The Definitive Guide to the Point-Factor Method of Job Evaluation
Author
Justin Hampton
Date Published
The Point-Factor Method of Job Evaluation
TL;DR — The point-factor method scores every job in your organization against a fixed set of weighted compensable factors (skill, effort, responsibility, and working conditions, broken into sub-factors). Each job gets a numerical total, jobs slot into bands by score, and pay decisions become defensible, transparent, and consistent. It is the most widely used quantitative job-evaluation method in the world, and AI is making it dramatically faster to run.
If you have ever been asked, "Why is my role a Level 6 instead of a Level 7?" and struggled to give a clean answer, this guide is for you. By the end you will know how to design a point-factor system, run a job through it, defend the result in a comp committee, and avoid the four mistakes that quietly break most implementations.
Table of contents
What is the point-factor method?
The point-factor method is a quantitative job-evaluation technique that scores each job against a fixed set of compensable factors — the dimensions an organization decides are worth paying for. Each compensable factor is broken into degrees or levels (1 through 5, for example), each level carries a defined point value, and every factor is weighted relative to the others. A job's total score determines where it sits in your pay structure.
Two facts about the method matter more than the mechanics:
It is quantitative, not narrative. Where a ranking method asks evaluators to compare jobs holistically ("Is the Senior Engineer bigger than the Product Manager?"), the point-factor method forces evaluators to score each factor independently and adds the numbers up. That structure is what makes the method defensible in pay-equity litigation and consistent across hundreds of jobs.
*The factors are your decision.* Two organizations using point-factor evaluation can produce very different results because the weights they apply express their business strategy. A research lab will weight "innovation" and "knowledge" heavily. A logistics company will weight "scope of responsibility" and "physical effort" more. The method is a frame, not a verdict.
When people search for "point factor system" or "point factor job evaluation," they are usually trying to solve one of three problems: pay disparity that they cannot explain, role inflation as the company has grown, or a looming pay-equity audit. The point-factor method addresses all three because it produces a numerical justification for every internal pay decision.
Definition for snippets: The point-factor method is a quantitative job-evaluation technique that scores each job against weighted compensable factors — typically skill, effort, responsibility, and working conditions — and uses the total point score to determine the job's relative worth within an organization.
Where the point-factor method came from
The method traces to industrial engineers of the 1920s — Eugene Benge in particular — and was popularized for white-collar roles by Edward N. Hay in the 1950s. Hay's variant, the Guide Chart-Profile Method, became the world's most-licensed job-evaluation system and is now owned by Korn Ferry (often referred to as the "Hay system" or "Hay points"). Mercer's International Position Evaluation (IPE) is another commercial point-factor variant, and gradar and PointFactors are modern SaaS implementations.
The vocabulary you will see in older textbooks ("guide charts," "know-how points," "accountability," "problem-solving") is Hay's. The vocabulary you will see in modern HR ("compensable factors," "sub-factors," "degrees") is the more generic point-factor language. Both describe the same underlying logic: define what you pay for, weight it, score every job, total the points.
Compensable factors: the building blocks
A compensable factor is a dimension your organization considers worth paying for. The classical point-factor model groups them into four families, each broken into sub-factors:
1. Skill
The know-how a job requires.
Education (formal degrees, certifications)
Experience (years and depth in the discipline)
Technical skill (specialized tools, languages, systems)
Interpersonal skill (negotiation, persuasion, conflict resolution)
Initiative & ingenuity (independent judgment)
2. Effort
The mental and physical demand the job places on the holder.
Mental / analytical effort (problem complexity, abstract reasoning)
Physical effort (lifting, standing, repetitive motion)
Visual or sensory effort (precision work, monitoring)
3. Responsibility
The consequences of failure if the job is done poorly.
Responsibility for financial resources (budgets, P&L, revenue)
Responsibility for people (direct reports, indirect influence)
Responsibility for equipment or materials (capital assets, hazardous materials)
Responsibility for information (confidentiality, data integrity, IP)
Responsibility for outcomes / customer impact
4. Working conditions
The environment in which the job is performed.
Physical environment (heat, cold, noise, outdoors)
Hazards (chemical, electrical, ergonomic risk)
Travel / unsocial hours
Emotional demands (handling distressed customers, crisis response)
Most modern point-factor systems use 8 to 14 sub-factors total. Fewer than 8 and you lose the ability to differentiate similar jobs; more than 14 and inter-rater reliability drops because evaluators get fatigued. PointFactors' default scorecard uses 11 sub-factors, which the comp research community treats as the practical sweet spot.
