After 30 years at Procter & Gamble (P&G), I tend to approach things like organizational design from a practitioner's perspective. I believe that theories and models are only as valuable as their ability to help order thinking and spark action. In other words, I focus on practical application. Of those 30 years at P&G, 22 were spent in Purchases—what other companies may refer to as supply management, sourcing, procurement, purchasing, or even supply chain management.

But no matter what you call it, the organization that interfaces with your supply base rarely receives the attention it deserves. While management theorists and business executives focus on marketing, finance, product design, and even manufacturing, the supply organization is often forgotten.

More and more, however, people have begun to realize that suppliers make up a growing proportion of a company's cost structure and operational capability. At a company like Procter & Gamble, for example, 60–70 percent of the corporation's cost structure depends on suppliers that provide a wide range of goods and services. A wrong decision in supply organization design can result in a rapid loss of expertise and competitive stature by reducing a company's leverage and splitting supplier relationships across multiple sites, businesses, and geographies.

Consider an example from P&G. In the late 1980s, we moved to an organizational design that combined purchasing, customer service, and plant engineering under a largely manufacturing leadership for each of our business categories in each of the four regions of the world. For Purchases, this meant decentralization and, within three to four years, a shocking loss of expertise.

When we "reassembled" our sourcing leaders—several of whom, like me, had transferred to manufacturing—we found our leverage gone and our purchase costs too high. We began bench-marking and found, to our dismay, that our manufacturing plant-centric designs had allowed others to pass us. Not only had our skills eroded, but others had taken sourcing and supply management beyond where we had been. By the early 1990s, when we fully realized what had happened, we knew that our work was cut out for us. What P&G needed was a disciplined, holistic way to think about the organizational design for sourcing so that we would not make the same mistakes again.

We found it in a theoretical model called the Organizational Performance (OP) Model. Developed by Dave Hanna, the OP Model builds off open systems thinking and Jay Galbraith's "design star" work. Hanna was a P&G employee when he published the model in his 1988 book, Designing Organizations for High Performance.

Shown in Exhibit 1, the model is used both for organizational assessment (is the organization "up to" the needs of the business?) and for organizational design (what organizational structure will meet our business needs?). Assessment moves clockwise around the model and design moves counter clockwise. At P&G, we found that the OP Model encourages managers to think holistically and create a design that delivers results and not just a bunch of lines and boxes that look good on paper. A simplified description of the five segments of the OP Model follows.

1. Business Situation—An environmental scan of business goals; competitive position; employee, stakeholder, and governmental expectations; and internal resources, products, and personnel. Both the design and assessment processes start with understanding the current and likely future business situations, and then defining the gap between the two.
2. Business Strategy—The key choices that determine how company resources interact with the market.
3. Design Elements—The six essential elements of any organizational design: tasks, people, information, decision making, rewards, and structure. When all six elements are included in the design process, they ensure holistic thinking. The bulk of this article will focus on these elements applied to supply organizations.
4. Culture—The operational behavior of the people doing the work and the norms that shape their efforts. The six design elements play out in an organization's culture. For example, if a supply organization consistently awards business only to the low bidder, its culture is price-driven, regardless of business strategies about quality or innovation.
5. Business Results—The outputs of the organization for example, cost, market share, profit, sales, and innovation.

The Organizational Performance Model is applied to the supply organization differently from how it is applied to most other functions. This difference relates to the fact that the supply organization is, by definition, a boundary function. Its role is to link the company's business needs with the capabilities of external suppliers. For this linkage to provide real value, supply must be able to interface with two very different worlds—the external world of the marketplace and suppliers and the internal world of business units, projects, and functions. The perspectives and approaches of these two worlds can be very different. Even the language can differ. At P&G, for example, we use a unit of volume called a "statistical case" to compare our wide range of consumer packaged goods products. This internal language means absolutely nothing, however, to our suppliers, who count in tons, kilograms, thousands of units, hours of service, and so on.

The dilemma of a boundary organization is that it must "face both ways" to interact with the very different environments on either side. This is the twist when the standard OP Model is applied to a supply organization. When the OP Model is applied to a supply organization, the company must look at the design elements both from an external perspective and from an internal perspective. Then the final design must find a means to connect the two views.

When designing an organization, most managers leap directly to the structure of the group— boxes, reporting lines, and responsibilities. But that puts the "cart before the horse." The place to start is with tasks—what people in the organization actually do. In supply, two sets of tasks quickly emerge:

1. Commercially linking with external suppliers and markets to deliver value to the company, its products, and customers/consumers. These are the external-facing tasks.
2. Operationally linking with internal customers to understand business needs in order to accurately present those needs to suppliers and to provide internal functions with external supply market intelligence and actionable opportunities. These are the internal-facing tasks.

