Job Classification

What job classification is

Job classification is an approach to job evaluation in which the grades are defined first, in written descriptions, and each job is then assessed as a whole and placed in the grade whose description fits it.

Its distinguishing feature is that its unit of measurement is a described level, not points and not money. The scale is built before any job is looked at, and each job is matched against it once. That reverses the approaches that start from the job and work towards its position: here the place is built first, and only then is the question what fills it.

Accounts of job evaluation describe three approaches: ranking, classification and points. Classification is the second of them. A fourth approach, factor comparison, is sometimes added to the three. The point method works on factors rather than on the job as a single whole; classification works on the job as a single whole, and that is the first difference between the two.

One word for three different operations

In this field the word classification is used for three different things, and mixing them produces a misunderstanding that is hard to repair because it does not show on the document:

• Job classification, in the sense used throughout: placing the job at a described level. It applies to the job regardless of who holds it.
• Grade placement: placing the person in a grade and fixing their point within that grade’s pay range, once the level of their job has been set. The two terms are not interchangeable within one document.
• The Saudi occupational classification (التصنيف المهني السعودي): an ordering of occupations at the level of the market, with its reference point outside the organisation and a purpose different from that of an internal scale.

The first orders jobs inside the organisation, the second puts people into what was ordered, and the third describes occupations in the market. Anyone who writes that the jobs were classified while meaning the second has said something other than what was done, and the document still looks correct.

How it is carried out, in order

• Decide the number of grades. This is decided before a single description is written, and its numerical effect is shown below.
• Write a description of each grade, with boundaries that separate it from the next: what is decided at that level without referral, what is referred upward, what happens if a mistake is made, and the nature of the supervision received and given. These are the same considerations that compensable factors weigh one at a time; in classification they are gathered into a single description of the level.
• Anchor the grades with known jobs. Place one or two benchmark jobs with settled content in each grade, so the grade is understood through an example rather than an abstract description.
• Read the remaining jobs against the scale. Each job is assessed from its documented description, looking for the grade closest to it as a whole, not to its strongest part.

All of this is preceded by job analysis. The approach matches documented work content against a written description; if the content has not been described, the job title gets matched against the description instead, which accounts of job evaluation count among the things that invalidate it.

The number of grades decides what the descriptions must do

Choosing how many grades to have looks like an administrative decision. In fact it decides how precise the writing has to be. Take a scale whose grade midpoints run from SAR 4,000 in the bottom grade to SAR 32,000 in the top one, rising by the same percentage from each grade to the next:

• With 8 grades, the step from each grade’s midpoint to the next is about 34.6%.
• With 12 grades, it is about 20.8%.

In the second case the descriptions have to separate two levels that are only about a fifth apart in value. That is much harder to write, and it is where the method breaks down in practice: twelve grades are written with descriptions fit for eight, the grades overlap, and which one a job lands in comes down to whoever places it.

A second figure belongs beside it. An organisation with 60 distinct jobs puts 12 jobs in each grade if it has five grades, and 5 if it has twelve. The first treats as equal jobs their holders see as different; the second creates distinctions the descriptions must carry. So the question is not how many grades are wanted but how many differences can be written down and defended. The figures in both examples are assumed, to show how the calculation is done.

How a grade description is written

A sound description can be read and tells you where a job stands. An unsound one can be read and agreed with, and yields nothing. The difference lies in four places:

• Write in boundaries, not degrees. Not “a degree of autonomy”, but “decides this without referral and refers that”. A boundary can be checked against the content of a job; a degree can only be checked against the reader’s judgement.
• Read it with the two neighbouring grades, not on its own. Test a description by setting it beside the one above and the one below and asking which sentence here could not be said there. If there is none, the two grades are one grade on paper.
• Do not name a particular tool or system. Tools change and the description stays; a grade defined by mastery of a tool has to be rewritten whenever the tool is replaced.
• Do not state years of experience. Years are an attribute of the person, not the job, and putting them into the description of a level is the back door through which the holder returns to the evaluation of their own job. The same attribute, written as a ceiling in a job advertisement, is one of the routes by which ageism enters hiring.