Each sub-factor is broken into degrees (typically four to seven), where Degree 1 is the lowest demand and the top degree is the highest. The point values assigned to each degree are usually geometric, not linear — Degree 5 might be worth four times Degree 1, not just five times. This curve matters because it forces the system to discriminate at the senior end of the job hierarchy where most pay variance actually lives.
The eight-step point-factor process
A clean implementation follows these eight steps. Most companies who fail at job evaluation skip steps 2 and 7.
Step 1: Pick your compensable factors
Choose the 8–14 sub-factors that match your strategy. A SaaS company should not be evaluating "physical effort" as a meaningful factor; an industrial manufacturer should not be evaluating "innovation" as one of the top three. Get this right and the rest follows.
Step 2: Define the degree descriptors
For every sub-factor, write a one-paragraph definition of what each degree looks like with concrete behavioral anchors. "Degree 3 — Responsibility for People" should not read "moderate supervisory responsibility." It should read "Manages 3–8 direct reports, including hiring, performance reviews, and termination decisions, within an approved headcount budget."
The behavioral anchors are what make the system defensible. Without them, evaluators score from gut feel and inter-rater reliability collapses.
Step 3: Assign weights and point values
Allocate 1,000 total points (a common total — anything from 500 to 1,500 works) across your factors. A typical SaaS weighting:
| Factor family | Weight |
| --- | --- |
| Skill | 40% |
| Responsibility | 35% |
| Effort | 15% |
| Working conditions | 10% |
Inside each family, distribute weight across sub-factors. Then assign point values to each degree, using a geometric curve. The math is straightforward — your free scorecard (below) does it for you.
Step 4: Calibrate on benchmark jobs
Pick 8 to 15 benchmark jobs that span the seniority and function range of your company — a junior individual contributor, a senior individual contributor, a first-line manager, a director, a VP, a finance role, a sales role, an engineering role, an operations role. Score them as a committee. Compare results against market data for those same roles. If your scoring puts a Senior Engineer at 720 points but the market pays Senior Engineers like your Directors who score 760, your weights are misaligned and you re-tune them.
This calibration step is where most internal point-factor projects get into trouble. Skip it and you produce a beautiful internal hierarchy that does not match how the labor market actually values your jobs.
Step 5: Score the remaining jobs
Use a panel of 3–5 trained evaluators per job. The standard practice is for each evaluator to score independently, then the panel meets to reconcile differences. Average the reconciled scores.
Step 6: Build bands from the score distribution
Plot every job by total points. Most distributions form a roughly continuous curve, not natural clusters. You impose bands by drawing horizontal lines — typically 8 to 14 bands across the whole organization. A common approach is to make each band ~10–12% wider than the one below it, so the absolute range gets bigger at senior levels where pay variance is higher.
Step 7: Link bands to pay ranges
For each band, set a salary range from a midpoint (anchored to your target market percentile — say, the 60th percentile of paying companies) plus a spread (commonly ±20% around the midpoint, ±25% at senior bands). This is where the point-factor system meets your compensation strategy.
Step 8: Document, communicate, and re-score periodically
Document every factor definition, every weight, every benchmark score. Publish the band structure (not necessarily individual job scores) to employees. Re-score jobs whenever they materially change, and audit the whole structure every 18–24 months.
A worked example: scoring a Finance Manager
Imagine you are scoring a Finance Manager at a 200-person SaaS company with the SaaS weighting above and a 1,000-point system using five degrees per sub-factor (40, 100, 200, 320, 480 point geometric curve).
| Sub-factor | Weight (%) | Degree | Points |
| --- | --- | --- | --- |
| Education | 8% | 4 (master's preferred) | 64 |
| Experience | 10% | 4 (7–12 yrs) | 80 |
| Technical skill | 12% | 4 (advanced financial modeling + ERP) | 96 |
| Interpersonal skill | 6% | 3 (cross-functional partner, no execs) | 30 |
| Initiative | 4% | 3 (operates with weekly check-ins) | 20 |
| Mental effort | 8% | 4 (complex non-routine analysis) | 64 |
| Physical effort | 2% | 1 (office) | 2 |
| Responsibility for finances | 14% | 4 ($50M revenue scope, 7-figure budget) | 112 |
| Responsibility for people | 10% | 3 (2–3 direct reports) | 50 |
| Responsibility for outcomes | 11% | 4 (board-visible deliverables) | 88 |
| Working conditions | 5% | 1 (office) | 5 |
| **Total** | **100%** | | **611** |
A Finance Manager scoring 611 points lands in Band 7 in this company's 10-band structure, with a salary range of $135K–$185K midpoint $160K. If a director-level Finance Director scores 740 points and sits in Band 8 ($175K–$240K), the relationship between the two roles is now numerical, defensible, and reproducible.
