Chapter 3 Workflow, Job Analysis, and Job Design

Workflow Analysis

Workflow is the way that people get work done. Every organization has at least one workflow (or business process). Often workflow entails a series of steps through which a piece of work passes, from initiation to completion. Workflows are often stable and broadly understood as “how we do things here.” A workflow map is a visual representation of this process. Workflows can be examined at different levels of granularity. Feature Box 3.1 provides a simple example of workflow at a family restaurant, expressed both as a narrative and as a map.

Organizations map workflows for several reasons. A clear workflow map is very useful for identifying the kinds of jobs that must be performed. For example, the stratified restaurant workflow in Figure 3.3 immediately tells us that three jobs must be done (greeter/cashier, server/table busser, and cook). Workflow can also be a useful training tool for new staff (e.g., “this is how work is done, and here’s how your job fits in”) and help supervisors and HR practitioners to determine how to assess worker performance. In the restaurant example, an employer might decide to monitor employee performance by tracking cash register shortfalls, customer complaints, and kitchen errors (performance standards are discussed in depth in Chapter 9).

Workflow mapping also makes explicit how work gets done. Making the process of work visible helps employers to exert greater control over the workplace, resulting in both reduced worker power and lower labour costs. For example, in the early part of the 20th century, Frederick Taylor pioneered the use of time-and-motion studies to determine the steps needed to produce items. He observed and measured where workers moved, how long it took them to perform a task, and in what order they performed them. Using this information, he redesigned the workflow to break down production into a series of discrete, sequenced steps. This process, today called Taylorism (or sometimes scientific management), captured workers’ knowledge and made it available to employers.

Scientific management also reduced the ability of workers to use their knowledge to control the pace of production. Once employers understood the workflow, they could rearrange it to their advantage. For example, knowing how long it takes to do each step allowed them to set and enforce production targets. Knowing the process of production also allowed them to resequence and simplify the steps in production and to hire less-skilled workers to complete some tasks (thereby lowering labour costs). Today’s highly regimented manufacturing jobs (in which a worker does a single task over and over) are the outcome of this workflow mapping and job (re)design. This has also spilled into service industry jobs, most obviously fast-food restaurants.

This historical example reveals that workflow analysis is not an entirely technical activity. Employers and HR practitioners make decisions (if implicitly) during workflow analysis about what sort of workforce they will have and the conditions that the workers will face. For example, Taylorism allowed employers to hire fewer skilled workers (at lower wages). The introduction of a moving assembly line by Henry Ford allowed employers to increase the pace of repetitive jobs by speeding up the line, in turn increasing the rate of injuries. Workflow decisions shape subsequent HR decisions, such as which KSAs must be sought during recruitment and what workers will be paid. They also affect the quality of workers’ lives. As shown in the opening vignette, all of these techniques and effects are evident in the implementation of scanners in grocery stores.

Although almost all organizations have established workflows, they are not always mapped or mapped accurately. Often the workflow exists only in the heads of those involved. Sometimes inaccurate or unmapped workflows reflect that the organization has simply not bothered to codify or maintain its workflow. For example, inaccurate written workflows can result from a long-term, incremental drift in processes, such as gradual changes in technology that make some steps redundant. Inaccurate workflows can also result from sudden, large changes that are not codified, such as jerry-rigged processes created in response to the COVID-19 pandemic. Inaccuracies can also be caused by employees who adjust processes on their own, such as deciding to skip steps to manage workload increases, change processes to make a job more enjoyable, or circumvent surveillance, such as by communicating via texts instead of through company email systems.

In other cases, the HR strategy adopted by an employer or the nature of the work can impede workflow analysis. For example, in a high-performance work system, jobs or tasks in which workers can (or must) exercise significant discretion can be challenging to map in a meaningful way. Consider the process of writing this textbook, a regular task for a university professor. Figure 3.4 maps this workflow at a high level.

