Profitably Healthy Companies: Principles of Organizational Growth and Development

WHEN OUR CHILDREN were younger, they participated in an annual Maypole ceremony. It was a well-choreographed tradition. It would be fair to say, however, that despite its ritualistic elements, no contemporary Maypole ceremony was identical to the year before. The individual players brought an element of eclecticisms to the proceedings, and intruder wasps, children’s missteps and stumbles, or snagged ribbons were cause for stylish maneuvers that could disrupt perfection as drawn on paper. Nevertheless, the event never departed from what could be identified as a Maypole dance, and observers never were disappointed with the spectacle as their maturing offspring marched to a tethered rule of order.

These annual variations on a theme worked because the dancers could deviate from scriptures, not despite the fact. These improvisations enabled the dancers to smoothly twist along and wind their way to the dance’s inevitable end. Indeed, the least adaptive and most error-prone dance would be one that merrily proceeded without regard to what was occurring in the surrounding area. The performance would be compromised if adjustments were not made for, say, a fallen comrade. Thus, we expect sensible departures from routine when the occasion calls for it.1

Most organizations thrive on the rhythms of routines. The orderliness of organization is the way companies make business plans, set budgets, and review performance. Routines define what must happen, by who, and when. These recurring cadences are like heartbeats that give institutions life. Routines reduce confusion and conflict and increase the speed of execution through plug-and-play orchestration. These patterns, however, invite a paradox. Efficiency necessitates rules and repetition, but shifting circumstances necessitate adaptation and change. Therefore, organizations must be at once disciplined and flexible. Organizations that can do both have the wherewithal to morph with circumstances and to regain their shape once the need for transformation has passed.

Companies must be pliant without becoming scattered and chaotic. They must prudently react to the unexpected during turbulent times and flexibly bend when rushes of demand are placed on them by internal and external customers. The aim of flexibility is to be profitably responsive to environmental conditions by adjusting to the variability in the volume of customers, the number of products and services customers desire, the speed with which goods or services are needed, or the magnitude of custom requests to standard offerings.² “Profitable” is the operative word. Therefore, throwing money and people at a process problem would not be considered solutions because of the huge penalty in cost that flexibility would impose. Specifically, flexibility refers to an organizational ability to change and react to circumstances without incurring negative costs in time, quality, expense, or performance. Indeed, we will go one step further and say that employee flexibility should have a positive effect on corporate performance in terms of both mitigating risks because of exogenous shocks as well as improving stock returns. Indeed, a recent longitudinal study of firms coded as more or less flexible showed that flexible firms were more resilient in times of uncertainty and more profitable over the period of the study.³

Not every process requires flexibility, but those situations are increasingly rare. As long as people accepted one Model T, the process for its assembly was straightforward. The assembly of automobiles became more complex when consumers began to demand new models of different colors. The greater product differentiation that followed from more expansive consumer tastes, the seasonality of new models, consumer purchasing habits, and such required more flexible forms of organizing. Burns and Stalker,4 and like-minded authors who followed them,⁵ pointed out that organizational processes needed to conform to the requirements of the environment. Burns and Stalker specifically contrasted mechanistic and organic forms of organizing, maintaining that the former was more conducive to stable, predictable environments and the latter was more conducive to dynamic environments (see table 5.1).⁶ And, indeed, this generally has been true. For example, a recent study of 167 manufacturing plants indicated that more flexible (flat, decentralized, multiskilled employees) organizational structures were more efficient in producing mass customized outputs than mechanistic structures. The more uncertain and variable the demands of the environment, the more organic and flexible a system needs to be.⁷

In this chapter, we discuss what it means for an organizational process to be flexible. We proceed with the following caveat: although the term “flexible” can pertain to a spectrum of process elements, our focus as organization development (OD) practitioners is on people’s skillsets and the methods by which staff counts are determined and employees are deployed. Specifically, we are concerned with a company’s ability to reconfigure its human resources in response to environmental demands to preserve service and production levels.⁸ Therefore, we note in passing that different forms of flexibility exist, but these will not be central to our discussion. Technologies and machines, structures, people, and processes all can be more or less flexible.9 For example, a Philips screwdriver has one intended purpose, whereas a Swiss Army knife has several that can be variously deployed. With that advisory, our commentary primarily will be directed to people and ways to enhance organizational flexibility by expanding the skills and abilities of the workforce or by increasing (or decreasing) their numbers.