Two practical observations from this example:
Notice how "Responsibility for finances" alone contributes 112 points. This is why the factor weights matter. If you weight financial responsibility lower (because, say, you are a research lab and innovation matters more than budget scope), this same Finance Manager would score lower and might sit in Band 6 instead.
Most of the differentiation between similar roles is in 2–3 sub-factors, not all 11. That is exactly the point. The method exists to surface those differentiating factors and price them.
Point-factor vs. the other three classical methods
The HR literature describes four classical job-evaluation methods. Here is how they compare.
| Method | How it works | Effort | Defensibility | When to use |
| --- | --- | --- | --- | --- |
| **Ranking** | Evaluators rank-order all jobs from highest to lowest. | Low | Low — no documented logic | < 30 jobs, single function, early-stage |
| **Classification** | Pre-written job-class descriptions (e.g., GS-1 through GS-15 in the US Federal system). Each job is matched to the nearest class. | Medium | Medium | Government, public sector, very large standardized workforces |
| **Factor comparison** | A hybrid: factor-by-factor ranking, then a rate per factor. | High | Medium | Almost never used today; superseded by point-factor |
| **Point-factor** | Score each job against weighted compensable factors; total the points; bands follow from scores. | Medium-high to set up; low to maintain | High | Most for-profit organizations with 50+ jobs |
For more on the alternatives, see our deep dive on the four methods of job evaluation and the standalone factor-comparison method article.
Point-factor vs. Hay vs. Mercer IPE
The most common confusion when companies are shopping for a job-evaluation approach is: are Hay and IPE different methods, or are they point-factor systems with different brands?
The answer is the latter. Both Hay (Korn Ferry) and IPE (Mercer) are commercial point-factor systems. Their underlying logic — score weighted factors and total — is the same. What differs:
Factor names and groupings. Hay uses Know-How, Problem-Solving, and Accountability as its three main factors, with sub-factors under each. IPE uses Impact, Communication, Innovation, Knowledge, and Risk. PointFactors uses the classical Skill / Effort / Responsibility / Working Conditions framing because it maps more cleanly to modern HR vocabulary.
Calibration database. Hay and Mercer both maintain proprietary databases that map their scores to market pay. This is part of what you pay for — and part of what locks you in.
Cost and licensing. Both are commercial methodologies that require training, certification, and ongoing license fees. A typical enterprise spend on Hay or IPE methodology, training, and benchmarking runs $50K–$250K per year.
Flexibility. Hay and IPE prescribe their factors. A modern point-factor system lets you choose factors that match your business.
For a side-by-side feature comparison, see our pages on the Hay methodology alternative and the Mercer IPE alternative.
Four common pitfalls (and how to avoid them)
After watching dozens of point-factor implementations across SaaS, manufacturing, healthcare, and professional services, four mistakes show up most often.
Pitfall 1: Too many sub-factors
The temptation is to be "thorough" and include 18, 22, even 28 sub-factors. Inter-rater reliability collapses past 14. Evaluators get fatigued, scores drift toward the middle, and the system loses discriminatory power exactly where you need it (between adjacent senior roles).
Fix: Cap at 14. If a factor is not differentiating, drop it.
Pitfall 2: Linear point curves
If your degrees are worth 20, 40, 60, 80, 100 points (a linear curve), your system will under-discriminate at senior levels. The difference between a senior engineer and a principal engineer is not the same magnitude as the difference between a junior engineer and a mid-level engineer — but a linear curve treats it as such.
Fix: Use a geometric curve (e.g., 40, 100, 200, 320, 480). The widening gap at higher degrees reflects how senior-level value actually scales.
Pitfall 3: Skipping calibration against the market
A common failure mode: an HR team builds an internally elegant point-factor system, scores 200 jobs, and discovers that their internal hierarchy puts the Senior Software Engineer below the Senior Accountant, while the market pays the engineer 40% more. Now what?
Fix: Calibrate on benchmark jobs in Step 4. Re-weight factors until the internal hierarchy lines up with market data for those benchmark jobs before you score everyone else.
Pitfall 4: Treating the score as the answer
The score is a starting position for a pay decision, not the decision itself. Tenure, performance, pay-equity adjustments, retention concerns, and geographic differentials all layer on top. Companies that publish point scores per employee invite endless arguments. Companies that publish band structures (and keep individual scores in the comp team) avoid most of them.
Fix: Communicate the system, not the individual scores. Train managers to talk about bands, not point totals.