It is not really possible to break down each high-level step further to create a step-by-step process that someone without our KSAs could easily follow. This reflects that the overall process is highly iterative. That is, we moved back and forth among steps 2 and 4 right up to the minute that the final manuscript was sent to the publisher. And, within each step, the exact process that we used was idiosyncratic, and if mapped it would appear to be chaotic. For example, portions of chapters were written piecemeal, often months apart, out of order, and by different authors. Ideas, examples, and concepts were continuously added, cut, altered, and moved around in response to our own evolving understanding of what worked, comments from reviewers, and random thoughts and ideas that we had or encountered. The appearance of a chaotic process, however, is misleading. We have decades of experience with this book’s content and teaching it to others. This expertise means that we can manage the complicated and demanding process of writing and revising a huge book (containing many interrelated parts) without losing track of either the details or the final goal. (A good editor also helps!)

Typically, workflow analysis entails performing three steps that can be broken into seven tasks. This process is set out in Table 3.1. Workflow analysis begins with answering the question of why it is necessary to analyze workflow. The purpose of the analysis will shape the focus and degree of granularity in the map. For example, mapping the workflow in a restaurant (e.g., Feature Box 3.1) to train new workers to do the job will likely require less detail than would mapping workflow to ensure that workplace hazards are identified and controlled. Consequently, the first step in a workflow analysis is to determine the reason(s) for it. This might require discussion among key stakeholders to ensure that this purpose is shared and understood.

The reason for analyzing workflow usually determines which workflow is analyzed, and more than one might be analyzed. For example, a quality control problem or a desire to contract out work (which requires a clear understanding of what is currently happening) can trigger a workflow analysis. Absent a compelling issue to trigger mapping, organizations tend to focus their resources on mapping only operationally essential business processes. This is because errors in these processes entail the greatest potential organizational risk.

Workflow requires information sources and a clear process for collecting the information. Many of the data collection techniques described below for job analysis (e.g., observation, interviews, questionnaires) can be used to gather data for a workflow analysis. The process of collecting and analyzing workflow data, however, tends to be a bit more iterative and visual than is typical of a job analysis.

Creating a new workflow map often begins with developing an initial sketch of the workflow by an HR practitioner. This process can be as simple as writing steps on sticky notes and then putting them in a sensible sequence, or it can involve using process-mapping software. This initial workflow can then be presented to a group of informants (e.g., workers, supervisors, clients) for review. This review can identify previously unrecognized tasks or sequences as well as interdependencies, contingencies, and feedback loops. If there is an existing workflow map, then it can be subjected to a similar process of reviewing and modifying the workflow. Eventually, the group will agree on a workflow map.

That map can be verified by observing the work process and comparing the actual practice to the map to identify differences or other errors. A map can also be refined by identifying which job is responsible for each task (see Figure 3.3) or what the inputs to the step are, how they are transformed, and what the outputs of the step are. For example, if an HR practitioner is mapping the process by which a university admits students to a degree program, then the process will include a step in which someone (likely a clerk in the registrar’s office) actually decides whether each applicant should be admitted or rejected. The inputs to this decision are student applications and admission criteria. For the sake of simplicity, we can assume that the criteria include a simple grade cutoff, such as 75%, based on five named high school courses. With this criterion applied, the applications are then transformed into two outputs (accepted or rejected applications). These outputs then move into separate processes at the next stage of the workflow. Identifying inputs, processes of transformation, and outputs can make clear how this workflow relates to other organizational processes.

At some point, a reasonably accurate workflow map will be produced. If the workflow is complicated, then it might take several rounds of revision and validation for an HR practitioner to be confident that the map is accurate. At this point, the map can be used to inform other HR activities, such as job analysis or (re)design. For example, a workflow map might identify that five jobs are related to a particular task. The workflow analysis might identify that this number of handoffs results in errors (e.g., incomplete tasks or dropped handoffs). Redesigning the workflow to reduce the number of handoffs might be desirable. This can entail redesigning jobs (so that one person completes the entire task) or applying technology to prevent and/or detect errors in handoffs.

Job Analysis

Job analysis is the process by which an HR practitioner determines a job’s duties, responsibilities, and working conditions as well as the requisite KSAs of workers in the job. The results of job analyses inform other HR tasks. For example, job analysis typically informs job descriptions and specifications used to help recruit and select the right worker for the job. Job analyses can also help to identify training needs among current staff, the best method and rate of compensation, and which standards should be used to assess worker performance.

