Software developer salary: pay by experience, industry and metro

If you have a software developer offer in hand, or you're deciding what to ask for, the number to anchor on is the federal benchmark: software developers earned a median of $135,980 a year as of May 2025, according to the Bureau of Labor Statistics (BLS) Occupational Outlook Handbook. This guide shows where that figure comes from, what it leaves out (stock and most bonuses), which industries pay above it, and how to check whether your offer is low, typical or high.

Quick reference: software developer pay, May 2025

These are base wage figures from the BLS Occupational Employment and Wage Statistics (OEWS) survey for SOC code 15-1252, Software Developers. "Median" means half of developers earned more and half earned less. The 10th and 90th percentiles mark the bottom and top of the typical range: one in ten earned less than the first figure, one in ten earned more than the second.

BLS released the May 2025 estimates on May 15, 2026, and they are the newest OEWS figures as of this writing. OEWS estimates are produced once a year, so check the BLS OEWS tables page for a newer release before you negotiate.

What software developers earn nationally

The national median of $135,980 is the cleanest single benchmark you have. It covers developers working for employers of every size and type, across the whole country, in May 2025.

You'll also see a second number floating around: the mean, or average, of $148,100. The mean is higher than the median because a minority of very well-paid developers pull the average up. When someone quotes you "the average developer salary," ask which one they mean. For judging a single offer, the median is the fairer comparison.

A few things the national figure does not tell you:

• Your level. OEWS does not split pay by junior, mid-level and senior. BLS's own guidance notes that more experienced workers tend to earn more and that workers in the same occupation can carry very different levels of responsibility.
• Your city. National figures blend high-cost metros with low-cost ones. More on that below.
• Your full package. The figure is base wages only. Equity, year-end bonuses and benefits sit on top.

It also helps to know how big this job category is. With 1,687,890 workers in May 2025, software developers were the bulk of the broader "software and web developers, programmers, and testers" group, which had 2,150,380 workers. That large sample is good news for you: the national figures are based on a lot of data, so they're a solid starting point.

The percentile range, and what the top 10 percent earn

The spread between the 10th and 90th percentiles is wide for this job. In May 2025, the lowest-paid 10 percent of software developers made under $82,460, while the best-paid 10 percent made over $214,670.

Here's a rough way to read that range for your own situation. BLS does not publish pay by years of experience, so treat this as a practical guide, not an official ladder:

One recent change matters if you're looking at the very top of the market. Starting with the May 2025 estimates, BLS began publishing exact figures for many high percentiles that used to appear only with a footnote saying the wage was at or above $239,200 a year. That makes the top end of the OEWS tables more useful than in past years, especially in high-paying metro areas.

When the range is wide, be honest with yourself about where you sit. A new graduate pointing to the 90th percentile won't get far. A developer with years of experience in a high-paying industry can reasonably aim above the median.

Pay by industry

Who you work for moves the number more than many people expect. BLS publishes median wages for software developers in the industries that employ the most of them. Here are the May 2025 figures alongside each industry's share of software developer jobs:

What this shows in practice:

• Software publishers sit clearly at the top. This is the industry most people picture when they think of "tech companies": firms whose product is the software itself. Its median is well above the national figure.
• Most other big employers cluster around the national median. Manufacturing, corporate headquarters (that's what "management of companies and enterprises" means) and finance all land within a few thousand dollars of $135,980 at the median.
• The largest employer pays slightly below the median. Computer systems design, which includes IT consultancies, contractors and agencies, employs 29% of software developers but has the lowest median in the table, at $132,050.

So if you're comparing an offer from a consultancy with one from a software publisher, a gap in base pay is normal, not a sign that one of them is lowballing you. Compare each offer to its own industry's median first.

Keep in mind that these are medians for whole industries. A large software publisher and a small one can pay very differently, and industry medians don't capture equity, which can be a big part of pay at software companies. For named employers, see our company pay coverage.

Top-paying states and metro areas

Location is the other big lever. BLS publishes software developer wages for every state and for roughly 530 metro and nonmetropolitan areas in its OEWS tables, including medians and percentiles for each.

This guide doesn't print a state or metro ranking. The May 2025 detail for this occupation lives in BLS's interactive OEWS profiles, and rankings copied secondhand from other sites often mix data years or use averages instead of medians. Pulling the official figures for your own area takes a few minutes:

1. Open the BLS May 2025 occupation profiles and choose "Software Developers" (15-1252).
2. Look at the state and metro area tables, which show employment and wage estimates for each area.
3. Find your metro area, not just your state. Big metros can pay far more than the rest of the same state.
4. Note the median and the 90th percentile for your metro. Those become your local benchmarks.
5. For a full spreadsheet, the May 2025 state tables and metro tables are also available as downloads.

Weighing pay against cost of living

A higher local median doesn't automatically mean you're better off. BLS itself notes that workers in large urban areas with higher living costs are more likely to earn higher wages, which is partly why metro pay differs so much.

When comparing two cities:

• Compare each offer with that metro's OEWS median, not the national one.
• Look closely at housing costs, which can swing a lot from one metro to another.
• For remote roles, ask whether the employer adjusts pay by location. Some employers do, and the answer changes which benchmark you should use.