External-facing tasks form the core focus of a purchasing or supply organization: commodity management, industry interaction, and supplier relationship management. If a company just looked at the supply organization's external tasks, it would design the organization around commodity/industry clusters for common purchased items and around supplier coordination structures, especially for large suppliers. However, any organizational design must also consider the supply function's internal-facing tasks.

These internal-facing tasks enable the purchasing organization to communicate corporate needs to external partners and to survive within the company's internal structure, which is often configured quite differently than the outside world. There are two types of internal tasks: business process management and organizational management. The first is the supply organization's version of the tasks described in Peter Drucker's classic Management: Tasks, Responsibilities, Practices: price, cost, and profit forecasting; project and new product launch processes; cross-functional team formation; budgeting; change management; and operational collaboration.

The second set of tasks is much closer to home—leading and managing the supply group itself. These tasks include: recruitment and staffing, coaching and people development, salary management, coordination of the supply organization, and clarification and deployment of the deluge of demands and information from the corporate center.

Successfully carrying out these internal and external facing tasks is only one aspect of the tasks element as it applies to supply organizations. A second dimension is how tasks are bundled within and across the organization to deliver needed results. This bundling depends on the business situation and strategy. For example, in P&G's coffee business, we realized that sourcing/commodity trading, logistics, and coffee tasting/quality were all intimately connected. This realization led us to broaden the supply organization to include people and tasks that previously had resided in manufacturing and product development organizations. At the other extreme, when P&G acquired Giorgio Cosmetics, we found that their supply group was focused on complex logistics flows rather than sourcing. The pressure of day-to-day product replenishment required us to divide the tasks between a small cadre that could focus on strategic sourcing and supplier relationship and a larger group that stayed focused on the critical logistics flow. In the first case, supply's tasks were expanded and integrated into a single organization. In the second, we split the function into smaller, more focused groups.

From the "what" of tasks, we now turn to the "who" of people. "People" includes not only individuals, their abilities, aptitudes, and skills but also the processes that acquire, retain, develop, and sometimes terminate individuals (through firing) or groups (through outsourcing). The people element also, includes the aggregation of individual skills into organizational strengths and weaknesses. At the organizational level, this element represents the sum of the individuals, the resources that support them, and the relationships both inside the purchasing/supply organization itself and across other functions in the company and suppliers. Just as with tasks, sourcing organization design must consider both those aspects of the people element that are necessary for interacting externally with suppliers and those aspects that are necessary for successfully interacting internally.

External-facing people factors are those skills, resources, and processes that are necessary for interacting successfully with groups outside one's own company. They include the skills your people need to successfully collaborate, negotiate, and communicate with these groups. These factors also involve externally acquired resources that enable your people to enhance their contribution to the company, such as e-procurement tools from external providers and external skill-building and training opportunities.

Internal-facing people factors are the resources and support required to enable your sourcing people to improve and effectively perform their key tasks. Often this means the money to staff effectively or to invest in external and internal tools and training. However, at the organizational level, we are also talking about hands-on coaching and peer skill support throughout the organization.

Using Hanna's OP Model to consider the people element of design has helped P&G Purchases to redesign its organization to respond to changes in the external and internal business environment. For many years, Purchases was a "functionally pure" organization. We had very few cross-functional transfers and, in the United States, we only recruited externally for entry-level positions. These external hires were typically either recent college graduates or ex-junior military officers (JMOs). By the late 1990s, however, a combination of external and internal people factors required that the organization change. Externally, our traditional entry-level hire pool was shrinking. The number of JMOs had dried up as military numbers dropped, and fewer students were seeking single-company careers. At the same time, there was a growing demand internally for Purchases talent. Yet an internal study revealed that P&G Purchases was able to retain only 42 percent of college hires for five years or more vs. 78 percent of hires with prior experience at another company.

Because of these factors, the Purchases leadership consciously decided to reduce campus recruiting from nearly 100 percent to about half, while expanding the number of cross-functional internal transfers and entry-level hires with one or two years experience at other companies. The goal became to balance these three "people sources." Today the organization has a blend of functional backgrounds at nearly all levels up to director and vice president. The corporate "promote from within" culture continued, while the mix, capability, and diversity of talent all expanded.

For supply, information is the raw material from which sourcing is fabricated; without information, sourcing professionals will make flawed decisions with catastrophic business results. As the building block of sourcing, information is a critical design element. Organizational structure needs to consider where information comes from, where it goes, where it resides, and how it is controlled.

There are several different types of information: transactional information (such as specifications, prices, and purchase orders), business management information (volume and cost plans, exception management, and new product introductions), and market information (commodity supply and demand and industry and supplier analytics). All of this information flashes through, across, and within the supply organization to internal and external parties (see Exhibit 2).

