Point factor job evaluation systems are always imperfect temporary mathematical models. Point plans were originally created to apply internal equity to jobs that could not be appropriately matched to competitive market benchmarks. Good survey data might exist for Widget Apprentice and Widget Expert while no other employer might have a job equivalent to your unique Widget Generalist. Without reasonably precise survey matches, pay decisions must hinge on some other process.
“Pegging” is one such option, picking a spot between two others. A unique job might fall between two known benchmark values, but you still need another form of job evaluation to determine exactly how much between them. Interpolations create arbitrary guesstimates of exactly where an intermediate occupation should be inserted between its higher and lower occupational cousins. Point plans are popular with those who prefer a precise number suggesting exactly how much more or less.
Point factor plans convert scalar information (above or below) to interval status (i.e., -8 or +5) with precise numbers. A mathematical language is created to rank job characteristics in terms of levels tied to weighted numbers that translate into pay. Numeric points are accumulated across a number of evaluation factor dimensions with both scalar and interval values and are applied to administer compensation. This permits the interval-based interleaving of unique jobs with benchmark jobs sharing similar overall value scores.
When any specific plan was first created is usually easy to discover. The lowest number of "points" you can establish for any job will generally be the minimum wage of the year it was initially designed. Bringing old point values up to date for modern use is more complex. Simple flat (or polynomial) formula increases applied to outdated weighted score numbers will never accurately capture true changes in those underlying factor values over the years. But the marketing lifetime of off-the-shelf prepackaged job evaluation plans is greatly extended by skipping factor revalidation steps. Also, it is much cheaper and easier to train sales people and support consultants on a relatively static uniform version of a point-factor plan. If the product being sold is substantially different for each client, its cost will be higher. In addition, any plan requiring an advanced mathematical vocabulary for explanation or maintenance will encounter resistance from all parties. It's much easier just to make and resell a cookie-cutter standard model rather than customize each system for optimal statistical precision. Obviously, those are arcane “secrets” generally hidden from the end users who usually don't care about it, anyway.
That said, most point factor systems survive because they overpay the jobs. No one will ever complain about being overpaid. The moment an evaluated job's internally equitable pay falls too far below the market-clearing rate, it generally gets re-evaluated upwards, reclassified by a content/title tweak or artificially promoted. Occasionally, it will be green-circled, placed in a superior range structure or given a higher-paying conversion formula. Without those accommodations, the company would hemorrhage workers and close down for lack of competent personnel. Underpaying brings obvious penalties while overpaying is universally embraced.
Other job evaluation systems that are less precise for interval positioning also exist: whole job, internal benchmarking, paired comparison, slotting, factor comparison, committee rankings, tabular classifications and variants.
It is important to recognize that "market pricing" (where dollars are the points) remains the simplest form of point-factor job evaluation. Ranking work by current market value is problematic, however, when no market match exists. Point plans that add unique rationales, standardize work valuation language and apply terms used as "cutters" can justify further differentiation and permit peer match refinements. Essential point-factor job evaluation system elements usually cover comprehensive work content details along dimensions of the skill, effort, responsibility and working conditions required for each incumbent.
I better stop, because this is probably already far more than anyone wanted to know about this topic.
E. James (Jim) Brennan is an independent compensation advisor with extensive total rewards experience, specializing in job evaluation, market pricing and pay budget distribution. After corporate HR jobs in chemical and pharmaceutical manufacturing, he consulted at retail, government, energy, IT, tax-exempt and other industries throughout North America before becoming Senior Associate of pay survey software publisher ERI until returning to consulting in 2015. A prolific writer (author of the Performance Management Workbook) and speaker, Jim gave expert witness testimony in many reasonable executive compensation cases both for and against the Internal Revenue Service and also serves on the Advisory Board of the Compensation and Benefits Review.
Image of Counting Scale by Yulius Adrianto, courtesy of Creative Commons