Job Leveling' Helps to Grade a Position's Value
Inaccurately evaluated jobs lead to pay dissatisfaction and wasted payroll dollars
Job leveling is a systemic method of objectively and accurately assigning value to individual positions within an organization. It is a process that defines and evaluates the knowledge and skills that are necessary to perform the job and establishes the job's duties, responsibilities, tasks and level of authority within the organization's job hierarchyJob leveling (or job classification or grading) is one of the most persistent and stubborn problems in many organizations. The effects of inappropriate leveling often have systemic and lasting implications. For organizations, jobs that are valued too high will result in additional compensation expense. For employees, jobs that are valued too low may be construed as personal affronts and a threat to status. Job leveling is, in fact, one of the most visible and pointed undertakings in the HR quiver. Although jobs are difficult to value, the methods described in this article will provide an objective approach to an otherwise subjective activity. When Jobs Are valued IncorrectlyIt all begins when HR finds itself in the crosshairs of dissatisfied employees who believe their work has been misrepresented, their jobs wrongly evaluated and their worth under-appreciated. In these situations, HR is repeatedly placed on the defensive and in the uncomfortable position of having to prove or disprove elusive facts pertaining to job content. (See the sidebar “Signs that an Organization Has Job Leveling Problems.”) Signs of Job Leveling ProblemsIt is likely time to take job design and leveling seriously when an organization is experiencing any of the following symptoms: • Persistent requests for job reclassifications. In the fuzzy world of job ambiguity, role confusion, and weak internal controls, employees quickly recognize that anything goes, and the spoils go for the most incessant and insistent complainers. • Too many job titles. Job title proliferation is generally associated with amorphous jobs whose responsibilities are specified unclearly. This results in the legally treacherous and organizationally damaging situation of having people who occupy the same role but have different job titles and grades, and, conversely, having people with the same titles and grade assignments who perform different activities. • Employee perceptions of uncompetitive pay. Pervasive feelings of dissatisfaction with pay may be traceable to a number of root causes, however a failure to accurately measure jobs to correctly portray their place in the organizational hierarchy and attribute the right market values to them is at the top of the list. • Redundant work/processes. Errantly measured and misplaced jobs may produce duplicative responsibilities and ambiguous accountabilities that contribute to an environment of distrust, miscommunication and confusion that, ultimately, erodes service and quality. • Financial “leakage.” Although not readily apparent until the problem is recognized and fixed, poorly defined and inaccurately evaluated jobs tend to increase payrolls. Under conditions of uncertainty, companies tend to overpay employees. • Staffing imbalances. A proliferation of “directors” and staffing ratios with top-heavy organizational designs is an indication that the job leveling system is being used as a piggy bank—a way to generate pay increases through faux promotions. Source: Sibson Consulting What Is a Job?A “job” is a convenient handle for the sum total of work performed by an individual. It cannot be put in a bag and counted or set on a scale and weighed. Nonetheless, a job can and should be measured as accurately as any physical object.As shown in the figure below, a “job” is a construct that relates elements of what jobs are believed to consist of (role, requirements and responsibilities) to critical organizational outcomes. In science, these unobservable intermediary constructs are called hypothetical variables. We know they are there because they help explain relationships between one set of measured variables and another. In essence, the idea of a job helps us understand how organizations work.For example, poorly composed jobs produce role conflicts, place difficult demands on employees and create perceptions of inequity (e.g., doing too much for too little). The better jobs are structured (all else being equal), the better the organizational outcomes. A well-articulated division of labor, logically distributed levels of authority, sufficiently achievable employee responsibilities and readily determined market value of jobs are the makings of an efficiently run organization.Job-Measurement MethodsOrganizations have traditionally approached job leveling through one of five measurement methods: • Slotting.The requisite characteristics of jobs are described for each grade or band within a salary structure and jobs are then sorted into the appropriate level according to the job’s major responsibilities and accountabilities. • Whole job ranking. Jobs are internally compared against one another based on aggregations of their job properties. • Factor comparison. Rather than ranking jobs holistically, they are broken down into their constituent elements and ranked against one another on each element. Results are subsequently combined into an overall rank. • Point-factor rating system. Each element of a job is assigned a value based on how much of that element a job is perceived to contain. The total points designate the overall size of the job. Jobs can be categorized by points or grouped by ranges of points. • External market pricing. Jobs are matched to similar jobs in the relevant market and, essentially, receive points as dollar values for their constellations of attributes. As above, the values can be left as is, or grouped. Market rates are established for benchmark jobs through matching, and non-benchmark jobs are slotted into the structure based on similarities to classified (benchmark) jobs. Market Pricing Jobs: Cautionary NotesMany companies mistakenly believe that they can circumvent