Why BLS Aggregate Occupations Across Industries (and When That Matters for Your Salary)

The BLS Occupational Employment and Wage Statistics program assigns every worker in the country to one of roughly 800 detailed occupations using the Standard Occupational Classification system. A software developer writing trading algorithms at a hedge fund and a software developer building internal tools at a regional hospital both land in SOC 15-1252. The national median for that single code — $135,980 (BLS OEWS, SOC 15-1252, May 2025 estimates, published May 2026) — is the number most salary articles quote, and it is accurate as far as it goes. What it does not tell you is which industry is pulling that median up and which is pulling it down, or where you personally sit relative to the industry you are about to enter or leave.

Understanding why BLS structures data this way — and where to look for the industry signal it does not surface directly — changes what you should put on the table in your next negotiation.

Why BLS Collapses Industry Into Occupation

The OEWS program is designed to answer a workforce-planning question, not a personal salary question. Policymakers and economists need consistent, comparable headcounts: how many software developers exist, what do they collectively earn, and where are they concentrated geographically? Splitting every occupation by industry would produce hundreds of additional cells, many too small to publish without suppression rules that would make the data even harder to use.

The result is a system that is excellent for benchmarking an occupation against the full labor market but requires extra steps when your real question is industry-specific. Those extra steps are worth taking before any negotiation.

What the Aggregated Number Actually Tells You

The national SOC 15-1252 (Software Developers) row from the May 2025 OEWS release shows:

• 10th percentile: $82,460
• 25th percentile: $105,210
• Median (50th): $135,980
• 75th percentile: $171,980
• 90th percentile: $214,670

All figures: BLS OEWS, SOC 15-1252, national, May 2025 estimates, published May 2026.

The spread from the 10th to the 90th percentile — $82,460 to $214,670 — is wide enough to contain at least two distinct labor markets inside one SOC code. A developer near the 10th percentile is almost certainly working in a lower-paying industry or geography. A developer near the 90th is almost certainly in a high-paying industry, a high-cost metro, or both. The aggregate median tells you neither story on its own.

For a deeper look at what the spread between percentiles actually signals, the guide on reading between BLS percentiles walks through why the 25th-to-75th range is often the more actionable number.

The Industry Premium You Cannot See Directly

BLS does publish industry-specific OEWS tables — cross-tabulations of occupation by NAICS sector — but they are not the default view most people encounter, and they are not available for every occupation-industry combination due to sample size thresholds. When those tables are available, they confirm what practitioners already know: the same SOC code pays differently depending on the sector employing it.

The mechanism is straightforward. Finance and insurance firms generate revenue per employee at a scale that manufacturing or healthcare systems typically cannot match, so they can pay a premium for the same technical skill set. That premium is real, persistent, and large enough to matter in a negotiation — but it disappears into the aggregate median the moment BLS rolls all industries together.

This is not a flaw in BLS methodology. It is the correct output for the question BLS is answering. The flaw is in treating an occupation-level median as an industry-specific benchmark when you are moving between sectors.

A Worked Example: Moving From One Context to Another

Suppose you are a software developer currently employed in a healthcare organization, considering an offer from a financial services firm. The national aggregate median of $135,980 (BLS OEWS, SOC 15-1252, May 2025) is the number your current employer is most likely to cite as the market rate. It is also the number the financial services firm may cite if they want to anchor low.

The more useful question is: where does the financial services firm's offer sit in the full percentile distribution for this occupation? If the offer is at $150,000 and the 75th percentile nationally is $171,980, the offer is between the median and the 75th — plausible for a finance employer paying an industry premium, but still below the top quartile. That gap is negotiable, and you now have a specific, citable floor to work from rather than a vague sense that finance pays more.

You can compare the full percentile range for Software Developers against other computer occupations to see exactly where any offer lands relative to the published distribution — which is the starting point for a defensible counter.

The same logic applies to any role that spans industries: accountants in public accounting versus corporate finance, project managers in construction versus technology, compliance officers in banking versus healthcare. The SOC code is the same; the industry premium is not.

What to Do With This Before You Negotiate

Three practical steps follow from understanding how BLS aggregates work:

1. Identify the direction of the industry premium. Before any negotiation involving an industry change, establish whether you are moving toward a higher-paying or lower-paying sector for your occupation. You will not find this in the aggregate OEWS table directly, but you can infer it from the percentile distribution: if the role and metro you are targeting are known to concentrate near the top of the distribution, an offer below the 75th percentile warrants scrutiny.

2. Anchor to the right percentile, not the median. The median is the midpoint of everyone in your SOC code across all industries. If you are targeting a higher-paying sector, the median undersells your position. Use the 75th percentile as your opening reference and explain the industry context explicitly.

3. Separate the occupation benchmark from the geography adjustment. Industry premiums and metro premiums compound. A software developer moving from a mid-tier metro to a financial hub in a high-cost city is picking up both adjustments simultaneously. The guide on the metro area trap covers how BLS geographic definitions can further distort the baseline you are working from.

None of this requires data that does not exist. The BLS OEWS percentile tables for SOC 15-1252 and every other detailed occupation are public, free, and updated annually. The discipline is in reading the aggregate for what it is — a cross-industry average — and adjusting your negotiation anchor accordingly.

Frequently asked questions

Does BLS publish separate salary data for the same occupation in different industries?

Yes, but not for every combination. BLS OEWS does publish cross-tabulations of occupation by NAICS industry sector, and these are available on the BLS website for many occupation-industry pairs. However, when employment in a particular cell is too small, BLS suppresses the estimate to protect confidentiality, so coverage is uneven. The national occupation-level tables are the most complete and consistently available starting point.

If the BLS median is a cross-industry average, why do employers use it as a benchmark?

Because it is the most defensible, publicly available number — and because it often serves the employer's interest to use the aggregate rather than the industry-specific figure, particularly when the employer's sector pays above the cross-industry median. Knowing this dynamic is part of negotiating effectively: the aggregate median is a floor, not a ceiling.

How much does the 90th percentile for Software Developers (SOC 15-1252) differ from the median?

According to the BLS OEWS May 2025 estimates (published May 2026), the national median for Software Developers (SOC 15-1252) is $135,980 and the 90th percentile is $214,670. The 90th percentile is the better reference point if you are targeting a high-paying industry, since the median includes all industries and pulls toward the center.

Can I use the occupation comparison tool to see where an offer lands in the distribution?

Yes. The Occupation Comparison tool lets you put two or three occupations side by side and view their full pay ranges — 10th through 90th percentile — in the same city, drawn from the same BLS OEWS data. It does not split by industry, but it shows you the full range of what the occupation pays nationally or in a given metro, which is the correct reference for evaluating where an offer sits.

Does this aggregation problem affect occupations other than software developers?

It affects any occupation that appears across industries with meaningfully different revenue structures. Accountants, financial analysts, compliance officers, HR specialists, and project managers all carry a single SOC code regardless of whether they work in banking, manufacturing, nonprofits, or government. The principle is the same: the aggregate median is accurate for the occupation as a whole and insufficient as an industry-specific benchmark.

This guide is informational only and does not constitute professional salary or career advice. Last reviewed: September 2026. Figures from U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics, May 2025 estimates, published May 2026.

Informational only, not professional or financial advice.