Efficient processes involve the adept merger of several elements. Every process has a logically related set of tasks to perform, and technologies, people, communications, and frameworks that enable the execution of these tasks. These elements are stuck together by culture and mechanisms that promote coordination. Several models of culture exist, but one we have seen explicitly applied in the operations literature was developed by Westrum.¹⁰ His descriptions of culture were designed to showcase the differences between generative and pathological cultures that spur or curtail process efficiencies in information technology environments. These divergent cultures are summarized in table 5.2 and, as can be seen, are illustrative of destructive or beneficial milieus that may pertain to any process found in any company. The generative, performance-oriented culture reflects workplaces in which new ideas flourish, cooperation is rewarded, and people feel jointly liable for the outcomes they achieve. Therefore, errors are public, nonincriminating affairs and are cause for collective problem solving, correction, and learning. Such cultures are evidenced by the nonthreatening way some organizations handle mistakes. For example, BambooHR asks employees to circulate their errors without fear of recrimination through an Oops email inox. The aims are to announce mistakes made, to make others aware of them, and to broadcast the steps that were taken to correct them. Therefore, the company communicates that mistakes are an inevitable part of the human condition, happen, and are fixable.

In contrast, pathological cultures crush initiative and produce alienated environments in which people keep their heads down and remain silent, still, and invisible. This self-imposed exile is a way to avoid detection and arouse the critical glowers of others. The less one interacts with others, the safer one is from extended hostilities. These repressive regimes propagate insular employee perspectives and encourage behaviors in which employees do their small, circumscribed parts and little more.

One issue that undoubtedly is familiar to anyone who has ever been a participant in a cross-functional process pertains to silos, or the issue of coordination among independently functioning units. Silos appear when people are comfortably sequestered in their specialized departments and are removed from happenings outside their secure environs. The mechanisms for managing their interdependencies are absent. Clearly, if people are not communicating and information flows are unconnected, a process will not be efficient regardless of anything else a company might do. In fact, what little semblance of structure exists when (internal) processes fail is further undone by workarounds—by individuals who circumvent processes to get what they need. When processes are broken, it is tempting to think that things could not get worse, but they can. Formal procedures can dissolve into free-for-alls. Companies sometimes euphemistically refer to this chaos as “relationship cultures.”

Therefore, organizations can help their cause by creating ways for information and materials to successfully travel across functional lines and tie the group’s activities together. In most ways, cross-functional teams succeed using the same facets of teamwork as any other team: through goal identification, strategy formulation, and system monitoring. Additionally, the work of cross-functional teams is further enabled by making the entire process visible to the people involved.¹¹ This visibility ensures that processors are working from the same mental model with the same understandings of the bigger picture. The display of the complete process also highlights everyone’s instrumentality toward, and accountability for, the end results. A comprehensive view of the entire process allows all parties—upstream and downstream—to see how they are performing as a unit. Too frequently participants are sheltered from the results of their work, for example, they never get to hear a customer complain about the atrocious service and, consequently, never have the same impetus to change as those who are directly exposed to customers’ dissatisfaction. In fact, one of the factors that makes systems reliable is the ready acknowledgment of situational failure and preparedness for learning and improvement, qualities that regular team briefings can help sharpen.

Relatedly, the connectedness of a system is furthered by sound knowledge management—by wrapping the requisite operational and technical knowledge around the entire process. In that way, those who have a role in the process are kept well informed and, consequently, can make intelligent, nuanced decisions and advance changes accordingly. Studies show that adept knowledge management, or the ability to acquire, convert, transfer, apply, and protect knowledge, is critical to the dynamic capabilities of organizations.12 Sound knowledge management promotes cross functional understandings, enhances sensitivities to changing conditions, and stimulates reengineering priorities, plans, and actions. An organization’s capability to process knowledge effectively reduces the impact of environmental uncertainty and increases effective responses. Therefore, companies that are able to generate new and relevant knowledge and diffuse, share, and apply it are more versatile and responsive to external conditions.¹³

Every process has inputs, throughputs, and outputs. The nature and issues concerning these processes, however, depend on the ingenuity of those involved. For example, bottlenecks are familiar in many U.S. airports (and organizations) because too many people converge on too few checkpoints at certain times of the year, week, and day. That is not a problem you will find at Singapore’s Changi’s International Airport because people go directly to their gates and are checked there.¹⁴ The customer’s burden of waiting is shifted to the responsiveness of security who must allocate labor to the right gates in a timely manner. In the case of Singapore’s airport, authorities make prudent use of floaters, or agents who serve in a generalist capacity and are assigned dynamically to go where the needs are greatest. In many companies, managers serve in the role of floaters, stepping in to aid their teams when and where the situation demands.