There are also three main flows of information. They are: (1) external flows between companies (such as information about markets, suppliers, economics, and currencies), (2) internal flows across internal boundaries (such as information about cost, volume, availability, and demand); and (3) intra flows within the supply organization itself (for example about budgets and staffing).

To design an effective supply organization, you need to know what information flows run through your organization and to envision how these flows interface. You also need to understand how the information would be used in practice. For example, sourcing managers do not need information about strategy development in real time, but they may need daily management information available in real time. Performing this mental hook-up will help you gain a more overarching analysis of your information needs, requirements, and uses.

Some questions to ask yourself when conducting this analysis include:

• What needs to be connected so that you source well? Be sure to think down, up, and across your organization.
• What information flows—external, internal, and intra—does your organization contribute to?
• What types of information does your organization need? Where do you get your information from?
• What are the gaps in your information and in your knowledge?
• How will you control information? What control points do you need in order to measure, monitor, and protect your information?
• How will you deploy information? How will you "get the word out" about organizational strategy, change management, and direction setting?

How decisions are made is a key design element—one that can make or break an organization. Supply organization design needs to consider both the internal and the external sides of the decision-making process. The internal piece involves ensuring cross-functional alignment, or getting all of the company's functions on the same page. The external piece involves all the choices the supply organization must make relative to suppliers. Such decisions include: Should we negotiate, competitively bid, or reverse auction? Which suppliers should be included? How does the supplier quote stack up against its competitors and relative to the market?

Decision making should be an interactive process between the external and internal sides—neither should be addressed in isolation. Who must agree, who actually decides, and how the internal/external iterations are sequenced all matter. Pity the buyer who cuts a deal and then must sell it to the internal stakeholders whose input was ignored or never gathered.

Getting internal cross-functional alignment is often the toughest part of the process. Ask any negotiator who has had to represent a range of internal constituencies in a supplier negotiation, keeping the various groups on the same page takes as much or more energy than the actual supplier interface. Before attempting a market intervention with suppliers, you need to understand who the internal players are. The loss of credibility and level of embarrassment when a commitment is overturned internally can undermine the ability to get best value—both "this time" and in the future. The supply organization must be able to paint the broader strategic and market pictures to the internal players while also understanding their needs. The organizational design must truly map out this interaction so that the supply organization can successfully create the right internal connections and expectations to get strong executable decisions.

The external side of decision making spans not only reviewing the final supplier and terms of purchase but also determining what levels of review and involvement are required from sourcing management. At P&G, typical criteria used to make such choices include:

• Size, implications, length, and risk of the business to be awarded.
• Skill, experience, and knowledge of the people closest to the work.
• Accountability for results. (Who is responsible, and how comfortable are they with the choice?)

An important concern here is the balance between a decision-maker's accountability, specific operational knowledge, and ability to apply general expertise to a particular situation. For example, new employees should not initially be expected to make many large decisions. As they grow, however, they should receive more authority to make commitments. Large, high-risk supply agreements will require more senior involvement, especially when they involve major strategic suppliers. However, elevating decisions to senior managers who are unfamiliar with the issues or do not possess strong expertise can result in poor choices. Because major companies have a large number of supply agreements, they need strong, savvy mid-level sourcing managers with the authority to make decisions. This ensures that decisions are made efficiently and are not bottlenecked at senior levels.

Organizational design should ensure that it encourages and rewards actions that meet the business's needs and are in line with corporate strategy. A student in one of my MBA classes at Xavier University, who was the human resources manager of a mid-size company's plant, captured the essence of what can happen when the reward element is poorly executed. She remarked that her organization was told to significantly change its structure and tasks but that "corporate" kept the same measures and continued to reward senior leaders for doing what had been done in the past. Consequently, results proved disappointing, and people were confused about what to do.

Too often, leadership is so close to its strategies and goals that they forget to explain the actions they want to get results. To them, it is obvious what must be done. Several levels below, however, things are rarely that clear. To avoid this disconnect, when P&G began global sourcing, it rewarded its global teams based on three criteria:

• Strong overall corporate results, or the global aggregate result.
• No single region having to take a huge loss so that others would gain. This avoids competitively disadvantaging one region.
• No single region consistently losing over time. This avoids incrementally creating a competitive disadvantage for a region in the long run.

Sometimes these criteria were at odds with maximizing local performance. In one case, globally sourcing for a particular spend item resulted in one region facing a $500,000 premium, while the rest of the world saved more than $5 million. That particular region, however, had had the lowest cost the previous two years and had the second lowest cost, even with the premium, in that year. The criteria were clearly met, senior management engaged, and the globally optimal choice made. The rewards had elements of both global and regional priorities, and supply chain/finance/purchases recognition was included in the overall reward system.

