What US labour statistics say about QA analyst and tester pay

Buyers who are deciding between an in-house team and a testing vendor often start with one number: what a tester costs to employ. In the United States, the public reference for that number is the Bureau of Labor Statistics (BLS), which tracks software quality assurance analysts and testers as a distinct occupation, code 15-1253 in the Standard Occupational Classification. This post sets out what the BLS figures say, what they leave out, and how to use them without turning them into a price benchmark they were never designed to be.

The headline figures

The BLS Occupational Outlook Handbook reports that the median annual wage for software quality assurance analysts and testers was $104,300 in May 2025. The median is the point at which half of workers earned more and half earned less. The same page gives the spread: the lowest 10 percent earned less than $61,440 and the highest 10 percent earned more than $167,010, again for May 2025.

The handbook also breaks the median down by the industries that employ the most testers. For May 2025 it lists:

Industry	Median annual wage, May 2025
Manufacturing	$129,930
Finance and insurance	$104,260
Administrative and support services	$102,660
Software publishers	$101,070
Computer systems design and related services	$99,960

On size, the handbook states that software quality assurance analysts and testers held about 187,600 jobs in 2025, and that employment in the occupation is projected to grow 6 percent from 2025 to 2035. Computer systems design and related services, the industry that includes outsourced IT and testing firms, employed 30 percent of them.

What the wage figure includes and excludes

The BLS wage figures come from the Occupational Employment and Wage Statistics (OEWS) survey. Its technical notes define what counts as a wage. Included are the base rate, cost-of-living allowances, guaranteed pay, hazardous-duty pay, incentive pay including commissions and production bonuses, and tips. Excluded are overtime pay, severance pay, shift differentials, nonproduction bonuses, the employer cost of supplementary benefits, and tuition reimbursements. The survey also leaves out the self-employed, so independent contractors are not in these numbers.

That exclusion list matters. A salary figure is a floor for what an employee costs, not the cost. The BLS Employer Costs for Employee Compensation release for June 2026 shows how large the gap can be at the level of the whole economy: for private industry workers, total compensation averaged $46.89 per hour worked, of which wages and salaries were $32.82 (70.0 percent) and benefits were $14.07 (30.0 percent). That release covers all private industry workers, not testers specifically, so treat it as an indication of the order of magnitude of benefits, not as a multiplier for your own team.

What the figure does not tell a buyer

It is not a vendor rate. A vendor's hourly or daily rate has to cover the tester's pay and benefits, plus time between assignments, recruiting, training, test management, account management, tools, facilities, and a margin. Comparing a vendor's rate with a median salary divided by working hours will always make the vendor look expensive, and the comparison says nothing about whether it is.

It is not a fair price. A median describes what employers paid in a survey period. It does not tell you what your project needs, which seniority mix fits your risk, or what a vendor should charge you.

It is national and occupational, not local and specific. The median hides differences by state, city, industry and specialism. Performance, security and accessibility testing, or testing in regulated domains, may sit well away from the median in either direction.

It is a point in time. The figures above are for May 2025. Check the reference period every time you reuse a number, and refresh it when BLS publishes a new release.

How to use it in a cost comparison

The BLS median is useful as one input into a fully loaded in-house cost, which you then set against vendor quotes on the same scope. A practical sequence:

1. Pick the wage point that matches the role you would hire, not automatically the median. A lead test engineer and a junior manual tester are both in code 15-1253.
2. Add your own organisation's benefit and payroll cost ratio, from your finance or HR team. Use the economy-wide ECEC split only if you have nothing better, and say so.
3. Add recruiting, onboarding, management time, tooling, devices and environments, and the months it takes to hire.
4. Compare the result with vendor proposals priced on the same scope, the same coverage and the same exit terms.

Keep the source and reference period next to every figure in your business case, so that anyone reviewing it can check the number.

What this means for your selection

Use public wage data to sanity-check your own assumptions, not to judge a vendor's price. Enter your own salary, overhead and hiring figures in the cost model, which puts in-house and outsourced options side by side on the same basis. When you receive proposals, ask each vendor to break its rate into the components listed above using the RFP builder, and score cost transparency in the scorecard. The guides on in-house vs outsourced QA cost, rate cards and what they hide and hidden costs of outsourcing testing cover the line items buyers most often miss.