A Guide to Compensation Percentiles for Supporting Fair Pay

Every pay decision inside your organization is, at its core, a statement about where you want to stand in the market. If pay is too low, you risk losing your top employees. If it's too high without a clear compensation strategy, you can end up straining your budget and later making cuts that hurt trust.

So instead of thinking about pay in isolation, it all comes back to one key choice: the compensation percentile you aim to match.

This guide covers everything you need to know about compensation percentile. You will learn what they are, how to find reliable market data, how to build them into your pay structures, how to keep pay fair and equitable, and how to apply all of this throughout the employee lifecycle.

Chapter I. What is a Compensation Percentile?

A compensation percentile is a simple way to see how a salary compares to others in the same role. It tells you whether pay is lower than, about the same as, or higher than what the market offers.

Think of it as a ranking. A salary at the 50th percentile sits right in the middle. Half of workers in that role earn more, and half earn less. A salary at the 90th percentile is higher than 90% of salaries for that role, with only 10% paying more.

Most companies look at four key points: the 25th, 50th, 75th, and 90th percentiles. Together, these give a clear picture of where pay stands and where adjustments may be needed.

And to get these numbers right, more and more organizations today use CompAnalyst® Market Data, giving HR and compensation teams instant access to percentile benchmarks across thousands of roles, industries, and locations to support smarter, more defensible pay decisions.

1.1 Common Percentile Benchmarks and What They Mean

The U.S. Bureau of Labor Statistics describes percentile measures as a useful tool for evaluating employee compensation across different jobs and industries.

The table below shows the most commonly used compensation percentiles and how they are typically interpreted.

Percentile	Meaning	Common Use in Compensation
P25 (25th percentile)	25% of the market earns less and 75% earns more	Often used as the lower end or minimum of a pay range
P50 (50th percentile / median)	Half the market earns less and half earns more	Most common market reference point for compensation decisions
P75 (75th percentile)	75% of the market earns less and 25% earns more	Used by organizations aiming to pay above market median
P90 (90th percentile)	90% of the market earns less and 10% earns more	Used for lead-the-market strategies in high-demand roles

1.2 How Percentiles Translate into Real Pay Numbers

Understanding percentiles alone doesn't tell you the actual salary amount for a role. That number comes from market pricing, which is the process of matching internal jobs to external salary data.

1.2.1 The Three Steps of Market Pricing

• Job matching: A compensation analyst aligns an internal job to the closest equivalent benchmark job in one or more salary surveys, using job descriptions, scope of responsibility, and seniority level as matching criteria.

• Data extraction: The analyst pulls the P25, P50, and P75 values for base salary or total cash from the matched survey job.

• Survey aging: Because survey data is typically collected 6 to 18 months before publication, the analyst applies a salary aging factor, a compound annual growth rate (CAGR) based adjustment, to bring values forward to the current effective date.

1.2.2 Two Key Metrics: Compa-Ratio and Range Penetration

Once pricing is complete, two supporting metrics show where an individual stands relative to the market:

• Compa-ratio

This is calculated by dividing an employee's pay by the midpoint of their pay range. The result tells you exactly where their pay stands relative to that midpoint.

A compa-ratio of 1.0 means pay is right at the midpoint. Anything below 1.0 means pay falls short of the midpoint, and anything above 1.0 means pay exceeds it.

• Range penetration

It shows where an employee's pay sits within the full span of a pay range, from minimum to maximum. While compa-ratio measures pay against the midpoint, range penetration looks at the entire band, making it easier to see how much room remains before an employee reaches the top of their range.

This makes it especially useful in broadband structures, where ranges are wide and the distance between minimum and maximum is significant

Chapter II. How to Find and Use Market Pay Data

A compensation percentile is only as reliable as the data behind it. The quality of your salary survey data determines whether pay decisions are grounded in reality or built on noise.

This chapter covers where market data comes from, how to evaluate it, and how to combine multiple sources into a single, defensible market reference point.

2.1 Where Market Compensation Data Comes From

Most organizations use a mix of sources, since no single one covers everything. Here's what each offers:

Commercial surveys (e.g., CompAnalyst® Market Data): The most trusted option. Employers submit pay data directly, giving you accurate, detailed results by job level, industry, and location. Requires a paid subscription.

