STRUCTURE AT UNITED PARCEL SERVICE (UPS)

Ideas about structure in organization have inherited some misleading baggage. One fallacy is the equating of structure with rigid top-down policies and rules, bureaucracy that impedes work and frustrates workers. In fact, when strategy, rules, policies, control, and measurement are right for an organization’s circumstances, people become more productive and satisfied. United Parcel Service provides a familiar example.

The main purpose at UPS—“Big Brown”—is to deliver packages on time to make customers happy. In the early days, UPS delivery workers were “scampering messenger boys”¹ who carried packages to department stores. In recent years, computer technology has replaced employee discretion, and every step from pickup to delivery is highly routinized. Every movement of drivers at UPS is studied, refined, and programmed: “In God we trust; everything else we measure.”² Detailed instructions specify where and in what order to place packages on delivery trucks. Drivers follow computer-generated routes (which minimize mileage and left turns to save time and gas). The number of steps to your door is premeasured by GPS. If a driver sees you while walking briskly to the door with your parcel, you’ll get a friendly greeting. Look carefully and you’ll notice that the driver carries an electronic locking device. That’s part of the delivery routine: get out of the truck, retrieve package, lock truck, place package in designated place, jauntily return to truck, unlocking door en route. Given that they’re on such a tight leash, you might expect demoralized employees. But the technology makes the job more predictable, helps drivers be productive, and keeps customers satisfied. As one driver remarked with a smile, “We’re happy robots.”

McDONALD’S AND HARVARD: A STRUCTURAL ODD COUPLE

A second misunderstanding about structure is that there is one best way to organize. There is no shortage of consultants hawking proprietary models. But the most effective social architecture—hierarchy of authority, division of labor, and coordination of work—depends on how leaders assess the situation. Consider two contrasting examples.

McDonald’s, the company that made the Big Mac a household word, has been enormously successful. Since Ray Kroc started to take McDonald’s across America in the 1950s, the company has become an almost unstoppable growth engine, dominating the worldwide fast-food business. McDonald’s has a relatively small staff at its world headquarters near Chicago; most of its employees are salted across the world in more than thirty-one thousand local outlets. Despite its size and geographic reach, McDonald’s holds things together in a centralized organization in which most major decisions are made at the top.

Managers and employees of individual restaurants have limited discretion. Much of their work is controlled by technology; machines time french fries and measure soft drinks. The parent company uses powerful systems to ensure that customers get what they expect. A Big Mac tastes about the same whether purchased in New York, Beijing, or Moscow. Guaranteed standard quality inevitably limits the discretion of people who own and work in individual outlets. Cooks are not expected to develop creative new versions of the Big Mac or Quarter Pounder.

All that tight structure might sound oppressive, but a major miscue in the 1990s resulted from trying to loosen up. Responding to pressure from some frustrated franchisees, McDonald’s in 1993 stopped sending out inspectors to grade restaurants on service, food, and ambience. When left to police themselves, some restaurants slipped badly. Customers noticed, and the company’s image sagged. Ten years later, a new CEO brought the inspectors back to correct lagging standards. But even as it centralized quality control, the company also gave regional managers more leeway to align offerings with their local market in response to globalization and a desire to serve customers better. So a burger will taste pretty much the same wherever you buy it, but you can also get breakfast porridge at McDonald’s outlets in England, veggie burgers in India, and burgers-on-wheels home delivery in traffic-choked cities such as Cairo and Taipei.

On the other end of the spectrum, Harvard University is at or near the top of almost every list of the world’s best universities. Like McDonald’s, it has a small administrative group at the top, but in most other respects, the two organizations diverge dramatically. Harvard is more geographically concentrated than McDonald’s, but it is significantly more decentralized. The bulk of Harvard’s activities occur within a few square miles of Boston and Cambridge, Massachusetts. Most employees are housed in the university’s several schools: Harvard College (the undergraduate school), the graduate faculty of arts and sciences, and various professional schools. Each school has its own dean and its own endowment, and, in accordance with Harvard’s philosophy of “every tub on its own bottom,” each largely controls its own destiny. Schools have fiscal autonomy, and individual professors have almost unlimited discretion over courses they teach, research they do, and university activities they pursue, if any. Faculty meetings are often sparsely attended. If a dean or a department head wants a faculty member to chair a committee or offer a new course, the request is more often a humble entreaty than an authoritative command.

The contrast between McDonald’s and Harvard is particularly strong at the level of service delivery. No one expects individual personality to influence the quality of McDonald’s burgers. But everyone expects each Harvard course to be the unique creation of an individual professor. Two schools might offer courses with the same title but different content and widely divergent teaching styles. Efforts to develop standardized core curricula founder on the autonomy of individual professors.

