Guide · Job evaluation

Which job evaluation method to use?

The lightest method that still defends equal value.

Short answer: Pick the lightest job evaluation method that can still show why two different jobs are or are not of equal value. At about 200 roles that is usually a short points plan with a few clear factors. Ranking is faster but cannot explain itself. Heavy plans look safest and are often never finished.

Three proposals, one company

Picture a medical device distributor based in Minneapolis with about 200 people. Around 40 of them work in its Dutch warehouse and sales office. The company has job titles, a rough set of levels and pay that grew offer by offer. Now it needs a way to show which jobs are of equal value, because its EU staff will soon be able to ask how their pay compares and its US offices post ranges in job ads.

Three proposals land on the HR lead's desk in the same month. The finance lead suggests ranking: Put every job in order on one spreadsheet in an afternoon and draw grade lines through the list. A board member who ran HR at a large company recommends the plan used there, with a dozen factors, sub-factors and a scoring committee. A third option comes from a free EU toolkit, which suggests comparing jobs two at a time.

Each proposal has a real argument behind it. Ranking is fast and everyone understands it. The big plan is familiar and thorough. The paired method is designed for smaller employers. The HR lead's question is simple: Which one is enough for a company this size, without being more than it can keep up?

This is a composite of a common situation, not a client. The choice it describes is one of the first real decisions in any job evaluation project. It sets how long the work takes, who has to be involved and what the company can say if a pay comparison is ever challenged.

Why choosing a job evaluation method is hard

Job evaluation means judging the relative size of jobs in a consistent way, so that pay can follow the work rather than the history of each hire. The methods differ in one main way. Some compare whole jobs by overall judgment. Others break each job into factors, such as knowledge, responsibility or working conditions and score each factor separately. Guidance calls these non-analytical and analytical methods [1] [2].

Two good goals pull against each other. The first is defensibility. Equal pay law in the EU, the UK and some US states, such as Oregon, asks whether jobs are of equal value or comparable character, judged on skill, effort, responsibility and working conditions [2] [3] [6]. A method that can show its reasoning factor by factor is easier to defend than a method that says "we judged this job to be bigger". The second goal is that the method must actually get done and stay done. A plan too heavy for the company is finished late, applied unevenly or quietly dropped. An unfinished method defends nothing.

Size makes the choice harder at around 200 people. A company with 15 jobs can compare them all by hand. A company with 5,000 people has a reward team to run a full plan. At 200, the company is too large for informal judgment and too small for a heavy program. It also changes fast: New jobs appear every quarter, so whatever method it picks has to handle a new role in days, not months.

The official guidance points in two directions at once. The EU's gender-neutral job evaluation toolkit, published in March 2026, sets out a full points method as its standard approach for organizations with more than 250 workers and for medium-sized ones with varied jobs and structured HR processes. For smaller and medium-sized organizations where that is not feasible, it offers a paired comparison method, which it calls non-analytical, yet structured [1]. The UK equality regulator, by contrast, says a job evaluation must be analytical to be relied on in defending an equal value claim [2]. The toolkit itself is not legally binding and does not interpret national law [1]. Companies close to the 200 mark, with staff in more than one country, sit right in that gap.

Five traps when choosing a method

Most of these traps come from choosing a method for how it looks or how fast it starts, rather than for what it will have to prove later.

"Ranking is enough at our size"

What it looks like: A small group puts every job in order from biggest to smallest, by overall impression, then draws grade lines through the list. Sometimes the ranking is made more careful by comparing jobs in pairs.

Why smart teams fall into it: It is fast, cheap and easy to explain. The results often look right, because the people ranking know the company well. For a first pass it can be a useful sanity check.

What it costs: Ranking can say that one job is bigger than another. It cannot say why. When an employee asks why the customer care lead sits two grades below the warehouse supervisor, the only answer is that the group thought so. That is weak in exactly the situation job evaluation exists for. The UK regulator's guidance says a job evaluation must score each job against factors to be relied on in defending an equal value claim [2]. The EU toolkit notes that judgment-based methods can unintentionally favor jobs mostly held by men and overlook work typical of jobs mostly held by women [1]. Ranking also scales badly. Comparing every job with every other job in pairs means 190 comparisons for 20 jobs and about 4,000 for 90. The EU toolkit itself says the paired method works best with a relatively small number of jobs [1].

"Buy the heaviest plan to be safe"

What it looks like: The company adopts a large factor plan built for big organizations: Many factors, many sub-factors, long level definitions and a committee that scores every job from a detailed questionnaire.

Why smart teams fall into it: More factors feel more thorough. Someone on the leadership team has seen the plan work elsewhere. A heavy plan also signals seriousness to a works council, an auditor or a board.

What it costs: Time first. Scoring a large plan well takes training, long job questionnaires and many committee hours. At a 200-person company the same five or six people end up doing all of it next to their real jobs. The project slips, some families are scored and others are left for later. The UK regulator lists being fully completed across the organization as one of the tests of a reliable scheme [2]. A plan that covers half the jobs fails that test however good its design. More factors also do not always mean more accuracy. In a company this size, many fine sub-factors end up scored the same way for most jobs, which adds work without changing the order.