Government data (e.g., BLS/OES): Free and credible, but typically 12 to 18 months behind the market and less detailed. Best used to validate other sources, not as your primary benchmark.

Job posting data: Reflects what employers are actively offering right now. Pay transparency laws have made this more useful, though range quality varies by employer.

Online salary platforms: Easy to access, but data is self-reported and unverified. Useful for a general directional read, not precise enough to rely on alone.

Industry association surveys: Trade organizations collect member-only data with strong sector specificity, a good complement for specialized industries.

2.2 How to Combine Multiple Data Sources into One Reference Point

Relying on a single salary survey introduces risk. Sample composition, participating companies, and data collection timing all vary by provider, meaning any one source may not fully represent your actual talent competitors.

Using multiple surveys and combining them into a single reference point produces a more balanced and defensible result.

2.2.1 Three Ways to Blend Survey Data

The Market Reference Point (MRP) is the blended percentile value used as the anchor for setting pay range midpoints. The more data points included in the blend, the more reliable and defensible the final reference point becomes.

Organizations typically use three common approaches to arrive at an MRP:

• Simple Average: This calculates the mean of P50 values across two or more surveys. It is the easiest approach and works well when the surveys are similar in scope and sample size.

• Weighted Average: This gives more weight to surveys with larger sample sizes or stronger industry relevance. It produces a more precise result but requires a clear rationale for how the weights were assigned.

• CAGR Aging Before Blending: This ensures all surveys are updated to the same date before combining them. Skipping this step is one of the most common reasons pay ranges become outdated.

2.2.2 Using Trend Lines to Build Pay Grade Structures

When organizations update a full salary grade structure, they often use regression analysis. This involves plotting market P50 values against job levels or point scores and drawing a trend line, often called the market line.

This helps compensation teams set consistent midpoints for all grades, even for roles without direct market data.

Chapter III. Building Pay Structures Around Percentile Anchors

Pay ranges, grade structures, and broadbands are the building blocks of a well-run compensation program. This chapter covers how percentile anchors fit into those structures and why the design choices matter.

3.1 Pay Grades, Ranges, and Where Percentiles Fit In

A pay grade groups jobs of similar market value into a single administrative band. The pay range attached to each grade defines the minimum, midpoint, and maximum base salary for all jobs in that grade.

To make this process easier, CompAnalyst® Software helps compensation teams build and manage pay structures using integrated market data.

3.1.1 How Percentiles Map to Pay Range Minimums, Midpoints, and Maximums

• Midpoint is set at the MRP or P50 for the grade. The market median is the structural center of the pay range.

• Minimum is often set near P25, reflecting entry-level or developing performers within the role.

• Maximum is often set near P75, representing a fully experienced, high-performing employees at or above market.

3.1.2 Choosing the Right Range Spread by Job Level

A typical salary range spread is around 30% to 40%, but the right range depends on the role, level, and industry. In practice, higher-level roles often use wider ranges:

• Administration and Operations: 40% or more

• Professional and Management: 50% or more

• Executive roles: about 50% to 65% or more

In general, the more senior or complex the role, the wider the pay range tends to be.

3.1.3 When Employee Pay Falls Outside the Range

Here are two special conditions that warrant documented action plans:

Red-circle pay occurs when an employee's pay is above the pay range maximum. This is common after a job re-evaluation to a lower grade or when the market has declined. The standard response is to freeze base salary increases until the range catches up.

Green-circle pay occurs when an employee's pay is below the pay range minimum. This requires a prompt adjustment to bring the person to minimum, as it represents both a legal risk in pay equity audits and a retention risk.

3.2 Determining a Job's Internal Value Before Comparing to the Market Rate

A compensation percentile describes where a job sits in the external market. Before that external question can be answered, however, the internal question must be resolved first: what is the relative worth of this job inside the organization? That is the function of job evaluation.

3.2.1 Common Ways to Evaluate a Job's Worth

Here are some common methods organizations use to evaluate job worth:

• Point-factor method assigns numerical weights to factors such as knowledge, problem-solving scope, accountability, and working conditions. This is the most widely used approach in large organizations.

• Whole-job ranking is simpler but less defensible for large or complex job populations.

