Finding the Average Salary for a Job: A Reliable Research Method

TL;DR

• Don't rely on one source — combine BLS, crowdsourced platforms, and live job postings.
• Use median, not mean — averages get distorted by high-earning outliers.
• Geographic pay gaps can exceed 20% (Pew Research) — always adjust for city.
• Pay transparency laws make live job listings in CA, NY, CO especially valuable.
• Total compensation (base + bonus + benefits + equity) can be ~30% more than base.

Walking into a salary conversation without data is how candidates leave thousands on the table — or worse, price themselves out before the first interview. The fix is straightforward: triangulate a few credible sources, adjust for location and experience, and bring a researched range instead of a single number. This guide walks through where to look and how to interpret what you find.

What "Average Salary" Actually Means

Most "average" salary references are really referring to the median — and that's the right number to anchor on. Means get skewed by a few high earners; medians describe the typical worker.

Two further distinctions matter.

Base vs total compensation

Base pay is the headline. Total compensation (TC) includes bonus, equity, retirement matching, healthcare, and other benefits — often adding 20–30% to the base number. When comparing offers or negotiating an offer, always compare TC to TC.

Market rate vs average

The average is a statistical calculation. Market rate is what employers are actively paying right now to attract talent. In hot markets, market rate runs ahead of the average; in cooling markets, behind.

The Three Sources Worth Using

No single source is sufficient. Combine three.

1. Government data

The Bureau of Labor Statistics is the most rigorous source — based on employer tax filings, methodologically sound, slightly lagged. As of late 2024, BLS reported median weekly earnings for US full-time workers at $1,145. Good for floors, ceilings, and broad trends.

2. Crowdsourced platforms

Glassdoor, Payscale, Salary.com, Levels.fyi — useful for company-specific patterns and current trends. Self-reported data skews — some entries are stale, top earners over-report. Treat as range estimates, not exact values.

3. Live job postings

Pay transparency laws now require disclosed ranges in California, New York, Colorado, Washington, and a growing list of states. Scanning recent postings — even outside your city — surfaces real budget data in real time. The most actionable signal of the three.

Using Real Job Listings Effectively

Three moves that extract maximum signal.

Check transparent-state postings

CA, NY, CO listings often show real ranges even for nationwide roles. Use them as a baseline for what employers are actively budgeting.

Look at ranges, not the midpoint

A "$120k–$160k" range tells you the company will pay $160k for the right person. The candidate the company actually wants usually gets paid in the top third.

Compare similar roles across companies

Three or four postings for similar roles produce a tighter band than any single posting. Outliers get visible quickly.

How Location Reshapes Salary

A $100k salary in Des Moines is a comfortable life. The same salary in San Francisco is a struggle. Pew Research shows geographic pay gaps within the same industry routinely exceed 20%.

For remote roles, two common employer approaches.

Location-adjusted pay

Companies pay based on where the candidate lives. Common at large tech companies; the trade-off is steady pay scaling with cost of living.

Location-agnostic pay

Companies pay a flat national rate regardless of location. Better for workers in low-cost areas; pressured at high-cost cities. Increasingly common among smaller, fully-remote companies.

Identify which approach a target employer uses before negotiating.

Adjusting for Experience and Skill

A "Manager" title means radically different things across companies. Three adjustments worth making.

Years of experience

Entry-level lives near the 25th percentile of the role's band. Mid-career hits the 50th. Senior or specialist roles approach the 75th–90th.

Hard skill premiums

Specific certifications, languages, or platforms can shift the band materially. Cloud architects, ML engineers, and security specialists routinely command premiums of 20–40% over generalists.

Salary compression check

Long tenure at one company can produce salary compression — new hires entering above your current rate. If your pay hasn't been benchmarked in 2+ years, run the analysis before your next review.

Common Mistakes

Three patterns to avoid.

Asking individuals instead of the market

Coworker salary conversations are awkward and produce limited signal. Aggregated market data is more useful and less complicated.

Comparing base pay alone

Benefits can be 25–30% of total compensation. Comparing $120k base + 10% bonus + great equity against $130k base + no bonus + no equity by base alone gets the answer badly wrong.

Using a single source

Government data is comprehensive but slow. Crowdsourced data is current but noisy. Live postings are real but role-specific. Each adds something the others miss.

How Employers Use Salary Benchmarks

Employers don't guess. They hire compensation specialists to run salary surveys every six months, often paying for Mercer or Radford data that covers their specific industries and locations.

Showing you've done your own homework — phrases like "based on current market data for this role and location" — meaningfully shifts the negotiation. The recruiter knows they're talking to someone with grounds for their ask, not a wish. See how recruiters screen candidates for the broader context on where these conversations land in the process.

The Bottom Line

Salary research isn't glamorous, but it pays directly. The candidates who consistently land at the high end of their target range have done the work — BLS for the floor and ceiling, crowdsourced platforms for company-level patterns, live postings for what's being paid right now, and explicit adjustments for location, experience, and skill premiums. The investment is a few hours. The return is years of compounding income.

FAQs

What's the most accurate way to find average salary?

Triangulate government data (BLS), crowdsourced platforms (Glassdoor, Payscale, Levels.fyi), and live job postings — especially from pay-transparent states. No single source is sufficient on its own.

Should I trust salary ranges on job postings?

Yes, when they're recent. Posted ranges in CA, NY, CO are particularly reliable. Remember ranges are intentionally wide — the company pays the top of range only for the candidate they really want.

How often does salary data change?

Typically every 6–12 months in stable industries. Faster in fast-moving fields (AI, security, specialised cloud roles) where the market can shift quarterly.

Is average salary the same as market rate?

No. Average is a statistical calculation. Market rate is what's actively being paid right now to attract talent — often higher than average in hot markets, lower in cooling ones.

What's the highest-leverage step?

Find three recent live job postings for your target role and location, compare their ranges, and use that as the floor of your negotiation. Live postings give you the freshest, most actionable data of any source.