The counternormative approach of how to check passengers at airports demonstrates an important, often overlooked, fact. Process development is a creative endeavor that often turns on a simple challenge to assumptions, like solving a riddle. For example, hotels were spared expensive repairs in just this manner. Tired of customer complaints of waiting for elevators, hotels once thought they needed more and faster elevators. A psychologist, however, suggested that perhaps what hotels needed were more patient customers. In response, hotels enabled customers’ persistence by putting in windows with views by the elevator banks, hanging artwork and mirrors, creating seating areas with reading materials, and such—things to make the time, not the elevators, go faster. The same sort of mental gymnastics transformed Honda’s assembly process at its most advanced plant in Thailand. The plant contains the world’s first moving cell assembly line called the Assembly Revolution Cell Line. Rather than standing at an assembly line performing a singular task, four workers step onto a moving platform that tracks with Honda Civic shells and, with carts of parts by their sides (only the parts they will need), install ten to twenty parts. The people walk less, become more versatile in assembly through the execution of multiple related tasks, and are more cognizant of the entire assembly process—passing on their observations to developers. Production efficiency has increased by 10 percent.15 This process would not have been possible if someone had not challenged the assumption that people are supposed to be stationary in assembly—that only the line could move.

Breakthroughs in process come when existing assumptions that frequently go unspoken are identified and questioned. It involves a mode of thinking, called divergent thinking, in which the uses of objects are reimagined or assumptions that have entrapped thinking are challenged and new ideas are set free (for an example, see box 5.1).¹⁶ Thus, novel design thinking is a prelude to enhanced processes. The same sort of originality is valuable to other aspects of process as well, as when designers consider how to balance the demands on a system with the system’s capacity to respond.

Box 5.1
Duncker Candle Task

Fix a lit candle on a wall without wax dripping on the floor using only a candle, a box of matches, and a few tacks.

Once people abandon the assumption that it is the candle itself that has to be attached to the wall and that the matchbox only as a receptacle for matches, the problem-solver gains new perspective. The box can be attached to the wall with tacks with the candle placed in it.

In New Haven, Connecticut, a line forms daily outside of Pepe’s Pizza, a landmark eatery in the city. Patrons stand outside and wait their turns to be seated and served, frequently indifferent to the weather. Pepe’s could move to another, larger location with greater seating capacity, but their current location has historical significance. Although the proprietors built a small spillover space nearby for those patrons who do not want to wait and are willing to pay for a lesser experience than offered at the original site of the restaurant, they have decided to marry demand with capacity in a different way than creating a larger space. Pepe’s knows that their lines are chic but that people will not stand in line forever. Therefore, they have more wait staff per table than typical, and service is based on variations of a single product so that tables turn rapidly. Those waiting outside know that their turn will soon come.

Emergency rooms take a different approach, although the goals are similar. Emergency departments must provide the right number and mix of personnel in a way that balances utilization and revenues against the acuity of the case load and wait times. Emergency rooms are notorious for their crowded waiting rooms and lines (although, interestingly, those waiting never really know where they are in the line and feel more like lottery winners when their names are called). In fact, many people who visit emergency rooms leave without ever having been seen. Emergency rooms manage demand by conducting triage to channel traffic based on the severity of patients’ conditions and attempt to siphon nonemergency cases out of the system by promoting patients’ use of telehealth phonelines, colocating (with the emergency department) general practice walk-in clinics, or encouraging the use of in-home care.17 Thus, effective processes are those that regulate input and output rates in a way that productively meets the needs of customers (or patients). These regulatory methods are enlisted with the aims of increasing demand, decreasing demand, rechanneling demand to other resources, or altering the timing of demand. For example, reservation systems are demand-influencing methods that inventory people so that they are aligned with the organization’s ability to serve them. In any case, the objective of the organization is to fulfill demand in the most cost-beneficial manner possible through prudent mutations in capacity.

The seeming simplicity of the problem has led to myriad complex analysis and possible solutions. Although theoretically optimal solutions exist to these problems, the problems do not necessarily reflect the realities of continuing operations, multiple opportunities to place orders, or incomplete information. Even so, people do not make mathematically optimal decisions and, therefore, the problem has provided a decades-long window into the thought processes of decision makers who make capacity decisions as experimenters manipulate different aspects of the problem. For example, when product profit margins are high, people tend to under-order, and when product profits are low, people tend to over-order. It may be that when profit margins are high, decision makers may see little to be gained (and a lot to lose) with incrementally more purchases and tend to be risk averse. Conversely, when profit margins are low, decision makers may see a lot to gain (and little to lose) with incrementally more purchases and tend to be risk seeking.18 Although people have a theoretical basis for the decisions they make, our general suggestion in determining capacity is more prosaic, yet, sound. Organizations frequently will not know what a correct decision is until it is too late and, given that mistakes are bound to happen, our advice is to choose the right mistake to make. The grid in figure 5.1 shows the types of capacity mistakes that can ensue from high- and low-demand situations, including the mistakes you might prefer to make.

A general rule for ensuring a capable response to fluctuating demand is to have as much variety in the organization as the environment in which its sits (this is known as Ashby’s rule).¹⁹ The more variety in demand in the form of volume, products, and services, the more variety in response capabilities the system should have. In general, a system should have a response repertoire that is consistent with the variety of forces acting upon it.