• Commercial frameworks are embedded in many enterprise HR systems and align directly with those firms' salary survey data. This creates an end-to-end connection from job evaluation score to external market percentile.

3.2.2 Job Levels and Career Paths as a Foundation for Pay Fairness

Job leveling is a way to apply job evaluation consistently across the entire organization. It creates a clear career structure that defines how roles progress over time.

This structure includes levels for individual contributors (IC1 to IC6) and managers (M1 to M4), showing different stages of responsibility and seniority.

Each level is tied to a pay grade, and each grade has a pay range based on market data and percentiles. With a clear leveling system in place, organizations can make fair and consistent pay decisions across different teams, locations, and job types.

3.3 Broadbands and Other Pay Structure Options

Some organizations, especially those with flat hierarchies or flexible work cultures, use broadbanding. This approach combines many traditional pay grades into a few wide salary bands.

Even in broadband structures, market percentiles are still used to guide pay decisions within each band. However, strict grade boundaries are removed, giving managers more flexibility to reward skills, performance, and experience.

The main tradeoff is that pay can become less structured over time. Without clear internal limits, salaries may spread unevenly, which can create pay equity challenges.

This is where ongoing analytics become critical. CompAnalyst® Pay Equity Suite allows compensation teams to continuously monitor internal pay distribution, surface disparities across employee groups, and take corrective action before gaps become systemic.

To manage this effectively, organizations also need strong manager training and reliable data tools to ensure pay decisions within each band remain consistent, defensible, and fair.

Chapter IV. Pay Fairness, Legal Requirements, and Talent Strategy

A well-built compensation structure does more than organize pay. It needs to support fairness, meet legal requirements, and hold up at every stage of the employee lifecycle, from the first offer to long-term retention.

4.1 Using Market Data to Support Pay Equity

Pay equity is the principle that employees performing the same or comparable work should be paid fairly regardless of gender, race, ethnicity, age, or other protected characteristics.

Percentile-based pay structures are a foundational tool for achieving this, because they create an objective external reference against which individual pay decisions can be audited.

4.1.1 How a Pay Equity Analysis Works

A pay equity analysis generally follows three steps.

Step 1: Group employees into comparable clusters

Employees are organized by grade, job family, and level to identify who is doing comparable work.

Step 2: Run a regression model

The model tests whether any unexplained pay differences are linked to protected characteristics such as gender or race, after accounting for legitimate factors like tenure, performance, and location.

Solutions like CompAnalyst® Pay Equity Suite can run this regression analysis automatically, flag disparities across protected groups, and help compensation teams move quickly from detection to action.

Step 3: Address the gaps

If a systematic pay gap is found, a remediation plan is developed with clear owners and timelines to resolve it.

4.1.2 How Pay Transparency Laws Affect Pay Strategy

Pay transparency laws in California, Colorado, New York, Illinois, and a growing number of other states now require employers to post salary ranges in job listings.

For organizations without clear, percentile-based pay structures, this creates two immediate problems. Current employees may raise concerns when posted ranges do not match their own pay. Job seekers may look elsewhere when ranges appear lower than what competitors offer.

Adoption is growing quickly. Around 60% of organizations now publish pay ranges in job postings as of 2024, up from 45% in 2023, driven by both legal requirements and candidates who expect pay transparency from the start.

The gender pay gap makes this even more pressing. Pew Research Center data shows that in 2024, women earned about 85 cents for every dollar men earned, based on median hourly wages across full-time and part-time workers.

Organizations without structured pay ranges are less equipped to spot these gaps, address them, and explain their pay decisions with confidence.

4.2 Looking Beyond Base Pay: Bonuses, Equity, and Total Rewards

Base salary percentile alone does not show the full picture of how competitive a compensation package is. To get an accurate view, organizations need to look at total compensation.

4.2.1 Total Cash and Total Direct Compensation as Benchmarking Levels

Total cash compensation (TCC) refers to base salary plus annual incentives such as bonuses and short-term incentive payouts.

Total direct compensation (TDC) refers to TCC plus the expected value of long-term incentives such as equity grants, restricted stock units, or stock options.

Salary surveys publish percentile data at both levels. If your organization benchmarks at the base salary level while competitors are benchmarking at TDC, you may be working with an incomplete picture.

