Salary Bands: Inside the Corporate Pay Grid

Last spring, a senior analyst at a Cleveland health system opened a spreadsheet that a colleague had forwarded by mistake. Her own name sat in the fourth row, beside a salary range her employer had never shown her: $82,000 to $123,000, with a midpoint of $102,500. She had been earning $86,400 for two years — about 3 percent above the bottom of a band she did not know existed, and 16 percent below its middle.

The analyst's situation is common, and it is changing. Nearly every employer with more than a few hundred workers organizes pay into bands: ranges with a minimum, a midpoint, and a maximum that define what a role is worth. Most employees never see the grid, yet the grid governs their raises, their promotions, and much of what they earn. This article explains how bands are built from market data, what the midpoint really means, and what to do below it, at it, or above it. The short version: your salary is not a verdict on your worth. It is a position inside a range that someone built, and positions move.

Where the numbers come from

Bands begin with market data, not opinions. Every year, compensation consultants at firms such as Mercer, Willis Towers Watson, and Radford run salary surveys in which thousands of employers submit pay for matched jobs — same title, same duties, same level — and receive back percentile tables. A company pricing a marketing-manager role reads off the 25th, 50th, and 75th percentiles of what comparable firms pay for that work. Public sources offer a rough cross-check: the Bureau of Labor Statistics publishes median wages for more than 800 occupations, and sites like Payscale and Glassdoor collect millions of self-reported salaries from employees checking where they stand.

Most companies price only a handful of benchmark jobs directly against those surveys. Every other role is fitted to a regression line — a curve drawn through the survey points that assigns a market value to any job, however unusual the duties. This is why two titles that sound nearly identical can land in different bands, and why the same title spans enormous pay differences across companies. Smaller employers, which cannot afford survey subscriptions, price against published data and advertised ranges — one reason their bands drift from the market. The band is a statistical artifact, assembled from thousands of data points, none of them about you personally.

The percentile choice is a policy decision, usually a visible one. A company that wants to attract stronger talent prices roles at the 75th percentile of the market; one that competes on brand or stability may target the 50th. On a role with a $100,000 midpoint, the two philosophies differ by roughly $10,000 — before anyone negotiates. Recruiters notice the gap, which is why the same job can be advertised at sharply different ranges in the same city.

"The band is a market statistic with your name on it," said a compensation consultant in Philadelphia who has designed pay structures for about 40 companies. "Companies pick a percentile the way they pick a budget line."

The anatomy of a band

Once the market value is set, the band takes shape. The midpoint is the anchor — the market rate the company intends to pay a fully competent performer. The minimum and maximum are derived from it using a spread, the distance from bottom to top expressed as a percentage of the minimum. For most professional roles the spread is about 50 percent, which means the maximum sits 50 percent above the minimum and the midpoint sits exactly between them. A $100,000 midpoint therefore produces a band of about $80,000 to $125,000. Entry-level and support roles run narrower, often 30 to 40 percent; executive bands run wider, sometimes 80 percent or more, because a chief financial officer's contribution spans far more than a coordinator's.

Bands are grouped into grades — the job architecture of the company — so that similar roles share a grid. A company might maintain a dozen grades, each band overlapping the one above it. The overlap is what lets a strong performer in grade six earn more than a weak performer in grade seven without a promotion. When someone is promoted, the move usually lands them at the minimum of the new band or at a 5 to 10 percent increase, whichever is higher. That policy — promotion to the minimum — is why so many promoted employees find themselves at the bottom of a new range, and why the next raise conversation starts from a different place.

Illustrative bands from a midsize company's grade structure, 2025 survey data.

Role	Grade	Minimum	Midpoint	Maximum
Customer service representative	4	$41,000	$51,000	$64,000
Staff accountant	6	$58,000	$73,000	$91,000
Senior software engineer	9	$110,000	$137,000	$172,000
Director of marketing	11	$150,000	$188,000	$235,000

The table is illustrative, but the shape is real: the midpoint is the number the company manages to, and the spread is the room to grow within a grade. Bands also age. Survey data are typically refreshed once a year, and companies reprice bands on that cycle, which is why a band can look stale by late winter though it was current in the fall.