the perils of leveling by market pricing jobs. However, consolidating jobs based on market values is a method of job leveling that uses dollars as the standard of measurement. Here are a few cautionary notes about using market pricing as a method of job leveling: • Market pricing is subject to error just like any other method and, therefore, the precision of the job analysis, job documentation and job-matching processes are essential for achieving accurate results. • The market does not necessarily align with the internal value of jobs to the organization and, therefore, may miscommunicate relative importance to the specific organization. A paradigm case involves HR: When the market is used to determine grades, the head of HR always is lower than many of his or her senior peers. Often, CEOs view that outcomes as incorrect and detrimental to the operations of the organization and, consequently, overrule the market and insist on parity in the executive ranks. • The market often does not react quickly to shifts in industrywide transitions and, consequently, under- or over-represents value. For example, for a long time, marketing in consumer banking consisted of basic communications and the printing of brochures. Banks that wanted to upgrade their marketing could not find ready solutions within the financial services industry since the industry as a whole under appreciated the function. In order to internally reposition marketing, banks initially had to make value decisions independently of the traditional marketplace until the market caught up.The point is not that market pricing is a bad approach, but that it is not a panacea for the organizational strife that can be produced by a disputable internal order of jobs. Source: Sibson Consulting Organizations use each of these methods of measurement and each is a viable way of scaling jobs as long the potential for error is minimized. In order to decide which approach is best, an organization would need to evaluate its ability to guard against common sources of error when using a particular job leveling method. In this regard, Sibson often recommends assessing potential approaches to leveling against a checklist of items that may enhance or diminish the accuracy of the leveling method. These include factors related to item content, the response format, the intensity of administrative procedures, time requirements, system maintenance, and such. There are many ways the leveling of positions can go astray, and assuring that the proper controls are in place as the process commences is a good way to prevent organizational mutiny when the leveling results are revealed.Getting Job Leveling RightAll job-measurement systems are susceptible to error. The task is to eliminate as much as possible. One way is to recognize where errors are likely to occur.For example, some leveling systems that rely on questionnaires to measure jobs will lay out descriptions for each of the job’s criteria in a clear low-to-high order, often asking the incumbents to self-rate. Consequently, raters are aware of how their answers will affect the grade assignments of jobs, which makes it possible to retrofit ratings to the desired grade.Transparent questionnaire systems such as this do not enhance the accuracy of job leveling. Moreover, there is another problem with this, or any system, that relies on one or more non-independent raters. Even if the incumbents complete the forms honestly and objectively, it is impossible to know if their answers are correct. There is no external referent to validate the answers.Since a job is not tangible, how can an organization determine how big (or small) it really is and whether its placement into a structure is correct? The answer is, “When two or more people independently agree.” This could be called “convergent vision,” the only way to see what is not visible. It follows the old saying, “perception is reality.” There is truth when multiple raters viewing the same jobs reach the same conclusions (statistical indices reveal how accurately people classify jobs based on their level of agreement). People from outside the categorization process can provide input, and they can offer their opinions, but they cannot impose their political will on the outcome at the risk of corrupting the system.HR can ease its pain when overwhelmed by requests for job reclassifications by: • Construing jobs as the intangibles they are—there is no right or wrong, just agreement or disagreement. • Training a team of people in the leveling process regardless of the leveling technique selected. • Keeping vested interests from intruding on the process. Allowing personal bias to infiltrate the system violates a crucial tenant of convergent vision and removes one of job leveling’s ablest allies: objectivity.ConclusionWhile job leveling may not be the most exciting assignment, it is one of the core tasks and remains a centerpiece of many HR programs. Organizations with ill-defined and misclassified jobs are often poorly structured, ponderously ineffective and burdened with excess costs. If job leveling is done right, HR can transform what many falsely believe is a subjective exercise into one that is quantifiably precise and, in the process, make their organization fairer, more effective and less costly. Michael O’Malley, a PhD in social psychology and quantitative methods, is a senior vice president and human capital consultant for Sibson Consulting in New York City. This article originally appeared in the November 2015 issue of Sibson Consulting's Perspectives e-zine under the title “Accurate Job Leveling Matters: Measurement Is the Key to Success,” and is reposted with permission from Sibson Consulting, a division of Segal. © 2015 by The Segal Group Inc. All rights reserved. Related SHRM Article:How to Counter Employee Perceptions of Income Inequality, HR Magazine, May 2016
An organization run by AI is not a futuristic concept. Such technology is already a part of many workplaces and will continue to shape the labor market and HR. Here's how employers and employees can successfully manage generative AI and other AI-powered systems.