Another aspect of the rewards element is balancing corporate, business unit (or regional), and functional rewards. In the global case above, corporate recognized that Purchases was fulfilling the business need of producing good overall savings. Additionally, the three regions that saw reduced costs were happy with their global sourcing team, while the fourth was recognized for the good work it had done for the company overall. Finally the finance, product supply, and purchases functions all saw the team's work (including those in the dissenting region) as excellent.

Organizational design needs to ensure that rewards are relevant to the individuals as well. For the short term, that typically means money and recognition. Longer term, however, rewards must come in the form of career development and growth. In our example, the global sourcing team leaders were regional business employees, but their skills were visible globally. Of the eight team leaders, six were promoted within two years based on broad recognition of results and personal skill growth. Our global (external) and regional (internal) reward systems were aligned.

Finally! Lines and boxes! "It's about time," you say. Structure is last for a good reason. Form follows function and must align with business strategy, culture, and rewards as well as the structures of other internal organizations. For supplier-facing organizations, this requires a three-part structural design.

1. An "extra" design, or the structure that interfaces with the outside world of suppliers and markets. "Extra" here means focusing outside the company.
2. The "inter" design, or the structure that interfaces with internal units, sites, and projects. "Inter" here means interdepartmental across the company.
3. The "intra" design, or the structure inside the supply organization that translates, coordinates, and integrates the extra and inter designs. "Intra" here mean within the supply organization.

The success of these designs depends on how compatible they are with the other five elements as well as the ability of people to maintain both internal and external viewpoints when doing their work.

In supply, the "extra" design is often structured around commodity. Why? So that the company can aggregate requirements and leverage its spend, maximizing the value it receives for the money it spends. This external design is often the formal, visible structure everyone sees.

On the internal side, supply design should recognize that each business is unique with its own needs, rules, and people. Therefore, supply needs to have an additional "shadow" organization that matches up with how the company's internal business units are structured. This shadow organization lies behind the visible, external organizational design depicted in organization charts. The internal linkage plays out in the task element as the internal tasks. (By the way, if you choose to make the internal ("inter") design the visible one, recognize that you will still need a shadow design to match up with the outside world. You must have both.)

Finally, the "intra" design represents the "control room" between the visible and shadow sides of the structure—it is the brains of the place. The intra design provides managerial oversight of the supply base, people resources, business results/need fulfillment, and conflict resolution across business units. Its job is to make sure the supply base delivers the right blend of aggregated scale and customized value so that all businesses win and the suppliers do not get pulled in conflicting directions that compromise results.

Conceptually, "structure" should represent the summary of the design process. The beauty of Hanna's thinking is that it allows the organization to blend elements of seemingly opposite designs into a single holistic unit. For example, in 2000, some P&G businesses created purchasing subunits to focus on new types of products with components we had never bought before—such as Swiffer implements and electric batteries for the Crest Spinbrush. These new P&G products also needed materials from traditional spends with existing commodity structures, like packaging or chemicals, but at extremely small volumes that could be lost in a larger aggregated requirement.

The use of the Organizational Performance Model allowed P&G to embed these "inter-designed" subgroups within the existing commodity-based organization. The new-product subgroup's shadow structure linked to the primary group's visible commodity structure, and the larger commodity structure's shadow organization linked to the visible new product group's internal business structure. This design allowed the new-product people to focus on new spends and coordinate the smaller "old" spends, while the primary group focused on our large existing businesses. The commodity structure still sourced the new business's small requirements for already existing commodities, but it let the new-products subgroup coordinate from there on. Little task duplication occurred between the groups.

The "intra" control room shepherded the process and, when some of these new businesses reached critical mass, transitioned the sourcing into the normal commodity structure. We used the model to get a hybrid centralized/decentralized structure under one roof that rewarded both types of work and results.

None of this is rocket science. Others have probably designed similar things into their organizations. However, P&G's experiences show that using a holistic design model, like Hanna's OP model, helps organizations recognize that structural success depends on nonstructural systems—in other words, those first five design elements. For a design to be successful and deliver results, all of these elements need to be incorporated.

The point of organizational design is to focus your organization to deliver results. While structure is surely part of that focusing process, just "drawing new lines and boxes" is not enough because results are delivered by people. How they do the tasks that make up the work is key. A successful design process involves holistic thinking that systematically looks at the nonstructural elements in concert with structure. People often operate differently than structural charts would predict; they are driven by things like rewards, access to information, and how decisions are made.

Lastly, we are competing in a constantly changing business environment that demands that organizations continually recreate themselves. The Organizational Performance Model provides companies with a repeatable thought/design process, which is linked to business needs and cultures. When applied periodically by business leaders, this model gives organizations a systematic way to evolve with changing situations. Building regular design renewal into your organizational plan can ensure that your supply design doesn't get left behind by the realities of changing internal and external environments.