"Tune it until it matches the order we already have"

What it looks like: The team picks a method, scores a first set of jobs and compares the results with the current pay order. Where they differ, the factor weights or level definitions are adjusted until the results look familiar.

Why smart teams fall into it: A result that matches today's structure causes no difficult conversations. Leaders trust it because it confirms what they already believe. Tuning is also easy, because weights look like technical settings rather than choices.

What it costs: It turns the method into a way of restating current pay, including whatever bias current pay contains. The EU toolkit warns that gender bias can enter through how factors are defined, how points are assigned and how weights are applied [1]. A method tuned to protect the current order will tend to protect the current gaps. Some differences between the method's result and current pay are the point of the exercise. They are where the company learns something. The sound way to handle them is to look at each one and decide whether the job or the pay is out of line, not to tune them away. Our guide on which pay gaps you can defend looks at the same problem from the pay review side.

"Score the person in the job"

What it looks like: Jobs are scored from what the current holder does, knows or has achieved. The job held by a strong performer with a degree comes out larger than the same job held by someone newer.

Why smart teams fall into it: At 200 people many jobs have one holder, so the job and the person feel like the same thing. Describing what a person does is easier than describing what the job requires. Managers also want their best people recognized.

What it costs: The method stops measuring jobs. The UK regulator's guidance puts it plainly: A job evaluation scheme evaluates the job, not the job holder [2]. Oregon's equal pay law defines work of comparable character by the knowledge, skill, effort, responsibility and working conditions the work requires, regardless of job description or job title [6]. A score that rises and falls with the holder cannot show that. Person-based differences belong in where someone sits within a pay range, which is a separate decision with its own reasons.

"A one-off project will settle it"

What it looks like: The method is chosen and run as a project. All jobs are scored, grades are set and the project team moves on. Nobody owns the method afterwards.

Why smart teams fall into it: Job evaluation is sold and budgeted as a project. Once grades exist, the problem feels solved.

What it costs: A growing company creates and changes jobs all the time. Within a year, new roles have been placed by analogy and changed roles have kept their old scores. The UK regulator's guidance expects job information to be revised when jobs change significantly and the outcomes for new and changed jobs to be monitored by gender [2]. The EU toolkit suggests re-evaluating a job when its content changes substantially and reviewing the whole system every three to five years [1]. So the right method is one the company can run for a new job in an afternoon, with the same rules it used in the project. A method only consultants can operate leaves every new job unscored or scored by guesswork.

The main options and their trade-offs

The table compares the method types most companies consider. It describes what each one does and where it fits, not how to run it.

Method typeHow it decidesEffort at about 200 rolesCan it show why two jobs differ?Where it fits
RankingWhole jobs ordered by overall judgmentLowNoA first look or a cross-check, not the record
Paired comparisonEach job compared with every other on agreed criteria; wins added upGrows fast with the number of jobsPartly; it records which job won, not by how much on each factorSmall sets of jobs; offered by the EU toolkit for smaller employers [1]
ClassificationJobs slotted into grades by matching written grade descriptionsLow to mediumOnly as far as the grade descriptions doStable organizations with similar jobs
Short points planA few defined factors, each with levels and points; totals set gradesMediumYes, factor by factorGrowing companies that need equal value defensibility
Full points planMany factors and sub-factors, detailed questionnaires, a standing committeeHighYes, in detailLarge organizations with a reward team; the EU toolkit's standard approach above 250 workers [1]

"Factor comparison" is used for two different things. In its classic form it ties each factor to the pay of benchmark jobs, which builds current pay into the values it is meant to test. The EU toolkit uses the name for a simple factor method for micro-organizations with fewer than ten workers [1]. Neither is a common choice for a company of 200.

What a sound method choice looks like

A sound choice is not the most sophisticated method available. It is the lightest method that can still answer the equal value question for this company. Several qualities show it.

It can explain any two results. Pick two jobs in different families, such as a customer care lead and a warehouse supervisor. The method should show, factor by factor, why they land where they do. If the only explanation is the total, or the panel's view, the method is too light for the question it will be asked.

Its factors cover the whole range of work. The factors reach the demands of every job family, including the ones that are easy to overlook: Communication, care, emotional demands, coordination. Equal pay guidance stresses that skills often associated with jobs mostly done by women are frequently left out or undervalued [1] [2]. Six well-defined factors that cover the full range beat fifteen that mostly describe management work.

It gets finished. Every job in the company is scored within the project, not most of them. The size of the method matches the people available to run it.

It handles a new job quickly. A new role can be scored with the same rules in a short session, by people the company has, with a written reason. That is what keeps the structure true a year later.

Its differences from current pay are looked at, not tuned away. Where the method and current pay disagree, each case gets a decision and a reason.

It is documented. Factor definitions, level descriptions, scores and the reason for each score are written down. Under US equal pay guidance, a job classification system used to explain pay differences has to accurately reflect job duties and be applied the same way to men and women [7]. Documentation is how a company shows that.

