Benchmarking CNA and Nursing Assistant Pay Against BLS Data

The Resignation Letter That Names a Competitor Down the Road

The Director of Nursing at a skilled nursing facility reads her third CNA resignation letter in five weeks. The exit conversation is short and polite. The aide is not leaving the work. She is leaving for another building across town that pays somewhat more per hour, and she says so plainly.

Back at her desk, the Director opens the pay scale. The RN and LPN bands were reviewed in the last budget cycle. The CNA rows were simply carried forward, because the review always seemed like it could wait until next quarter.

This is a common sequence in SNF and long-term care settings, and it is fixable with public data. The Bureau of Labor Statistics publishes occupation-level wages for nursing assistants every year, free, by state and by metro area. What most facilities lack is a repeatable way to read that data and turn it into a pay band they can defend.

This article walks through a repeatable method for CNA wage benchmarking: which occupation code to use, which geography to trust, how to read the percentiles, and how to avoid the mistakes that make a benchmark misleading. You will leave with a process you can run in an afternoon.

Why Nursing Assistant Pay Gets Benchmarked Last

Registered nurses draw most of the attention in workforce reporting. The turnover headlines are about RNs: according to the 2026 NSI National Health Care Retention & RN Staffing Report, RN turnover was 17.6% in calendar year 2025, up from 16.4% in the prior-year (2025 report, CY2024) comparison. Three habits explain why CNA pay slips down the list.

• Pay scales are inherited. Aide bands are often set years ago and adjusted by a flat percentage at each budget cycle, which keeps the structure intact but quietly detaches it from the local market.
• Local competitors are the real benchmark, and they are hard to see. Nearby facilities and non-nursing employers compete for the same workers, and none of them publish their pay scales.
• The data feels harder to find. Widely quoted wage figures tend to cover RNs, but nursing assistants have their own occupation code in the BLS system.

Federal staffing rules add to the pressure. The May 2024 CMS final rule set 3.48 hours per resident day (0.55 RN plus 2.45 nurse aide) and 24/7 on-site RN coverage, and a December 3, 2025 interim final rule repealed both mandates effective February 2, 2026, while the facility-assessment requirement remains. Confirm current requirements with CMS and your state survey agency.

A Practical CNA Wage Benchmarking Method, Step by Step

The core of any benchmark is picking the right comparison. The steps below use only public sources and need no software.

Step 1: Use the right occupation code. BLS reports wages by Standard Occupational Classification (SOC) code. RNs are 29-1141 and LPN/LVNs are 29-2061. Nursing assistants are reported under SOC 31-1131. Confirm the code against the current BLS Occupational Employment and Wage Statistics (OEWS) occupation list before you pull data. If you employ medication aides, patient care technicians, or restorative aides, check which BLS category each title fits, and keep that mapping written down.

Step 2: Pull the current OEWS release. The BLS OEWS program publishes national, state, and metropolitan-area estimates, including mean wages and a percentile spread. Record the release year with every number you copy. A benchmark with no date is a benchmark nobody can audit. If you want a refresher on the tables themselves, our guide to reading BLS OES wage data covers the columns and the common traps.

Step 3: Choose state or metro, deliberately. We return to this in the next section, because it changes the answer more than most people expect.

Step 4: Convert units once. OEWS reports hourly and annual figures. The standard conversion is 2,080 hours per year, which you can verify with a figure from this data: the BLS lists the LPN/LVN national median at $62,340 a year, or about $29.97 an hour (May 2024 Occupational Outlook Handbook), and $62,340 divided by 2,080 is $29.97. Pick hourly for aides, since that is how you pay them, and convert everything else to match.

Step 5: Build the band. Place your midpoint against a chosen percentile, then set a minimum and maximum around it. As a worked example only, using a round placeholder wage and not a BLS figure: if your chosen reference wage were $20.00 an hour and you set the band at 90% to 110% of that midpoint, the band would run $18.00 to $22.00 with a $20.00 midpoint. Swap in the real figure from the release you pulled.

State Averages Versus Your Metro Area

The most useful single decision in this process is which geography you compare against. State figures are easy to find. Metro figures are closer to your actual labor pool.