As stated above, a job comprises a group of related activities and duties. When an organization has several distinct jobs with similar duties and responsibilities, those jobs can be assigned to a job family. They might have similar KSAs and training needs, have similar compensation structures, and provide pathways for career advancement. For example, a car dealership might have three job families: sales, administration, and servicing. The specific jobs in each job family are set out in Table 3.2. As you review these job families, assess whether they make sense from the perspective of the KSAs that they likely require and the potential job advancement within a job family. You might also wish to consider the best way to compensate work in each job family, such as commission, salary, or piecework.

The job analysis process has three main phases: preparation, data collection, and data analysis and use. Within each phase are specific steps as set out in Table 3.3. The first (preparatory) step is to determine the purpose(s) of the job analysis. For example, is the purpose simply to document the status quo in order to create accurate job descriptions and job specifications? Or is the purpose to identify areas where a job could be redesigned to streamline a work process? Understanding the purpose of the job analysis ensures that the right information is collected.

The purpose of the job analysis will also shape which jobs will be analyzed. Job analysis can entail a significant investment of organizational resources, particularly staff time. Using these resources to perform job analysis means that they cannot be used for another purpose. The forgone benefit that would have been derived from using these organizational resources for some other purpose is called the opportunity cost. The opportunity cost of performing job analysis suggests that HR practitioners should focus job analysis where there is a clear organizational need. Areas of need might include the following.

• • Jobs that have changed because of environmental or technological factors can benefit from analysis to ensure that job descriptions, specifications, and compensation remain correct.
• • Jobs that are difficult to fill or have high turnover might have structural problems such as requiring an uncommon set of KSAs, being too difficult to perform, or being undercompensated.
• • Jobs that are key to organizational success because of the tasks that they perform; a workflow analysis that breaks down workflow by job (e.g., Figure 3.3) can be helpful in identifying jobs key to organizational success.
• • Jobs that pose organizational risk; for example, if a government requires contractors to demonstrate gender parity in their workforces, and a particular job family is male dominated, then a job analysis can suggest ways to address gender parity and attenuate the risk of non-compliance.

Once a job has been selected for analysis, an HR practitioner can identify the available sources of information about the job. Such information can be derived either from people or from documents. Table 3.4 summarizes the main sources of information about jobs: either people or documents. Feature Box 3.2 identifies that the potential for workers to provide inaccurate information exists, suggesting in turn that it is sometimes necessary to use multiple and independent sources of information. Designing the data collection process entails selecting which data sources to use, determining how best to get the data from each source, and deciding how the information will be recorded.

Selecting which data source to use entails trade-offs. For example, an HR practitioner might get a thorough job analysis if three workers, a supervisor, and three customers are interviewed and then a worker is observed for a week. But the opportunity cost of this approach would be high. The practitioner might get almost the same information by interviewing one worker and one supervisor and reviewing an existing job description or training manual. Whether the extra accuracy of the first approach is worth the additional cost is context dependent. It might be worth spending more organizational resources if there is a significant organizational risk attached to error in the job analysis. The practice of HRM is rife with these sorts of cost-benefit decisions. These trade-offs help to explain the difference between prescription (what a textbook says should happen) and description (what actually happens).

Data collected during a job analysis is typically recorded on a job analysis form (sometimes called a job analysis questionnaire). Forms ensure consistent data collection. A form typically includes the following information.

• • Job identification: This section sets out which job is being analyzed (title, organizational unit), when the analysis took place, and who was involved.
• • Purpose, duties, and responsibilities: This section identifies the major purpose(s) of the job as well as the major duties (and perhaps the frequency and importance of each). This section determines the job’s key responsibilities (e.g., equipment operation, supervision of others) and whether they are major or minor responsibilities.
• • KSAs and working conditions: This section identifies specific knowledge, skills, and abilities required by the job. It might also identify notable working conditions, hazards, or demands of the job.
• • Performance standards: This section sets out how performance or success is measured.

Feature Box 3.3 provides an example of a job analysis form. The exact content of a form will vary among organizations and depending on the purpose of the job analysis.