Leveling frameworks, where they exist, usually help. A company with clear levels already has much of the job information a method needs. What levels rarely provide is a common yardstick across families. Our guide on whether the EU will make you evaluate jobs looks at that gap in detail.

The calls only you can make

Which factors say what you value? Every method weights some kinds of work above others. Whether customer-facing skill counts as much as technical depth, or how much physical and emotional demands count, is a statement about the company. It should be made openly by leadership, inside the limits of the law, not left to a default.

How many grades do the results feed? The method sizes jobs. The company decides how finely to cut the results. Our guide on how many job levels you need covers that choice.

Who scores the jobs? A panel of managers, HR alone, an outside team or a mix. The UK regulator's guidance recommends a panel of three or more with broad knowledge of the jobs, trained in avoiding bias and reflecting the workforce [2]. Who sits on it affects how far the results are trusted.

How tightly do grades drive pay? Some companies tie each grade to one pay band. Others use the evaluation as a check on pay set mainly by the market. Either can work. The choice changes what the evaluation will be asked to defend.

What happens to jobs that score below their pay? Before scoring starts, decide in principle how those cases will be handled. Deciding after the results arrive makes every case look personal.

The law, the guidance and the numbers

  • EU. The pay transparency directive requires pay structures that let employers assess whether workers are in a comparable situation for work of equal value, using objective, gender-neutral criteria that include skills, effort, responsibility and working conditions. For employers that must report pay gaps, an unexplained gap of 5% or more in any category of workers doing equal work or work of equal value can trigger a joint pay assessment [3].
  • EU toolkit. The EU-wide guidelines on gender-neutral job evaluation, published on 26 March 2026, offer a graduated factor comparison for organizations with fewer than ten workers, a paired comparison for smaller and medium-sized organizations where the standard approach is not feasible and a points method as the standard approach above 250 workers. The toolkit is not legally binding [1].
  • UK. The equality regulator's guidance says a job evaluation relied on in an equal value defense must be analytical, thorough and impartial, gender neutral and reliable [2].
  • US. Oregon's equal pay law covers work of comparable character, defined by the knowledge, skill, effort, responsibility and working conditions the work requires, regardless of job title. Pay differences must be fully explained by listed factors such as seniority, merit, education, training or experience [6]. Federal guidance requires a job classification system used as a defense to reflect duties accurately and be applied uniformly [7].
  • Readiness. In a September 2026 poll of more than 200 HR professionals in Ireland, 48% named job evaluation and grading as their biggest challenge in preparing for the directive [4]. In an October 2025 survey of 391 HR professionals in Ireland, 1% felt fully prepared and 25% named categorizing employees who do the same work or work of equal value as a main challenge [5].

This is general information, not legal advice. Rules and their national versions differ. Check the rules where you employ people.

Our view

We think a company of about 200 people should pick the lightest job evaluation method that can still explain, factor by factor, why two jobs in different families are or are not of equal value. In most cases that is a short points plan: A handful of clearly defined factors that cover the whole range of work, applied to jobs rather than people and documented well enough to explain.

Ranking has a place as a first look and a cross-check. It should not be the record a company relies on, especially with staff in the UK or EU. Paired comparison is a reasonable choice for a small set of jobs. It becomes hard work quickly and still cannot show why one job outweighs another.

We would argue hardest against the heavy plan that is never finished. It looks like the safest choice and often leaves the company with half its jobs scored and no way to place the next one. A lighter method, finished and maintained, defends far better.

Sources

  1. European Institute for Gender Equality, EU-wide guidelines on gender-neutral job evaluation and classification: Step-by-step toolkit (26 March 2026), checked 2026-10-26
  2. Equality and Human Rights Commission, Equal pay: How do I carry out job evaluation? (3 September 2020), checked 2026-10-26
  3. Directive (EU) 2023/970 on pay transparency, Articles 4, 9 and 10 (EUR-Lex), checked 2026-10-25
  4. Mason Hayes & Curran, Only one in five Irish employers have made real progress on EU pay transparency rules (8 September 2026; poll of more than 200 HR professionals), checked 2026-10-26
  5. Law Society Gazette, report on Ibec's HR survey (23 October 2025; 391 HR professionals), checked 2026-10-26
  6. Oregon Revised Statutes 652.210(16) and 652.220 (text via Oregon.public.law), checked 2026-10-26
  7. US EEOC, Compliance Manual Section 10: Compensation Discrimination, 10-IV.F.2.d (job classification systems), checked 2026-10-26

Are two different jobs of equal value?

Titles and job families do not settle equal value. The demands of the work do. Five traps in proving the comparison and what sound evidence looks like.

Job family or level: Which comes first?

Levels first, families second. Why job architecture projects stall on family trees and what an order of work that pays back early looks like.

What triggers a joint pay assessment?

A 5% gap in any category of workers can trigger an EU joint pay assessment. Five ways mid-size employers get caught and what a sound position looks like.

See Resources for every guide and the self-check.