In early 2000, President Larry Summers ran into the predictable challenges of trying to tighten up a professional organization. In attempting to achieve greater control over a fractious faculty, he inadvertently set off one bomb after another. In one case, he suggested that superstar African American studies professor Cornell West redirect his scholarly efforts. Summers gave his advice to West in private, but West’s pique soon made the front page of the New York Times. Summers’s profuse public apologies failed to deter the offended professor from decamping to Princeton. Summers resigned under duress in 2006 after the shortest tenure for a Harvard president since a long-forgotten incumbent died in office in 1862.

The examples of McDonald’s and Harvard illustrate the central idea of the structural lens: no organization can perform very well without strategies, roles, relationships, and coordination that are workable and appropriate for its circumstances. The right structure helps ensure that individuals know what they’re supposed to do and how they’re expected to work with others to get it done.

A basic leadership responsibility is to shape structure to fit the situation. In doing that, leaders always face three key questions: What are my strategies and circumstances? How do I allocate responsibilities across different people and units? And, once I’ve done that, how do I integrate diverse efforts in pursuit of common goals? We’ll explore these basic questions and describe options leaders consider when designing an arrangement that will work.

ELEMENTS OF SOCIAL ARCHITECTURE

Every structure is designed and crafted using a particular configuration of basic elements. One is the hierarchy of authority, or chain of command, typically with three levels: executive, managerial, and operational. Authority for making decisions can be concentrated at any of the three levels. A second element is the division of labor. Executives, for example, monitor the environment, set long-range strategy, and keep their eye on the bottom line. Managers set goals and objectives, supervise workers, and check short-term results. Workers at the operational level perform basic tasks.

Managers also have to decide how to group individuals into work units. They can choose among six basic options:

• Functional groups based on knowledge or skill, as in the case of a university’s academic departments or the classic industrial units of research, engineering, manufacturing, marketing, and finance.
• Units created on the basis of time, as by shift (day, swing, or graveyard shifts).
• Groups organized by product: detergent versus bar soap, wide-body versus narrow-body aircraft, smartphones versus tablets.
• Groups established around customers or clients, as in hospital wards created around patient type (pediatrics, intensive care, or maternity), computer sales departments organized by customer (corporate, government, education, individual), or schools targeting students in particular age groups.
• Groupings around place or geography, such as McDonald’s retail outlets in different countries or neighborhood schools in different parts of a city.
• Grouping by process: a complete flow of work, as with “the order fulfillment process. This process, as in UPS, flows from initiation by a customer order, through the functions, to delivery to the customer.”³

Once authority is established and roles and responsibilities defined, structural design needs to provide ways to link the parts together. The challenge is to develop an appropriate mix of vertical and lateral coordination.

• Those at higher levels provide vertical coordination by exercising authority, setting policy and strategy, and establishing planning and control systems.
• Lateral coordination happens through a variety of formal and informal roles, meetings, and groups. Individuals in coordinating roles have diplomatic license to span boundaries across specialized groups and areas. Matrix structures cross business and product lines. Digital technology provides rich channels of informal communication.

CONTEXTUAL FACTORS

Tinkering with structural arrangements requires a clear understanding of your situation. In developing the right social architecture to fit specific conditions, every organization needs to respond to basic contextual factors, outlined in Exhibit 2.1.

APPLYING THE STRUCTURAL FRAME

Suppose for the moment that you have become a regional sales manager for an investment bank.⁸ You manage a group that sells sophisticated financial instruments (bonds, mortgage-backed securities, options, futures, and so on) to institutional clients such as banks, insurance companies, and pension funds. Your office has been profitable and successful, but the business has become more competitive and complex in recent years. New low-cost competitors are entering the market and siphoning away sales of low-end, simpler products. Your firm is trying to shift its marketing mix to sell more complex products that have less competition and higher profit margins.

Most of your sales are generated by a small group of major account representatives who have the experience, people skills, intellect, and street smarts the work requires. Each of them sells your full line of products to a specific group of a dozen or so big institutional clients. A strong relationship with customers is vital, and each salesperson has a large expense account for entertaining. Top salespeople can make as much as $1 million a year in salary and commissions.

Here’s what worries you. As the products keep getting more complicated, it’s becoming very hard for your major account representatives and their customers to keep up. Your salespeople can get help from in-house specialists, but when a customer asks about a particular product, the account representative would rather give an answer than say, “I’ll have to get back to you on that.” Even more important, it’s hard to sell what you don’t understand, but the most profitable products are the hardest to decode. Market data suggests that competitors are gaining on you because of this expertise gap. What should you do now? Your first thought might be to put in some kind of training or motivational program for the salespeople. Managers often try to fix problems by fixing the people, but miss structural solutions that are more effective and easier to implement. Three simple questions can help guide a structural analysis:

1. What’s going on? What’s working and not working?
2. What’s changing (in your organization, your technology, or your environment) that creates an opportunity, a threat, or both?
3. What problem do you need to solve? What options should you consider?