The business goal of flexible labor is to retain the company’s ability to align human capital with the temporal requirements of the organization, including customer demand, machine running times, and the like. It costs money to be flexible, however. Therefore, flexibility must be introduced in a way that creates value for the organization while preserving employees’ enthusiasm for their work. This is easier said than done since flexibility and employee satisfaction often are at odds with the anesthetizing monotony of repetition and efficiency. A dedicated OD practitioner, however, will insert employee satisfaction into the staffing calculus and balance the interests of employees with management and other stakeholders in the design of work.

Organizations add labor flexibility in two broad ways: in numbers or by functional enhancements, appropriately labeled numeric and functional flexibility, respectively.20 Some typologies make finer discriminations (e.g., differentiate between casual and permanent employees), but these are easily handled by our basic dichotomy. In both instances of numeric and functional flexibility, the goal is to align talent with the business need in a cost-beneficial way to avoid understaffing and impairing performance or overstaffing and adding to costs. Given that goal, sometimes wage–reward flexibility is included in the labor–flexibility bailiwick. Wage flexibility becomes most prominent during recessionary periods when organizations must rapidly adjust to dwindling fortunes. Without making permanent reductions in staff, firms can decrease employees’ salaries or furlough employees. We prefer the latter approach. Planned and phased furloughs can be made in a way that maintains sufficient employee counts without visiting undue hardship on employees. In this case, pay is reduced but so are work hours. In contrast, lowering wages may be better than the alternative, but it requires people to do the same work for less. If the change is temporary, we see no likely ill effects. If the practice persists, however, employees will begin to view the practice as exploitative, particularly if executive pay plans keep executives whole.

Regarding numeric flexibility, the numbers of people (full-time equivalent, FTE) can be adjusted by adding time using the existing workforce or by adding people. Adding time frequently involves working hours past the normal thirty-five to forty hours per week. For many employers around the world, this can be an expensive option because eligible employees are entitled to extra pay over their hourly rate for working beyond a set number of hours. In the United States, overtime is defined as hours worked in excess of forty hours during seven consecutive twenty-four-hour periods. Those who meet the legal standards for overtime pay are entitled to a minimum of 50 percent above their computed hourly wage. Companies can pay overtime to anyone, but they must pay overtime to workers who are not exempt from the labor regulations. Thus, overtime is an easy but potentially costly means of introducing flexibility into the workplace.

Overtime has additional costs other than direct monetary ones. The dangers of persistent overtime among workers—whether paid or unpaid—are well known. Excessively long hours have negative consequences on health, family, well-being, and leisure. These hours do not have to be as onerous as those worked in Japan (see box 5.2) to be physically and psychologically debilitating.²¹ Although no magical number of hours exists that delineates the line between “too many” and “just right,” our bodies know the difference. Those who work lengthy hours without respite are susceptible to burnout. Burnout is a condition typified by exhaustion and physical distress. Common symptoms include tension, fatigue, depression, confusion, and anger.²² Additionally, people who experience burnout become more negative toward the organization and find it increasingly difficult to perform effectively, with every objective a tiring, uphill climb.

Box 5.2
Death by Overwork

Kenichi Uchino dropped dead at 4:00 am while at work. He was a quality control manager who worked on the production line but also spent many extra hours attending to various duties, routinely working eighty-hour weeks. Mr. Uchino was thirty years old when he died.

Those who support change in the number of hours people work in Japan place the annual death rate due to Karoshi (death by overwork) at ten thousand per year. This figure does not count fatalities due to Karojisatsu (suicide due to overwork).

Although the literature shows that prolonged workdays in succession lead to exhaustion and health risks, the relationship does not always obtain. Specifically, if people like their work, are appropriately rewarded, and have some control over their schedules, the effects of burnout are attenuated. In contrast, when the job demands are strenuous—for example, people are working under severe deadlines, concentration is intense and unrelenting, the work is laborious, and the overtime is unexpectedly mandated—the added hours can take a severe toll on individuals’ longer-term ability to perform. The problems of mandatory overtime, for example, are well known in nursing with well-established relationships between overtime and nurse fatigue, practice errors, injuries, adverse patient outcomes, and increased absenteeism and turnover.²³ Although several states are taking actions to restrict mandatory overtime in health care, these results illustrate the general point that poorly planned overtime may complicate the problem the organization is attempting to solve by increasing absenteeism and turnover. As the work time becomes more burdensome and absences and turnover increase, the need for more time ensues until the vicious cycle is broken. This downward spiral highlights the fact that flexibility becomes increasingly difficult to achieve if the workforce is unstable. Therefore, one productive way for a company to become more responsive to environmental instability and uncertainty is to keep absenteeism and turnover at low levels.

