Insight · Classification & Compensation

Pay Range Design: Range Width, Midpoint Progression, and the Five-Year Refresh Rule

By Tabitha McKenney Weinstein · September 18, 2026 · 8 minute read

A pay structure is architecture, not a table of numbers. Three design decisions carry almost all of the load: how wide each range is, how far apart the grade midpoints sit, and how often the whole structure gets rebuilt against the market. Get those three right and a pay plan holds for a decade. Get them wrong and the structure drifts into compression, off-scale placements, and red-circled salaries within three years, at which point every pay decision becomes a negotiation instead of an application of policy. This is how the three decisions get made.

What a pay range actually is

A single pay range has three anchor points: a minimum, a midpoint, and a maximum. The minimum is the entry rate for a fully qualified new hire. The maximum is the ceiling a fully seasoned incumbent can reach without a promotion. The midpoint is the target rate for a competent, experienced employee performing the full scope of the job, and it is the point that gets anchored to market. Everything else in range design is a relationship between these points, expressed in two numbers.

The first number is range width, also called range spread: the maximum expressed as a percentage above the minimum. A range that runs from $50,000 to $70,000 has a 40 percent width, because the maximum sits 40 percent above the minimum. The second number is midpoint progression, sometimes called the midpoint differential: the percentage gap between the midpoint of one grade and the midpoint of the grade directly above it. These two numbers, applied consistently across every grade, are what turn a list of jobs into a coherent structure.

Range width discipline: 35 to 50 percent

Range width answers a single question: how much room does an employee have to grow in pay without changing jobs? Too little room and there is nowhere to reward experience, which forces managers to promote people out of jobs they should stay in just to give them a raise. Too much room and pay stops tracking to the work, cost becomes uncontrolled, and two people doing identical jobs can sit $40,000 apart with no defensible reason. Pinnacle's house standard sets typical range width between 35 and 50 percent, and the number is not arbitrary.

Narrow ranges below 30 percent

Narrow ranges suit step-based, highly routinized work where the difference between a new hire and a veteran is small. Public safety recruit-to-officer progressions and entry clerical series often live here. The risk with narrow ranges is that employees hit the maximum quickly and then stall. A workforce clustered at range maximum has no place to go, and the structure starts delivering annual increases as red-circle exceptions rather than as movement within range. When most of a grade sits at the top of a narrow range, the range is too narrow for the actual career length of the job.

Wide ranges above 55 percent

Wide ranges suit professional and managerial work where the distance between a competent performer and an expert is large and takes years to travel. Senior professional and management series justify wider ranges. The risk is the mirror image: a range wide enough to hold a $45,000 gap between two incumbents in the same grade invites pay decisions that cannot be explained by tenure or performance, which is exactly the fact pattern a pay equity analysis flags first. Width buys flexibility, and flexibility uncontrolled is how inequity enters a structure.

Why 35 to 50 percent holds

The 35 to 50 percent band is wide enough to reward a full career of growth within a grade and narrow enough that pay stays tethered to the work. It also produces healthy overlap between adjacent grades when paired with disciplined midpoint progression, which matters because overlap is what lets an experienced incumbent in a lower grade out-earn a brand-new hire in the grade above without breaking the structure. That overlap is a feature. It is the shock absorber that keeps promotions and lateral moves from creating cliffs.

Midpoint progression: 8 to 12 percent

Midpoint progression is the vertical spacing of the structure. If range width is how tall each floor is, midpoint progression is how far apart the floors sit. Pinnacle's house standard sets typical midpoint progression between 8 and 12 percent between adjacent grades, and the discipline is to keep that progression consistent so the structure reads as a ladder rather than a pile.

Progression that is too small, under about 7 percent, collapses grades into each other. When two grades sit only 5 percent apart at the midpoint, the promotional increase from one to the next is smaller than a strong merit increase, and employees correctly perceive that the promotion is not worth the added responsibility. The grades stop meaning anything. Progression that is too large, above about 15 percent, opens gaps the structure cannot bridge, which forces off-scale hiring, special salary adjustments, and the exact one-off exceptions that erode structural integrity over time.

Here is the arithmetic working correctly. Take a grade with a midpoint of $60,000, a 40 percent range width, a minimum of $50,000 and a maximum of $70,000. Apply 10 percent midpoint progression and the next grade up has a midpoint of $66,000. Hold the 40 percent width and that grade runs from $55,000 to $77,000. The two ranges overlap from $55,000 to $70,000. An experienced incumbent near the top of the lower grade earns more than a new hire at the bottom of the higher grade, the promotional increase to the new grade midpoint is a meaningful 10 percent, and no one has to be placed off-scale. That is a structure doing its job.