When documentary evidence is used, typically an HR practitioner reads the documents, extracts the information relevant to the job analysis, and enters it on the form. It can be useful to indicate the source of the information in case it is ever necessary to go back to the original source.

There are four main ways to collect data from people (e.g., workers, supervisors, customers).

• • Questionnaire: A written or electronic questionnaire is an inexpensive way to collect data. The main risk of questionnaires is that respondents might misunderstand written questions (even carefully written ones). It can also be necessary to follow up on questionnaires to generate or clarify responses. It is often useful to supplement questionnaires with interviews.
• • Interviews: Asking questions of knowledgeable sources, either individually or in a group, can elicit useful information. Semi-structured interviews use a questionnaire to guide them but also give the job analyst latitude to explore unclear or unexpected answers. Separately interviewing people allows an interviewer to cross-check facts.
• • Diary: Workers can be asked to track their activities using a diary (sometimes called a work log). Typically, they record their activities at regular intervals (e.g., every 20 minutes) over a long enough period (e.g., two weeks) to provide a representative picture of their jobs.
• • Observation: Job analysts might be able to observe or shadow a worker. Doing so allows them to record which job duties and activities are performed and how frequently. Such observation can also reveal some of the KSAs that a worker must use on the job. Observation can be time consuming. It can also trigger an observer effect, by which the presence of an observer can affect the behaviour that the subject exhibits. This can undermine the validity of the information gathered by the observer.

Once data for a job analysis have been collected, they can be analyzed. Typically, this entails creating a job description and a job specification. A job description is a written statement of the duties of a job. It typically contains a job title, a job identification section, and a list of job duties. It can also contain information about the job’s compensation. A job description can also identify important working conditions as well as contain a job specification. A job specification is the list of qualifications required to perform the job. Feature Box 3.4 contains a sample job description for a construction labourer on a road crew.

There are three main audiences for job descriptions.

• • Workers and potential workers can know that the job entails semi-skilled manual work using hand tools and small machines in the outdoors. The work also requires long shifts, unusual shift schedules, and extended time away from home. This description allows potential workers to decide if they are qualified for and interested in the job.
• • Supervisors can use the job specification to make hiring decisions and the job duties to assign tasks to labourers consistent with the skills and terms of their employment contracts. Supervisors can also use the job description as the basis for assessing worker performance (see Chapter 9).
• • HR practitioners can use the title and job specification to write job advertisements, classify the job, assign compensation, and perform other HR tasks. HR staff can also ensure that the requirement for workers to pass periodic worksite drug and alcohol testing is performed within the boundaries of the law (testing is further examined in Chapters 5 and 8).

In developing a job description and a job specification, it is important to be mindful of three things. First, some aspects of jobs are more readily apparent than other aspects. For example, physical demands and activities (e.g., lifting) are often easier for job analysts to “see” than mental demands and activities (e.g., navigating office politics). Activities not readily apparent can sometimes be left out of job descriptions and specifications. This can result in workers being undercompensated and supervisors failing to hire workers with the necessary skills to do the job. Having a draft of a job description verified by workers and supervisors can identify these sorts of gaps.

Second, job descriptions and specifications are social constructions in that analysts must decide what to include and what to exclude (i.e., which KSAs, activities, and duties are important and which are not). This winnowing process can be a source of error. For example, an analyst might recognize lifting as a requirement for a building maintenance job but fail to include it as a requirement for a child-care worker. Given the gendered nature of these jobs, failing to recognize lifting as a routine aspect of child care renders such tasks and abilities organizationally invisible. Consequently, workers can be hired who are unable to do the job, supervisors might not provide workers with the equipment necessary to control the hazard posed by lifting, and organizations might be reluctant to acknowledge injuries caused by lifting. Again, knowledgeable reviewers can identify errors and omissions. Feature Box 3.5 explores the related issue of gendered emotional labour frequently overlooked in job analysis.