An extension of the workday also can be achieved through on-call work. One way to think about on-call work is as conditional overtime: work that is conditional on the need for the employee’s services arising after regular hours. Substitute teachers are the prototype of on-call workers, but physicians, IT specialists, midwives, utility workers, and marine and airline pilots also are commonly on-call. People on-call may receive extra compensation, often in the form of a stipend or flat amount. It may look like companies are potentially paying something for nothing, but the human costs of on-call duties are substantial, including sleep disturbances, interference with personal lives, mental health problems, and cardiovascular and gastrointestinal dysfunction.24 Indeed, many retailers have discontinued the practice altogether, admittedly in response to legal probes that have accused businesses of abusing the practice.²⁵ These probes followed studies that showed the stranglehold of on-call work on people’s lives when the on-call time is lengthy and calls into service are last minute.²⁶ The law, however, may have saved retailers from an awful business practice as a massive study recently found that last-minute calls reduce productivity in restaurants by more than 4 percent.²⁷

Unless an urgent need can be handled remotely, on-call work confines the location of off-duty workers to where they must remain available if called upon. These employees also must restrict their activities to those that will not disrupt their ability to work. Airline pilots on-call, for example, should not drink. The restrictions placed on employees’ personal lives coupled with the low-grade anxiety of never knowing when one will be called into work are stressful conditions. Therefore, we advocate the use of on-call time be limited to “sporadic” or “occasional” (e.g., place employees on a rotational on-call roster).

If work routinely extends past normal working hours, one alternative to overtime is the use of part-time employees. In contrast to what is popularly believed, most people work part-time voluntarily.28 A small percentage of people work part time who are available for and want full-time work. Basically, most people who work part time have other obligations they must meet, which makes working a full-time schedule unfeasible—such as tending to an aging parent. At least for those whose part-time work fits their lifestyle, the work is seen as satisfying while, at the same time, adding allocative efficiencies to the workplace. Recent studies with pharmacies and retailers showed that a mix of part-time and full-time employees increases profits—up to a point.²⁹ That is, profits rise with increased blends of part-time to full-time work until an optimal level is reached before profits plateau and then decline. One explanation for the fall-off in efficiency, productivity, and profits as proportions of part-time employees to full-time employees increase concerns a net decline in the workforce’s experience and skill—and in the case of temporary part-time help, such as employees hired during holidays, a potential lack of cumulative commitment to the organization. The lower hourly pay of part-time workers partially reflects this lesser experienced and trained faction (this pay gap between full- and part-time employees is called the “productivity penalty”). These ostensive deficits in the capabilities of part-timers partially can be fixed (part-timers still will not have the same range of on-the-job experiences) by including part-time employees in the full suite of developmental programs that companies offer. Any cost savings derived by excluding part-timers from these programs is offset by declines in employees’ productivity and devotion to their work—a feeling that can be further eroded by the common omission of these employees from companies’ lucrative benefits programs.

One possible alternative to part-time work is the use of flextime work schedules. Like part-time work, flexible scheduling is a way to fit more hours into a nonstandard workday and, with well-coordinated timing, affords corporations more coverage during periods of heightened demand. Flextime sets the boundaries of the starts and stops of typical days and allows employees to select blocks of time within that framework to work. The day can be scheduled continuously or discontinuously—for example, a person could work early mornings, skip midday, and return to work late afternoons. With appropriate planning, the day can be lengthened through the aggregate time selections of employees and overlaps in personnel can be scheduled during periods when the work needs are greatest. In this regard, flextime is one way to control overtime costs because everyone is working standard, yet overlapping, hours.

Longitudinal studies find that the implementation of flextime increases productivity, particularly when implemented with a focus on aiding employee conflicts with work and home versus for purely instrumental reasons to the company—that is, when there is a mutual benefit to employers and employees.30 Nonetheless, the positive effects of flextime can be thwarted by cavalier schedule changes that are at cross-purposes with the aims of the program. To be effective, these programs require some give and take among management and staff. For flextime to obtain its almost-too-good-to-be-true benefits, there must be an ample supply of group civility, communication, openness, and trust so that people can work out schedules in ways in which they feel that the benefits and burdens of the arrangement are being distributed fairly over time.

In general, employees like flexible labor arrangements because of their ability to arrange times that fit their work and personal lives. Indeed, an econometric study indicated the flexibility of working from home is worth about 8 percent of a person’s salary.³¹ The flexibility of tele-work and the kindred concept of ROWE (Results Only Work Environment) allows employees to work when and where (if possible) they want as long as they meet individual and team objectives. Given the exploding use of telecommuting in the Covid-era, however, the practice has become controversial with both ardent detractors as well as enthusiasts. As with many differences of opinion, the benefits of tele-work appear to be realized when there is a modicum of use, rather than at the extremes. The data clearly show that in the right doses, telework is effective. Employees with greater latitude to attend to work and family duties have greater job satisfaction, lower stress, better well-being, and higher morale.³² Employers also benefit from installing flexible work practices through greater employee engagement and lower absenteeism and turnover.