This can lead to offers that are lower than the market, particularly in roles like sales, executive, and investment management, where variable pay makes up a significant portion of total earnings.

4.2.2 Annual Pay Increases and Budget Constraints

Merit cycles are the main way individual pay moves within a grade over time. Most organizations use a merit matrix to decide how much each employee receives, based on two factors:

• Worker's performance rating

• Where their current pay sits within the range

A high performer paid near the bottom of the range receives a larger increase than a high performer already near the top. This helps move pay toward the market midpoint while keeping increases within budget.

Budget pressures are real. WorldatWork data covering nearly 17 million employees across 22 countries shows that average salary increase budgets held at 4.1% in 2024, with a slight decline projected to 3.8% in 2025.

4.3 Applying Market Data Throughout the Employee Lifecycle

Percentile data delivers the most value when applied consistently at every stage of the employee journey, not just during annual pay reviews.

4.3.1 Using Market Position to Guide Job Offers

Every offer is a practical test of whether your percentile targets are set correctly. Before extending an offer, benchmarking it against the market distribution helps confirm it is competitive. A candidate offered the pay range midpoint is being offered at market P50, while one offered near the third quartile is closer to P75.

This matters more now that salary history ban laws are in effect across most U.S. states. When past pay cannot be used as a reference, the percentile-based pay range becomes the primary anchor for the offer conversation.

4.3.2 Spotting and Responding to Retention Risk

Employees whose pay has fallen below P25 compared to peers in similar roles are at a higher risk of leaving. A retention risk analysis helps identify these individuals early, so adjustments can be made before a resignation happens.

The cost of inaction is real. One study found that 73% of employees would consider leaving for higher pay. While counteroffers, retention bonuses, and off-cycle increases can help, they are often more costly and disruptive than keeping pay competitive from the start.

4.3.3 Managing Pay Across Remote and Multi-Location Teams

Organizations with distributed teams need to decide how to handle geographic pay differences, whether to use a single national rate, local market rates, or a tiered geo-pay model.

The right benchmark is cost of labor, not cost of living. Cost of labor reflects actual talent supply and demand in a given area, while cost of living measures housing and consumer prices, which are not the same thing.

For organizations managing pay across multiple locations, CompAnalyst® Global Market Data provides location-specific compensation benchmarks across thousands of geographies, making it easier to set accurate, defensible pay ranges for every market without relying on fragmented or outdated local data.

Chapter V. FAQs

Getting compensation right is not a one-time exercise. It takes the right data and a clear philosophy, applied consistently across your organization. The sections below answer some of the most common questions organizations have as they put all of this into practice

5.1 How does compensation percentile differ from average salary?

A compensation percentile shows where a pay value falls within the full range of pay rates for a role. An average salary, on the other hand, is simply the sum of all pay rates divided by the number of employees.

5.2 How often should organizations refresh their percentile data?

At minimum, review your percentile data once a year, ideally before the annual merit cycle. Most commercial salary surveys publish annually, making this a natural checkpoint.

For fast-moving roles in technology, AI, and cybersecurity, an annual review may not be enough. A mid-year check is a practical necessity for these job families. Real-time data tools can also help monitor market movement between formal survey cycles.

5.3 How do you choose the right salary percentile to target?

Start with your pay philosophy. Most companies paying the 50th percentile (market median) aim to stay competitive without overspending. Some target the 65th to 75th percentile for high-demand roles, while others use the 25th percentile for lower-priority positions.

Match your target to your budget and business needs. Hard-to-fill or critical roles often justify higher compensation percentiles; common or entry-level roles may stay closer to the median.

Finally, compare current pay against market data and adjust over time to keep your structure aligned.

5.4 What do percentiles in a salary survey mean?

Salary survey percentiles show what employers are actually paying for a specific job in a given industry and location. They are useful because they reflect real market behavior, not just averages. Use them as a guide for where your pay stands today and where you may need to adjust.

5.5 Do all jobs need to be anchored to the same percentile?

No, and they probably should not be. Different roles carry different levels of risk, scarcity, and strategic value. The goal is not to apply one compensation percentile across the board but to define a clear compensation philosophy and use it consistently. That way, your decisions stay fair and easy to explain, even when the percentiles vary by role.