The second of these is the practical test that saves a great deal of review, and it is carried out on paper before a single job is assessed. It applies the requirement that levels be described in terms that distinguish them, since a difference with no description is left to each evaluator’s own judgement.

A grade description also answers a different question from the performance standards of a job. The description says at what level the work sits; the standards say what any holder of the job must reach in it.

The job that falls between two grades

This is where the approach is tested. A job can fit the higher grade’s description in one respect and the lower grade’s in another, and the approach cannot combine the two into a number, because it produces no numbers.

There are exactly three written ways of handling this case, and each is declared in advance rather than left to circumstance:

• Decide by the bulk of the description: place the job where most of the description fits, and record what tipped it.
• Decide by the heaviest aspect: the policy states that one named aspect, such as the consequence of error, takes precedence when the two conflict.
• Revise the description itself: jobs repeatedly falling between two grades show that the descriptions do not discriminate, and the remedy lies in the scale, not in the job.

The third is the one that gets neglected. As long as the case is handled by tipping alone, the scale keeps its weakness, and individual decisions pile up that nobody can make sense of two years later or trace to a basis.

How it differs from the point method

• Cost. Classification is quicker, because it needs no questionnaire, no weights and no rating of each factor in each job.
• What can be shown when a result is challenged. The point method is the slowest approach, and what it offers in return is a detailed reason that can be shown when a result is disputed. Classification gives one reason, that the description fits; if the holder disputes that, there is no detail between them to go back to.
• Where human judgement sits. It is present in both. With points it is spread over factors that can each be argued separately; in classification it is gathered into one judgement on the whole job.
• What moves with the market. A description does not move when pay rates move, and in that it is like points and unlike the monetary scales used in factor comparison.

That is why the approach suits settings where jobs are similar in nature, the number of grades is limited, and the organisation needs a quick decision more than a detailed reason it can show. It does not suit settings where the nature of the work varies so much that one description of a level cannot be used to assess two jobs within it.

What governs the scale

In the sources we reviewed, we found no provision of the Saudi Labor Law (نظام العمل) that imposes a particular grade scale on establishments in the private sector, a particular evaluation method, a number of grades or the gaps between them. What governs those is what the organisation writes into its own documents and commits itself to. Job classification is an internal management instrument, and its purpose is that every level has a written reason that can be shown and reviewed.

The definition of job classification does not settle the pay amounts attached to each grade or how wide each grade’s range is. Those belong to the salary structure and to the spread of each range within it. None of this is legal advice.

What happens to a new job once the scale exists

The scale is built once and then new jobs are created, and that is its second test. A new job is compared with the existing descriptions, and one of two things happens: either a grade fits and the job goes into it, or nothing fits.

The second case is the one that gets mishandled. The usual response is to attach the job to the nearest grade as a concession, and to keep doing so until the grade fills up with jobs its description does not describe. The right response is to ask first whether this is really a new job or duties moved from existing jobs. If it is the latter, it belongs in the grades those duties came from; if it is the former, the scale is incomplete and its descriptions need revising.

It follows that the number of jobs attached as a concession is a figure worth counting: when it rises, the scale has become narrower than what sits beneath it, before that shows in anyone’s complaint.

What invalidates it

• Writing the descriptions after looking at current pay, so they are tailored to what already exists. Accounts of job evaluation count this as invalidating it: carrying it out to justify existing pay instead of fixing the level and deriving pay from it.
• Descriptions in words that do not distinguish. A grade described as carrying broad responsibilities, with the one below it carrying moderate responsibilities, leaves the difference to every reader.
• Placing jobs by title, when a title can change without anything underneath it changing.
• Doing it once. Jobs change, and a scale that is never reviewed becomes a description of the past.
• Placing by the jobholder. A person’s seniority or performance belongs to grade placement and what follows it, not to setting the level of the job.

Related terms

• Salary Structure
• Performance Standards
• Competencies
• Ageism
• Graphic Rating Scale