Our salaries by city section covers how pay and living costs compare across metros.

Base wage vs total compensation: what BLS does and does not count

This is the most important thing to understand before you compare an offer to BLS data. OEWS figures are straight-time, gross pay, before taxes, without premium pay. They are a measure of base wages, not of what you take home in a strong year at a company that pays in stock.

According to the OEWS definitions, here is what is in and out:

Two more details from BLS's methods: annual figures assume a full-time, year-round schedule of 2,080 hours, and the survey does not cover the self-employed. Freelance and independent contractor rates are not in these numbers.

Why this matters for developers

For many developers, especially at software publishers and large public companies, a big chunk of pay arrives as restricted stock units (RSUs), stock options, sign-on bonuses or annual performance bonuses. None of that shows up in the OEWS median. BLS does not publish total compensation figures for software developers that include equity.

That means:

• Compare base to base. Put your offered base salary next to the OEWS median and percentiles. Don't add your stock grant first.
• Value the rest separately. Then work out what the extras add per year, and judge them on their own terms.

How to annualize an offer (worked example)

Example only: say an offer letter lists a base salary, a target annual bonus stated as a percentage of base, a one-time sign-on bonus, and an RSU grant that vests over four years.

1. Start with the base salary. This is what you compare to OEWS.
2. Add the target bonus: base multiplied by the target percentage. Remember it's a target, not a promise.
3. Add the yearly equity value: the grant value divided by the number of vesting years. If the stock is not publicly traded, treat this value with caution.
4. Count the sign-on bonus in year one only, and check whether you'd have to repay it if you leave early.
5. Note the benefits separately: retirement match, health plan costs and paid leave all have real value, but they are not in OEWS either.

The result is your estimated year-one total. Offers with similar totals can look very different underneath, so keep the base comparison in front of you.

Related roles: QA analysts, web developers and programmers compared

If your job title is fuzzy, make sure you're benchmarking against the right occupation. BLS separates software developers from three close neighbors in the same May 2025 release:

All figures are May 2025, from the BLS Occupational Outlook Handbook (median and percentiles) and the OEWS May 2025 news release (mean and employment).

What the comparison tells you:

• Developers out-earn the neighboring roles at every point shown. Software developers lead on median, mean and both ends of the range.
• Industry patterns differ by role. For QA analysts and testers, manufacturing had the highest industry median in May 2025, at $129,930, ahead of software publishers at $101,070. For web developers, the information industry led at $119,940. For computer programmers, software publishers led at $111,710.
• The outlook differs too. BLS projects software developer employment to grow 10% from 2025 to 2035, while it projects computer programmer employment to decline 7% over the same period.

Your title doesn't decide which row applies; your duties do. If you design systems and build applications end to end, you're in the developer row even if your title says "engineer." If your work is mainly testing, writing test plans and finding defects, the QA row is a more honest comparison. Using the wrong row can lead you to under- or over-ask.

How to benchmark your own offer

Put it all together with this checklist:

1. Pick the right occupation. Match your duties to 15-1252, 15-1253, 15-1251 or 15-1254.
2. Place your base on the national range. Is it near the 10th percentile, the median or the 90th?
3. Check the industry median. A consultancy offer near $132,050 is typical for that industry. A software publisher offer at the same number is below that industry's median of $164,550.
4. Check your metro. Pull the OEWS figures for your metro area from the occupation profiles. This is often the biggest adjustment.
5. Check posted ranges. A growing number of states and cities, including Colorado, California, Washington and New York, require many employers to include pay ranges in job postings. Look up postings for the same role at the same company and at competitors. If you're in a state without such a rule, postings for the same employer's jobs in those states can still give you a useful signal. Rules on which employers are covered vary, so check your state labor department.
6. Value the extras separately. Annualize bonus and equity using the steps above.
7. Decide your ask. Pick a specific base number you can justify with steps 2 to 5.

Scripts you can use

At offer stage, when the base is below the local benchmark:

"Thanks, I'm excited about this role. Based on BLS data for software developers in this area and the ranges I've seen posted for similar roles, I was expecting a base closer to [your number]. Is there flexibility on base salary?"

When the base is fixed but the package isn't:

"I understand the base is set by band. Could we look at a sign-on bonus or a larger equity grant to close the gap?"

When the company asks for your expectations first:

"I'm looking for a total package that's competitive for this level in this market. Could you share the range budgeted for the role?"

None of these guarantee a yes. But a request backed by a specific, sourced number tends to get a more serious answer than "can you do better?" Our negotiation section has more scripts for both offers and annual reviews.

Common mistakes

• Comparing total comp to the BLS median. You'll think a typical offer is generous. Compare base to base.
• Using the mean as your benchmark. The mean runs above the median, which can make a fair offer look low.
• Ignoring industry. A consultancy and a software publisher are different pay markets.
• Using national data in a high-cost metro. Always check the local table.

What to do next

Write down three numbers before your next pay conversation: the national median ($135,980), the median for your industry, and the median for your metro from the OEWS tables. Then find two or three posted ranges for similar roles. If your offer's base sits below all of them, you have a clear, sourced case to ask for more.

For more on what different jobs pay, browse salaries by role, and follow pay trends for new BLS releases and pay transparency rules as they change.