collective creativity."

In order to see if these skills were portable, Mollick re-examined the data, looking only at individuals who moved between companies. He found that middle managers who switched employers had an even larger impact on performance than those who remained within organizations. "This is not about a person being a good fit in just one specific organization. Their skills are useful anywhere." It's more evidence that managers are not "cogs in a machine. There is something innate in them that makes them good at what they do."

Beyond Dilbert

"It's amazing that the effect of these middle managers on a project is not only larger than the creative people, but larger than the rest of the organization," Mollick says. "We tend to think of companies as all about systems and not enough about people." He suggests that companies pay more attention to filling middle levels of management, figuring out who the best ones are and rewarding them appropriately.

Middle managers, he adds, "have a tough job." They are managing a finite set of resources, they don't have control over everyone's actions, they can frustrate people around them who are not interested in changing direction when necessary, and they must go in a direction – even if it's an unpopular one -- that ensures the project's success. Finally, the project has to fit in with the goals of the company. "It's always easy to think about the worst manager you have had, the ones you see in the Dilbert cartoons," Mollick says. "But it's important to recognize the vital role these middle managers play in making sure that information flows and that creativity happens."