The right amount of time of working from home appears to be 15–20 hours per week.³³ Too little flexibility and the benefits are lost. Too much time at home creates too much social and professional isolation, derails careers, and contaminates the boundaries between work and home, increasing work-life conflict.³⁴ Thus, exclusively working from home does not work. Indeed, studies show that employees who work exclusively from home work longer hours (by, in part, repurposing former commuting times), but that extra time does not produce gains in productivity. Rather, that added time is expended on meetings and after-hour texting and emailing.35 This expanded day may attest to the human inclination to use all the time available: if you have the time, take it, and when you don’t, work smarter and waste less time. Thus, a five-year study of 3,000 employees in Iceland revealed that a reduction of 4 to 5 hours per week had no effect on productivity.³⁶ Wherever and whenever one works, then, one can appreciate the Economist’s long-time admonition in Bartleby’s Law that 80 percent of meetings are a waste of time for 80 percent of the people in attendance.³⁷ These ravenous gatherings may be a hard habit to break, however, especially for managers who do not know what to do when they cannot see their direct reports who are working from home. Their insecurities and lack of felt purpose and control might be enough to fill the calendars of Zoom-weary remote attendees. Therefore, the quality of management is one of a number of factors (along with other factors such as the state of technology) that can facilitate or undermine the benefit of working from home.

When the workload becomes too great to handle through modest reconfigurations of a workday or prudent increases in time or part-time help, companies often have to add numbers by adding shifts of employees. In our global order, sometimes these shifts can occur in various places in the world so that everyone is working a daytime shift regardless of the time of day for the customers. For service providers that operate customer service centers, this worldwide staffing and geographic flexibility offers 24-7 coverage. If additional crew are needed to augment shift workers, companies have two options. They can add facilities and run several day-shifts, or use the same facilities for two or more shifts. The hardy corporate vote goes to the latter alternative.

The extra pay associated with shift work is marginal to the company and readily embedded in the price of the product. A shift differential generally increases employee pay by 7.5 percent to 10 percent per hour (government provisions set these rates for qualifying employees in the United States, but most employers adhere to these rates regardless of eligibility). Nevertheless, companies may incur additional costs in other ways. Shift work results in employee health problems, disruption to their personal lives, and higher risks of on-the-job accidents. For the roughly 20 percent of the people who work outside of standard daytime hours, shifts aggravate sleep, and are associated with weight gain and higher incidences of type 2 diabetes, coronary heart disease, stroke, and cancer.38

Studies on shift workers have found that some of the negative effects can be reduced. For example, providing longer breaks, introducing robust wellness programs, and maintaining accommodating ambient conditions, such as good ventilation and proper lighting, can be helpful to overall employee wellness. If the employee is on a rotational schedule, a two-day break before the changeover is made can ease employees’ transition into new bodily rhythms.³⁹ Additionally, because the newest and least skilled employees often are assigned the worst shifts, organizations should provide a training bridge to ease their deployment and make sure that more experienced employees are available for assistance in the early stages of their employment to reduce the frequency of accidents.

Companies also can expand their numbers through external sources, such as temporary agencies, independent contractors, or outsourcers. The use of outsiders to supplement staff often is attended by an implicit or explicit assumption about the workforce: that “core” employees have valuable firm-specific skills and are central to the organization’s success, and “peripheral” employees supplement the work of the core through the use of more generic and fungible skillsets.⁴⁰

Temporary help originally served as a shock absorber function with, most often, office workers supplementing the regular permanent staff during upticks in work (see box 5.3).⁴¹ Temporary workers form a cadre of workers known as contingent staff because their work is contingent on the completion of a project or the expiration of a contract. Temporary workers are employed by other host organizations who rent out their employees’ time to organizations on an as-needed basis. Nevertheless, if contractually permitted, these temporary jobs may become permanent internal positions when an organizational need for a full-time slot arises and money is available to fund it. In these cases, organizations will have had the chance to observe the performances of temporary workers and to select the best.

Box 5.3
The Rise of Temporary Agencies: Manpower

Manpower was the first temporary agency. Employment agencies at the time (1948) only supplied permanent workers. When Elmer Winter, a lawyer in private practice who could not type, needed a long brief for the Supreme Court, he and his partner Aaron Scheinfeld called on a former secretary who resigned when she had her first baby. The temporary agency Manpower was born. It struggled to expand services trying to convince the model obtained to any discipline in which there were spikes of activities or seasonality, such as retail at Christmas, florists at Easter, or accounting during tax season. Employers at the time were largely wedded to the idea that employees should be permanent, as a moral matter. That changed when paper-based systems needed to be digitized. Northwestern Mutual insurance Company had its documents transferred onto IBM punch cards and stored, requiring a borrowed crew of key punch operators. The industry for temporary help—which expanded to wider use of consultants—was changed.