Compa-ratio: the number that governs your raises

Compensation professionals rarely talk about your salary in dollars. They talk about your compa-ratio — your pay divided by the midpoint of your band. A compa-ratio of 1.0 means you are paid exactly the market rate for a fully competent performer. A ratio of 0.9 means you sit 10 percent below it; 1.1 means 10 percent above. The normal working range runs from about 0.8 to 1.2, and where you sit inside it determines what happens at raise time.

New hires usually land between 0.8 and 1.0, below the midpoint, because the company takes less risk on an unproven employee. Someone below 0.8 is a green circle in the jargon — paid under the band, common for recent graduates in training. As tenure and performance accumulate, raises push the ratio toward 1.0, where a fully competent performer is meant to sit. Above 1.0 the arithmetic changes, because merit budgets are small: most companies budget about 3 to 4 percent of payroll for raises each year, according to Mercer and Willis Towers Watson budget surveys, and a raise that keeps a 1.1 performer competitive is a different decision from one for a 0.85 performer.

Consider two analysts in the same band, midpoint $90,000, each rated solidly. The first has a compa-ratio of 0.85 and earns $76,500. The second sits at 1.1 and earns $99,000. The manager's raise pool is 3 percent of payroll. The first analyst can plausibly receive 5 percent — it closes a market gap and costs little. The second analyst at 5 percent would land at $103,950, nearly 16 percent above the midpoint, where the band's logic stops supporting raises. The manager gives the second analyst 2.5 percent and calls it a good year. Same review. Same band. Different numbers, and the difference is arithmetic, not favoritism.

Where bands break

This is the quiet part of pay that employees rarely hear. Raises shrink as the ratio climbs, which is why two coworkers with identical performance reviews can receive different increases — and why the employee who job-hopped into a higher band often keeps outpacing the one who stayed. The system is not punishing loyalty; it is enforcing a range.

Bands break in predictable places, and the most common is compression. When the market moves faster than the annual refresh, new hires are priced at the current market while existing employees sit on bands a year or more old. A 2023 hire can earn more than a 2021 hire in the same grade; a new graduate can outearn a tenured specialist; a manager can earn less than the people they supervise. Compensation surveys and pay-equity analyses find this pattern across industries, and it is usually invisible until someone leaves — and the replacement is hired at the new rate.

Internal equity audits are the employer's answer. Many companies run them annually, comparing pay across employees in the same band, same grade, and similar tenure, looking for gaps that market pricing alone cannot explain. The audits find real problems — women and minority employees overrepresented below the midpoint in some firms, hires from hot markets paid far above tenured peers — and correcting them is one reason market adjustments exist as a separate budget line from merit raises. When an audit turns up a compa-ratio of 0.82 next to a peer at 1.18 doing the same work, the fix is usually a midyear adjustment, not a lecture about performance.

Red circles and market corrections

What happens when an employee's pay passes the top of the band? In the jargon, they are red-circled: the position is paid above its maximum, and the company treats the excess as a problem to manage. Some employers freeze the salary until the band catches up; others cap raises and pay a lump sum instead; a few quietly stretch the band when the market moves. The most common fix is the market correction — a midyear adjustment, outside the raise cycle, that resets a band or a salary when survey data show the company has fallen behind.

Market corrections are why the annual raise is no longer the only moment pay changes. Compensation-platform data show a rising share of pay movement happening off-cycle: sign-on bonuses, counteroffers, spot adjustments for scarce skills, and promotion bumps. The practical consequence: an employee who waits for the calendar review to discuss money is negotiating against a system that no longer moves on a calendar.

The top of the band is also where counteroffers happen. A resignation from someone paid at or above the maximum forces a different calculation: the company can break its own grid to keep them, or watch them leave. Retention data from staffing and outplacement firms show that a large share of counteroffers go to people already paid near their band's ceiling, because that is where the market pressure is most visible. Whether the counter is worth accepting is another question — Robert Half surveys find many who accept one leave within a year anyway — but the counter itself is proof that the band bends when leaving is credible.