A statewide average blends every market in the state into one number. The cleanest illustration in the public data is for RNs: according to O*NET local wage data for May 2025, the Ohio statewide RN average was $82,510, while the Columbus, OH metro average was $83,900. That difference is modest, but it shows the mechanism. A facility in Columbus that benchmarked to the statewide figure would be anchoring to a number that does not describe its own market.

The same logic applies to nursing assistants, and we are not publishing a CNA figure here because the exact numbers belong to the current release. Pull both the state and your metro for SOC 31-1131, then ask which one your recruits and your leavers actually come from. For a facility inside a metro area, use the metro figure as the primary benchmark and the state figure as context. For a facility outside any metro area, use the nonmetropolitan area estimate for your state, which BLS also publishes. Our deeper walk-through of state versus metro wage benchmarking covers the edge cases, including facilities near a metro boundary.

One more caution: OEWS covers nursing assistants across all employer types, not only SNFs. Hospitals, home health agencies, and other settings are blended into the number. Treat the BLS figure as the occupation's market reference, then adjust using local job postings and recruiter feedback.

Reading Percentiles and Choosing a Target

A mean is a starting point, not a pay policy. OEWS gives you a spread (10th, 25th, 50th, 75th and 90th percentiles), and the spread is where the decision lives.

To see what that looks like, consider RNs, where the numbers are published in our reference set. According to BLS OEWS (May 2025, national), the RN 10th percentile wage was $68,940, the 25th was $80,330, the median was $97,550, the 75th was $112,350 and the 90th was $137,470. An occupation average hides a wide band, and aides have their own spread to read the same way.

A simple way to choose a target:

1. Decide your position. Many SNF leaders target the median for the occupation, then move above it for night shifts, high-turnover units, or roles that take on extra responsibility. Others hold at the 25th percentile and compete on schedule and culture. Either is a legitimate choice. The mistake is making the choice by default.
2. Write the reason down, with the release year, so you can answer a CFO or administrator who asks why the aide budget moved.
3. Compare your actual wages to the band. List each aide's hourly rate against the band you built. The people sitting below the minimum are your resignation risk. The people above the maximum are a budget question, not a retention question.
4. Look at the ladder, not just the entry rate. A benchmark that ignores tenure can fix the starting rate and still lose your most experienced aides.

For context on how the neighboring occupations compare and how the gap affects your career-ladder design, see our explainer on the RN versus LPN wage difference.

Joining BLS Wages to O*NET Job Content

Wage data tells you what the market pays. It does not tell you what the job involves. That is where ONET helps. The BLS and ONET systems share the same SOC codes, so you can join them: the wage release gives you the price, and the O*NET occupation page gives you the tasks, skills, and knowledge that define the role.

This is useful when you are deciding which of your job titles belong in the same band. If your lead aide, restorative aide, and standard aide all map to the same SOC code but do materially different work, the O*NET task list gives you an outside reference for documenting the difference. Our overview of O*NET nursing occupations explains how to search it, and O*NET OnLine is the primary source.

O*NET Database by U.S. Department of Labor, Employment and Training Administration, used under the CC BY 4.0 license.

Keeping the Benchmark Current

A benchmark is a snapshot, so schedule the review rather than react to resignations. Ongoing CNA wage benchmarking works best as a fixed calendar item.

• Re-run the method each time a new OEWS release is published, and file the old version with its date.
• Check the aide band at the same meeting as your workforce plan. Wage decisions and staffing decisions affect each other, and our guide to SNF workforce planning shows where pay fits into the plan.
• Document sources and conversions so anyone inheriting the spreadsheet can reproduce every cell.

If you would rather not rebuild this each year by hand, the NursingWorkforce dashboard compares your pay bands against public BLS OES wage data and raises a wage-gap alert when a band falls behind. It works from the numbers you enter, holds no PHI, does not connect to your EHR or payroll, and does not touch scheduling. The spreadsheet method above works with or without it. For the broader approach across nursing roles, start with our BLS nurse wage benchmarking guide.

To get new benchmarking walk-throughs as they are published, subscribe to the NursingWorkforce newsletter. Each issue covers one practical method built on public data, with the arithmetic shown on the page.