An organization also may lop off entire pieces of work, such as payroll, janitorial services, food services, benefits administration, call centers, and even parts of the research and development process, and give the associated responsibilities to others to perform in so-called outsourcing arrangements. Outsourcing broadly applies to the attainment of goods and services from outside parties that could be provided internally. When outsourcing work is given to third parties, companies typically discriminate between activities that are central to its money-making capability and those that are peripheral to it, ceding the peripheral to others and keeping the corporate nucleus of work to themselves.

At times, companies require work from independent contractors that is central to the organization’s mission but that constitute skills that companies do not need in the long term or that they will acquire through the transfer of knowledge from the contractors. This arrangement can get a little complicated in two ways, especially for longer-term projects. First, many countries have strict legal definitions for independent contractors. These laws prohibit employers from treating independent contractors like employees. Independent contractors are expected to have multiple client relationships, have their own tools of trade, have other customers, and regulate their own time with the project aims and deadlines in mind. This type of agreement offers advantages and disadvantages. The advantage is that the laws protect independent contractors from abusive practices of organizations that confine and direct workers like employees but that do not pay them benefits. Conversely, well-intentioned employers have run afoul of the rules by trying to incorporate independent contractors too completely into the company’s operations and treating people the way one would imagine people would like to be treated—as appreciated allies of the employer’s mission. Good intentions aside, to avoid the long arm of the law, the relationship between the company and the contractors must remain at arm’s length. The temporal impermanence of the work and the interpersonal distance in its performance mean that organizations can be convivial and accommodating to contractors, but not to the extent and favor that they treat everyone else.

This leads to a second matter. Although independent contractors must be accorded a certain degree of independence, this does mean that they cannot be managed. They were hired for a purpose and, consequently, to receive the full benefits of contractors (at the lower hiring, training, and replacement costs than full-time staff), the relationship with vendors must be competently supervised. The company needs to assign a person to manage the vendor relationship who is at a comparable skill level as the contractors, understands the outcomes to be achieved, and can push the contractors to accomplish things they otherwise might have thought impossible. Furthermore, the project overseer needs to be adept at team-building, be business savvy with the keen ability to understand and enforce contracts, and possess good interpersonal, communication, influence, and problem-solving skills.42

The main way to add functionality to a given workforce is to cross-train, or add skills to current members of the workforce. The training frequently is focused on technical know-how, procedural improvements, and different aspects of the operations. Often, too, training is concentrated on equipment used and complementary tasks like computer skills, set-up skills, minor repairs, or basic troubleshooting. These latter skills are critical because they can lower switchover costs: the time and effort it takes for one person to step into a job vacated by another. For example, when you are at the supermarket and you are in line when a cashier change is scheduled, you want the new cashier to plug into the system quickly to get the line moving again. For that to occur, the cashier not only must know how to operate a cash register but also how to establish themselves as a new primary user; efficiency will be further heightened if that same cashier can conduct basic fixes like replacing an uncooperative roll of tape.

Cross-training is closely aligned with process redundancy but it is not quite the same. Redundancy is a way to ensure the robustness, or resilience, of a system: to maintain the functionality of the process despite changes in the environment.43 Thus, running duplicate systems when converting enterprise technologies or having available backup (e.g., when Elyse is out, William can do it) are ways to ensure that processes remain immune from operational disturbances. Thus, redundancy usually concerns some form of replication across people or systems. Cross-training pertains to the multifunctionality of an individual or to the number of tasks a given worker can perform. The more a person can do, the more flexible they are, and the more options an organization has for deployments. A recent study on utility players in baseball illustrate the point.⁴⁴ A capable starting player who is skilled at two or more positions allows managers of those teams to make more flexible lineup selections and provide more options if a player, say, is injured and players need to be moved around. Researchers calculate that this flexibility adds a couple of wins to a team’s record per season.

The illustrations in figure 5.2 show different degrees of cross-training.⁴⁵ The illustration in figure 5.2(a) is the least flexible arrangement, all else being equal. A certain task must be performed by a given individual. Still, even in these restrictive circumstances, organizations have flexible options. First, the burgeoning, often frustrating, use of automation can be used to scale back demand or redirect traffic to more suitable providers. This same automation permits routing flexibility so that if Person A is unavailable for a given task, the activity can get moved to Person X who is trained to perform the same task. If, however, certain tasks can be performed only by certain people (e.g., provide a signature), then it helps if the tasks in figure 5.2(a) can be batched (several forms needed for signature are assembled and presented at the same time to the signatory every week), or if the order of operations as depicted in figure 5.2(b) can be changed (e.g., currently, Person A waits until accounting approves a position for hire before Person B posts the position; however, the logical sequence does not imply that the same operational sequence must be followed—Person A can create the requisition and have it go live once formal approval is given). Processes are structured the way they are for many reasons, but processes in which the order of operations can be changed will be more flexible. And so, too, will processes such as those rudimentarily exhibited in figure 5.2(b). This shows that a given person can accommodate more than one throughput. If Person A can help the caller with say, auto as well as homeowner’s insurance, the more flexible the process.