Range width and midpoint progression are not independent settings. Width without progression discipline produces overlap so large the grades blur together. Progression without width discipline produces cliffs. They have to be designed as a pair.

The five-year refresh rule

A pay structure is built against a market that does not hold still. Even a perfectly designed structure ages, because the external labor market it was anchored to keeps moving while the structure sits. Pinnacle's standard is to rebuild the structure against fresh market data on a cycle no longer than five years, with a lighter midpoint check at the midpoint of that cycle. The five-year outer bound is not a preference. It is the point past which a structure stops being evidence of anything.

Markets move unevenly. Some job families run hot for a stretch while others stay flat, so a structure that was internally consistent at launch develops soft spots as specific families outpace the general adjustment. A single across-the-board cost-of-living increase applied every year does not fix this, because it moves every grade by the same percentage and preserves the exact relationships the market has already broken. Only a rebuild against current survey data re-anchors the midpoints to where the market actually is.

The signs a structure has gone stale

A structure past its refresh date announces itself. The tells are consistent: growing salary compression, where the gap between supervisors and their direct reports narrows toward zero. Rising red-circle counts, where more incumbents sit frozen above their range maximum. Off-scale hiring, where the posted range no longer attracts qualified candidates and hiring managers negotiate exceptions to fill roles. And midpoint lag, where the structure's midpoints have fallen measurably below the market rate for the same jobs. Any one of these is a signal. Two or more together mean the refresh is overdue, not upcoming.

The discipline of a scheduled refresh is what keeps a pay plan maintainable rather than perpetually patched. A structure that is rebuilt on a known cycle is one the organization can defend and run without a consultant on retainer, which is the entire point of building it correctly in the first place. That maintainability is the same principle behind a sustainable classification framework: a structure the jurisdiction can carry forward on its own is worth more than an elegant one it cannot maintain.

A pay structure with no scheduled refresh is not a structure. It is a snapshot that gets less accurate every year until someone mistakes its age for a budget problem instead of a design problem.

How Pinnacle designs a pay structure

Pinnacle designs pay ranges under the firm's four-layer reference framework, so the structure holds up in front of an elected body, an auditor, or a grievance hearing. Federal and industry authority anchor the method, house discipline sets the numbers, and statistical rigor tests the result before it is delivered.

Federal authority comes first. The Office of Personnel Management Position Classification Standards and the federal General Schedule provide the reference logic for grade relationships and the principle that structure follows the evaluated level of the work, not the incumbent. Fair Labor Standards Act exemption status is confirmed for every grade so the structure does not embed a misclassification. Industry methodology comes second: the WorldatWork total rewards framework for pay structure design and market pricing, and IPMA-HR public-sector classification guidance for how ranges map to job families in a government setting.

Pinnacle's house standard comes third, and it is where the numbers get set: range width in the 35 to 50 percent band, midpoint progression in the 8 to 12 percent band, consistent grade overlap, market-anchored midpoints, and a documented decision rule for every deviation from the standard bands. The house discipline includes a refusal to solve a single hard placement with a one-off exception that quietly breaks the structure for everyone else. Every deviation is written down and justified, because an undocumented exception is a future grievance.

Testing the structure for pay equity

The fourth layer is statistical, and it runs after the structure is built but before it is adopted. A pay structure can be internally coherent and still produce disparate outcomes once real incumbents are placed into it, so Pinnacle tests placements against protected-class distribution using EEOC pay analysis frameworks and OFCCP compliance review standards. Where the data supports it, a multivariate regression isolates whether pay differences track to legitimate factors like tenure, credentials, and evaluated grade, or whether an unexplained gap correlates with a protected characteristic.

This test is not optional rigor. A structure delivered without it can hand a jurisdiction a defensible-looking pay plan that produces an indefensible pay equity profile the moment someone runs the numbers. Running them first, inside the engagement, is cheaper than having them run later by a plaintiff-side expert. The structure that survives the equity test at design time is the one that survives it in litigation.

The through-line

Range width, midpoint progression, and refresh cadence are three decisions that look technical and are actually strategic. They decide whether a pay plan is a durable asset the organization can run for a decade or a depreciating snapshot that has to be renegotiated every budget cycle. The width sets the room to grow. The progression sets the meaning of a promotion. The refresh keeps both tethered to reality. Designed together, under a documented method, they produce a structure that answers the questions an auditor, a council, and a hearing officer will eventually ask. Designed carelessly, they produce compression, exceptions, and a pay plan no one can defend.

← All insights