Third, job analysis often has difficulty grappling with activities that a worker is not formally required to perform but that are practically important to the organization’s operation, such as observing social rituals (e.g., acknowledging birthdays or the births of children) or undertaking relatively thankless organizational tasks (e.g., sitting on committees or informally mentoring new hires). These tasks can enhance the functioning of a group of workers and are often performed by women. But these tasks are rarely considered “work” and tend to be ignored, trivialized, and undervalued in job analysis. This pattern suggests that organizations take advantage of women, who traditionally have been taught a moral outlook that emphasizes solidarity, community, and caring, by expecting them to do this work for free.6

Job Design

As noted above, job design is the process of structuring work and assigning specific tasks and duties to jobs in order to achieve an organization’s objectives. In established organizations, job design is typically about redesigning jobs (or even entire workflows) to improve efficiency and/or workers’ experiences of work. Both outcomes are expected to lead to increased productivity. Increased efficiency means that more products or services can be produced with the same inputs. Increasing worker satisfaction is expected to result in greater productivity.

The notion that increasing job satisfaction increases worker productivity is based on the job characteristics model. That model was originally proposed in 1976 by Richard Hackman and Greg Oldham and asserts that improving job characteristics positively alters workers’ psychological states. That, in turn, results in better psychological and performance outcomes. The basic logic of the job characteristics model is intuitively attractive: better jobs make happier workers likely to work harder and more effectively. But is the underlying model correct? Feature Box 3.6 explores this model and its validity.

The process of job (re)design typically seeks to meet four objectives:

• • achieving the organization’s goals to which the job contributes;
• • increasing workers’ satisfaction through manipulating job characteristics;
• • maximizing efficiency through industrial engineering; and
• • minimizing the negative impacts of work on workers through ergonomics.

In practice, job (re)design can take many forms. The steps in a workflow can be altered, and work tasks can be assigned to different jobs. Frederick Taylor’s time-and-motion studies of factory work increased efficiency by making tasks simpler and more repetitive. Redesign can also include introducing new technology that can increase the productivity of workers or replace them entirely. For example, the introduction of machines and robots (e.g., the self-checkouts discussed at the beginning of this chapter) has increased the efficiency of (or replaced) human workers. Feature Box 3.7 examines an example of a job redesign involving the implementation of a call centre to manage customer requests and reduce wage costs.

Job redesign can also entail physical alterations to the workplace. For example, an employer might replace individual offices with smaller cubicles to reduce the office space required. Or the employer might direct workers to work from home, thereby offloading office costs onto workers. Finally, job design can take the form of outsourcing some aspects of the work. For example, many employers have outsourced the HR functions of payroll and benefits administration to specialist firms.

There are several common ways that organizations attempt to redesign jobs to increase productivity and/or worker satisfaction.

• • Flexible scheduling: Some organizations might be able to provide workers with flexibility in when they work. This can increase satisfaction by allowing the workers to manage competing demands. Organizationally speaking, though, such flexibility can complicate scheduling, supervision, and payroll.
• • Teleworking: Allowing workers to work from home (or another location) can improve their satisfaction by reducing commuting time and allowing them to manage competing demands better. The benefits of teleworking are gendered. Caregivers, mostly women, can experience more stress, less satisfaction, and less productivity when working from home.16 Teleworking can also shift costs associated with providing workspaces from the organization to the worker. It can add complexity to HR tasks such as performance management, health and safety, and information privacy.
• • Job rotation: Moving workers among jobs adds variety and can increase their skills. This can reduce, in turn, the impact of worker turnover because several workers will be able to perform a single job. Job rotation can increase costs given additional training requirements, and it does not fundamentally alter the core job dimensions identified in the job characteristics model. Workers can also resist cycling through jobs in the rotation that entail undesirable tasks or negatively affect their home lives (e.g., shift changes).
• • Job sharing: Allowing two workers to fill a position part of the time can increase their satisfaction by allowing them to meet other demands in their lives. Job sharing can also reduce worker turnover because an employer can ask the remaining worker to cover temporarily a vacated position. Two workers, though, can complicate workflows because they must hand tasks back and forth and will slightly increase compensation costs, particularly for benefits.
• • Job enlargement: Job enlargement entails adding tasks to a job of approximately the same complexity. This is sometimes called horizontal loading. It increases skill variety and can increase task identity. Job enlargement can also require additional training and increase compensation costs because workers must be more skilled. They might resist job enlargement if it affects the wage-effort bargain. For example, for workers who have short breaks between tasks, job enlargement can fill that time with other work. That would take away time that workers would otherwise use to recover or relax. Constantly shifting between tasks can also degrade workers’ performance.
• • Job enrichment: Increasing the autonomy of workers in how they do their work is called job enrichment (or vertical loading). It can increase skill variety, task significance, autonomy, and feedback. Thinking back to the high-performance work system (HPWS) discussed in Chapter 1, we can see that the creation of workers’ teams with expanded authority is another form of job enrichment. Again, training and compensation costs can rise, workers might resist this “opportunity” to work harder, and they might be reluctant to share their KSAs with peers because they view their knowledge as a source of power.
• • Employee empowerment: Although job enlargement and enrichment are supervisor-controlled changes to jobs, employee empowerment allows workers to direct changes in their jobs. This approach is common in HPWS and reflects that workers can be best positioned to make changes in their jobs and workflows because of their knowledge. This approach shifts power from an organization to its workers. Although an organization can always veto changes made by employees, such an action can undercut their willingness to engage.