In this case, your thinking might take you in a direction similar to the following:

1. What’s going on?

   Start by examining how the job of the account representatives is defined. Consider the options for dividing work (function, time, product, customer, place, process), and you’ll see that the role has been defined by customer. All the representatives sell the same products, but each has a unique group of clients. Examine the advantages and disadvantages of this arrangement, and you’ll see that it’s a good way to create strong relationships between representatives and their clients while giving customers the convenience of one-stop shopping. Whatever question they have or product they want, clients only need contact information for one person. It’s a straightforward structure that’s easy to understand and reduces the need for coordination among the sales staff.

2. What’s changing?

   A central idea in the structural frame is that structure needs to align with circumstances. If the environment changes, there’s a good chance that the structure has to adapt. The representatives have been generalists who cover the entire product line. That used to work well enough, but it is turning into a liability as the firm tries to shift to more sophisticated products. The representatives can’t always answer clients’ questions, and may miss opportunities because they don’t see the connection between client needs and the more specialized offerings. It’s tempting for them to keep selling the simpler, lower-margin products that they understand, but the high end is where competitors have been eating market share.

3. What problem do you need to solve? What options should you consider?

   The problem you’d like to solve is that of shifting the sales mix toward more high-margin products. What structural change would help solve that problem? If the goal is to increase the representatives’ ability to sell more complex products, a promising fix is to define the representatives’ job by product rather than customer. That would let each of the representatives specialize and become an expert on a segment of the product line. But there are trade-offs. Customers might be unhappy because it would make things more complicated; they’d have to contact different people depending on which product they were interested in. Moreover, it’s a structure that would require more coordination among the sales representatives in order to keep track of what’s happening with each customer. You don’t want customers feeling neglected because no one calls or harassed because everyone does. A third risk is that the new structure could fail if the representatives resist it or can’t adjust to it.

What to do? The analysis suggests that it makes sense to explore a move from a structure based on the customer to one built around products, but it’s not likely to work unless the representatives support the change and work as a team. That suggests involving the representatives in a discussion of how to solve the current problem, knowing that a solution that they believe in has a good chance of working and that the odds are against anything that doesn’t make sense to them.

In a real-world case similar to this one, that’s basically what happened. The regional sales manager held extensive discussions of the issue with the major account representatives, who eventually agreed to restructure so that each became a specialist in a particular product line. Seeing the importance of working together in this new structure, they defined themselves as the “key account team,” with one of their members serving as the coordinator.

After structural change, it is often the case that things get worse before they get better, and that was true in this case. There was short-term confusion as everyone tried to adapt to the new arrangement. Some customers saw the new structure as an improvement; others preferred the old one. Some of the representatives struggled with the transition. The new structure wasn’t perfect—no structure is—but six months later, the team was achieving the central goal: profits were up because they were selling more high-margin products to satisfied customers.

CONCLUSION

When leaders neglect structure and strategy, or buy into the two structural fallacies we discussed, they and their organizations pay a price. Units perform far below their potential. They waste time and money on training programs in a vain effort to solve problems that have more to do with social architecture than with people’s skills or attitudes. The right social architecture depends on prevailing circumstances and considers an organization’s strategies, technology, people, and environment. Understanding the complexity and variety of design possibilities can help you create patterns and prototypes that work for, rather than against, both your people and your purposes.

Ideally, leaders would be fluent and skilled in all four frames, but few of us are perfectly balanced. If structural thinking is one of your strengths, your natural inclinations can make you a valuable contributor in any team or organizational context. You can help diagnose structural gaps or overlaps and suggest ways to fix them. However, devotees of structural thinking often have difficulty seeing and dealing with messier and less rational human, political, and symbolic issues. Structure is important, but it is not the only dimension critical to top performance. Leaders who lean heavily toward structural thinking will typically feel comfortable organizing tight, well-managed teams focused on the task. But leaders need to be aware of the risk of becoming rigid, authoritarian micromanagers. They may have noble intentions and admirable concern for getting the job done, but may overlook human emotions, politics, and the cohesion that comes from cultural bonds rather than structure. We explore these important dimensions of leadership in subsequent chapters. In Chapter Three, we will home in on the structure of teams and small groups. More and more work gets done in such small units. But the call for more leadership talent grows stronger as these teams encounter structural issues that dampen performance.