How AI is changing point-factor evaluation
The slow part of a point-factor implementation is not the math — it is the human scoring. Five evaluators, 200 jobs, 11 factors each, multiple reconciliation meetings: it is hundreds of hours and weeks of calendar time.
AI changes the economics in three ways.
1. First-pass scoring at scale. An LLM trained on a company's own degree descriptors and behavioral anchors can read a job description and produce an initial score for every sub-factor. Human evaluators review and adjust, but they no longer start from a blank scorecard. Companies are reducing scoring time by 70–85% with this pattern.
2. Consistency checks. AI can flag inconsistencies a human panel would miss — for example, two roles in different functions that scored differently on "Responsibility for People" despite having identical span-of-control language in their job descriptions.
3. Continuous evaluation. Traditional point-factor systems get re-scored every 18–24 months. AI lets you re-score whenever a job description changes, so band assignments stay current without a full project.
This is the bet PointFactors is making: not that the point-factor method needs to be replaced, but that it has been waiting 70 years for tooling that matches its rigor. (If you want to see this in action, book a 20-minute demo.)
When the point-factor method is the right choice
Point-factor is the right method when:
You have 50+ jobs across multiple functions
You need defensibility for pay-equity audits or union negotiations
You want internal pay decisions that follow consistent logic across teams
Your company is growing fast enough that role inflation, title compression, or band creep is becoming a problem
You expect a regulatory or compliance reason (EU Pay Transparency Directive, US state pay-disclosure laws) to require a documented internal pay logic
Point-factor is not the right method when:
You have fewer than 30 jobs (use ranking)
You are a government or quasi-government body with a pre-existing classification scheme (use classification)
Your entire workforce is in one function and one level (no method really helps — just benchmark to market)
For more on which method fits your situation, see the four methods of job evaluation compared.
Free point-factor scorecard
We built a free Excel scorecard that implements everything in this guide: 11 sub-factors, configurable weights, geometric degree curves, automatic band assignment from total scores. Use it to run a pilot on 10 benchmark jobs in a couple of hours.
[Download the free Point-Factor Scorecard →](/templates/point-factor-scorecard/)
FAQ
What is the point-factor method in simple terms? The point-factor method is a way to evaluate jobs by scoring each one against a fixed list of weighted criteria — like skill, effort, responsibility, and working conditions — and adding up the points. The total decides which pay band the job belongs in.
What is the difference between the point-factor method and the factor-comparison method? The point-factor method assigns numerical points to each level of each factor and adds them up. The factor-comparison method ranks jobs against each other on each factor and assigns money values directly. Point-factor is more common today because it scales better and is easier to defend.
What is the SHRM definition of the point-factor method? SHRM defines the point-factor method as "a quantitative job evaluation procedure that determines the value of a job by calculating the total points assigned to it." Compensable factors are identified, weights are assigned, and each job's total point score determines its placement in the pay structure.
What are the four major compensable factors? The four classical compensable factor families are Skill, Effort, Responsibility, and Working Conditions. Each is broken into 2–4 sub-factors. Most modern point-factor systems use 8–14 sub-factors in total.
Is the Hay method a point-factor method? Yes. The Hay Guide Chart-Profile Method is a commercial implementation of the point-factor method, owned by Korn Ferry. It uses three main factor families (Know-How, Problem-Solving, Accountability) but follows the same scoring-and-totaling logic.
How many points should a point-factor system have? Total point ranges are arbitrary, but 500–1,500 total points is typical. The most common configuration is 1,000 points with 8–14 sub-factors using geometric degree curves.
How often should I re-score jobs? Re-score a job whenever its description materially changes. Audit the full system every 18–24 months. With AI-assisted tools, continuous re-scoring is becoming the new standard.
Does the point-factor method help with pay equity? Yes — significantly. Because every pay decision can be traced to a numerical score against documented factors, point-factor evaluation provides the audit trail regulators and plaintiff attorneys ask for. It does not eliminate pay-equity risk on its own; you still need to audit and remediate. But it transforms the conversation from "we couldn't really say" to "here is the score, here is the band, here is the range."
Key takeaways
The point-factor method scores jobs against weighted compensable factors. The total drives band placement.
Most successful implementations use 8–14 sub-factors, geometric degree curves, and calibration against market data.
The Hay system and Mercer IPE are commercial point-factor variants. PointFactors is a modern AI-assisted implementation.
The four most common failure modes are: too many sub-factors, linear point curves, no market calibration, and treating scores as final pay decisions.
AI is collapsing the cost of implementation by 70–85% — making point-factor evaluation practical for companies that could not previously justify it.
Want to run a point-factor evaluation on your own jobs in under a week? [Book a 20-minute demo →](/demo/)