Many of these approaches to job design seek to increase the satisfaction (and thus performance) of workers by placing increased demands on them. The degree to which increasing demands is associated with greater productivity is questionable. Some research suggests that there is an optimal level of workload, after which workers begin to throttle back their effort per task to cope with higher volumes of work.17 This suggests that the impact of job redesign on overall productivity might be limited.

Other limits to job redesign can include the following.

• • Worker availability: Employers can struggle to find workers willing to accept and capable of performing a job. Employers can influence workers’ willingness to work by altering wages and/or working conditions. Employers have less ability to address absolute shortages of adequately skilled workers caused, for example, by an unexpected boom in an industry absorbing existing skilled workers. The evidence for absolute labour shortages (i.e., there truly are no more workers available) is weak, and additional workers often can be enticed into the labour market by making jobs more attractive.18
• • Contractual or social norms: Employers can face contractual limits on job redesign. The common law might view radical job change as a violation of an employment contract (see Chapters 2 and 9). Provisions in a union’s collective agreement can also limit an employer’s ability to redesign jobs (see Chapter 10). Furthermore, workers might informally resist job redesign through individual or collective actions such as work slowdown, gaming, or turnover (see Feature Box 3.7). Customers can also resist job redesign, as shown in the opening vignette about self-checkouts.
• • Legal requirements: Employers can also face legal limits on job redesign. Redesign that makes it impossible for a worker (or a class of workers) to perform a job based on a protected ground might run afoul of human rights legislation (see Chapter 2). For example, penalizing workers who refuse to be on call because of child-care obligations might be found to be discriminatory (depending on the circumstances).

Conclusion

Workflow analysis, job analysis, and job (re)design are foundational elements of the practice of human resource management. The information provided by job analyses in particular is required for HR practitioners to complete basic HR tasks such as undertaking recruitment and selection, determining wage rates, and organizing training. Workflow analysis also informs HR practices such as human resource planning and performance assessment.

These HR functions require practitioners to have technical skills. But they also require practitioners to exercise political acumen. The requirement for acumen reflects that information about how work is done is a source of power. Specifically, workers might object to an organization knowing exactly how fast they can work or redesigning workflow to increase productivity. This objection reflects that workers (quite reasonably) expect their employers to use that knowledge to change the wage-rate or wage-effort bargain (perhaps through job redesign). This change will almost certainly benefit employers’ interests (i.e., minimizing labour costs to maximize profits), but the degree to which it benefits workers is less clear.

Ironically, exercising political acumen (and perhaps even trading off employer and worker interests) in the performance of HR tasks is often left out of the job descriptions and specifications of HR practitioners! This highlights that the contents of workflow maps, job descriptions, and job specifications are social constructions. That is, they are created by people within an organization. What is recorded in these documents is shaped, at least in part, by the interests, power, and biases of the various contributors to them. Leaving political acumen out of the job description and specification for HR practitioners allows employers to pretend that administering employment relationships is a technical task that warrants a lower level of pay. The consequences of omissions in job descriptions and specifications can affect workers unevenly, depending on their social characteristics.