What transparency laws changed

The band was once a guarded document. That has changed. Colorado began requiring employers to post salary ranges on job listings in January 2021; New York City followed in November 2022, and California and Washington in January 2023. More than a dozen states now have some version of a pay-range law, and the newest rules took effect in January 2026. What companies post is not always the full band: many publish an expected range for the specific role, which can be narrower than the band, and some post ranges wide enough to cover anyone. SHRM surveys and analyses by the Economic Policy Institute suggest the laws have pushed ranges into the open without collapsing the internal structure that produces them.

The laws also changed what workers ask. Recruiters in pay-transparency states say the first question from candidates is now routinely about the range — and that candidates who ask for it before discussing expectations get better information than those who answer the expectations question first. The range is a floor for the conversation. Knowing the band's midpoint before you name a number changes the negotiation in ways that favor the candidate. In interviews, the range question also signals something about the employer: companies that answer it directly tend to have bands they are willing to defend, and companies that deflect usually have reason to deflect.

What to do with your position

If you sit below the midpoint, the case is mostly arithmetic. Find your band — ask your recruiter, your manager, or human resources; many employers share the range on request, and California law and a growing list of states require it. Compute your compa-ratio, then make the market case: your pay relative to the midpoint, the survey data for your role, and what it would cost the company to replace you. Timing matters. Raises and market adjustments are decided during budget season, usually in the fall, so the conversation belongs in late summer — not in March, when the pool is already spent.

If you sit at or near the midpoint, the raise conversation is harder, because the system is working as designed. The lever is not a bigger merit increase but a different slot: a promotion to a higher band, a title change into a role priced against a different market, or a market adjustment if your duties have grown beyond the band's description. A recruiter in Austin who has negotiated more than 200 offers put it plainly: the midpoint is where companies stop paying for tenure and start paying for the role. If the role has not changed, the pay will not either.

If you sit above the midpoint — a compa-ratio above 1.1 — the realistic moves are promotion, equity, bonuses, and flexibility, not base-salary increases. Asking for a raise when you already earn above market invites a polite no; asking for the next grade invites a conversation. And if an employer refuses to show you the band at all, that is information too. Companies that publish ranges internally tend to manage pay more deliberately, and the data suggest their employees negotiate more successfully. Secrecy, by contrast, usually protects a grid the company does not want examined.

Run the numbers before you speak

Whatever your position, the practical tool is the same: run the numbers before you speak. The site's salary calculators turn an annual figure into an hourly one and back again, and the salary inflation calculator shows what a 2020 salary is worth in today's dollars. Then bring the compa-ratio into the conversation. "I'd like to understand where this offer sits relative to the midpoint" is a sentence that works in an interview, at a review, and at a counteroffer, because it asks for information rather than a favor.

The ask itself should be a question about information, not a demand. The compensation consultant in Philadelphia advises clients to frame every salary conversation around the grid: Where does this role sit in the band? What compa-ratio does this offer represent? What would it take to get to the midpoint? Each question forces the other side to explain the band, and people who explain the band are people who can change it. He has watched that script work from first-job offers to executive negotiations, and the reason is simple: bands are built on data, and data can be argued with. A number pulled from thin air cannot.

Where pay is heading

Three forces are pushing bands further into the open. Pay-transparency laws keep spreading, state by state. Job-posting sites display ranges by default, making the band a public fact rather than an internal one. Employers, facing retention pressure, are experimenting with wider career bands that let people grow in pay without a title change — a direct answer to the complaint that the grid makes raises impossible above the midpoint.

The mechanics are changing too. Compensation software now prices jobs continuously, pulling from live market feeds rather than once-a-year surveys, and several large employers say they are moving to career bands that span several grades. The direction of travel is the same everywhere: the grid is becoming less rigid, more visible, and harder to hide behind. For employees, that is good news. A system that runs on published data is a system you can prepare for.

The Cleveland analyst who found her band by accident asked for a market adjustment last fall and received it. Her compa-ratio moved from 0.84 to 0.92. She still does not know who built the grid, or why the spreadsheet reached her desk. But she knows where she sits inside it now, and that has turned out to be most of the battle. The salary number on an offer letter is a position, not a verdict. Positions move.