The illustration in figure 5.2(b) shows blocks of people who can respond to different demands. This is a realistic depiction of cross-training in which a person can perform two discrete tasks. However, as the number of learned increases, productivity declines. These diminishing returns are because of increased training times, deficits due to a lack of practice, and forgetting tasks. Thus, productivity erodes as people take on too much.⁴⁶ In many instances, organizations use a variation of what is shown in figure 5.2(b) that builds career progression into the process formulation. Thus, figure 5.3(c) depicts a step system of nested skill sets. People are progressively given greater responsibility and move into higher ranking positions as they acquire higher order skills after a certain number of contact hours and training.

Job rotation often is used in conjunction with cross-training. Technically, job rotations have been defined as lateral transfers that do not affect a person’s place in the organizational hierarchy or their salary grade.⁴⁷ Although job rotations mostly are lateral moves, they do not have to be. They can be up, down, or sideways. Job rotation is implemented for different reasons.⁴⁸ The most straightforward and customary reason is to foster employee development. At the same time, it gives organizations a chance to see how employees perform in new, challenging situations. The ease and effectiveness of job rotation depends on the nature of the tasks and the learning or overlap and transferability of skills. Job rotation is a favorite practice of lean manufacturing because operators or technicians acquire a better overview of the production process, which facilitates problem solving and increases flexibility in response to labor shortages and fluctuations in demand.

Despite the best of plans, people have scheduling conflicts and external events can unpredictably impose pressures on organizations. Organizations must respond in ways that the best of plans cannot accommodate. In these instances, staff improvisation, or schedule patching, comes in handy.49 Given that the management of schedules often leaves little window between planning and the execution of changes, solutions depend on actions outside of the formal plans. In these instances, researchers have found that the best solutions may be for employees to make up solutions as they go along by plugging scheduling holes opened by personal or work-related needs.⁵⁰ Thus, for irregular occurrences in home- or work-life, the voluntary changes made by the employees themselves in instituting impromptu adjustments is frequently the best, and only, course of action.

The issues in this chapter are significant because the labor costs within some industries consume between 50 percent and 80 percent of operating costs. Because of this, staff-planning should be a central concern and activity of OD practitioners. OD spends a great deal of time on organizational design, but forecasting staff needs has not become a stable quill in the OD arrow. Given the primacy of human resource planning to organizational success, determining an ideal number of workers (internal and external) who have the proper compositions of skills and are correctly assigned to jobs should be within the OD arena.

Large-scale workforce planning can be a mathematically intensive affair that may account for some reluctance of practitioners to wade too far into this area. Some planning models require advanced calculus and linear goal programming, The precision these optimization models afford, however, are rarely necessary for estimating staff requirements. Indeed, pure mathematical models often oversimplify real-life conditions and rarely acknowledge the many uncertainties that can infringe on desired organizational outcomes.51 In this regard, OD shines as practitioners are familiar with the many variables in addition to worker costs that can affect results, such as the timeliness and quality of care or service. These variables include the current organizational structure, the skills and experiences of employees, the availability of the labor pool, turnover, absenteeism, and retirement rates, anticipated leaves of absence, availability and quality of training, regulatory and task-related changes, and more.

These analyses will not answer some questions, such as how much, if any, slack to maintain within a system. That is, should companies maintain a surplus of human resources? Like having excess financial resources, having slack in personnel has been associated with an ability to exploit competitive opportunities, build new capabilities, and innovate, as well as to buffer companies against unexpected threats and disruptions to operations.⁵² Our experience on this matter corresponds to the results of a recent study.⁵³ As might be expected, when organizations are small or in the start-up phase, slack is a luxury ill-afforded. The strategy is set and there is no incentive to experiment. As organizations grow and the numbers of companies’ offerings and potential competitors multiply, having slack is advantageous to those companies that have clear, sound long-term objectives and understand the best ways to use their surplus capital. Unfortunately, no definite line exists that demarks where the value of slack ends. Although some slack seems beneficial to organizational development and innovation, there comes a point when too much slack seems to produce idle hands.

Mechanistic	Organic
Centralized control	Decentralized authority
High task specialization	Multifunctional employees
High standardization	Some flexibility/adjustments
Vertical communications/Reporting paths	Network/lateral communications
Tight supervision	Participative management
Rigid rules and procedures	Personal discretion
High formality	Informal exchanges

Pathological	Generative
Low cooperation	High cooperation
Messenger shot	Messengers trained
Responsibilities shirked	Risks are shared
Bridging discouraged	Bridging encouraged
Failure leads to Scapegoating	Failure leads to inquiry
Novelty crushed	Novelty implemented