how to figure out if you're underpaid without just trusting glassdoor

You Google your job title plus "salary" and Glassdoor spits back $72K. You're making $68K. So you're underpaid, right?

Maybe. Or maybe that $72K average is from San Francisco in 2021 and you're in Charlotte in 2025. Or it's mixing mid-level and senior. Or it's self-reported by people who lie up. Glassdoor is a starting point, not an answer.

Here's how to actually figure out if you're underpaid.

triangulate from 3-5 sources, not one

One salary number is an anchor. Three is a range. Five is a market rate.

Start with Glassdoor. It's the floor, the thing everyone checks. Note the range, not the average. If it says $65K–$85K, you're looking at a 30% spread. That tells you the role has variance.

Then add:

levels.fyi if you're in tech or adjacent (product, data, design). It's crowdsourced like Glassdoor but the data skews newer and the users verify via offer letters. You can filter by company and location, then narrow by years of experience. If your title isn't there, look for the closest match: "associate product manager" and "junior PM" pull from the same market.

LinkedIn salary insights. Hit the job search, type your title, click "Show salary" if it's available. LinkedIn's data comes from members who've shared comp, so it's self-selected, but it's often more current than Glassdoor and it lets you filter by your actual metro.

Payscale or Salary.com. Older tools, clunkier UX, but they let you adjust for cost of living and years of experience. Useful if you're in a non-tech role or a smaller market where levels.fyi has nothing.

Bureau of Labor Statistics (BLS) Occupational Employment and Wage Statistics. This is the actual government data. It's not sexy and the categories are broad ("market research analysts and marketing specialists" is one bucket), but it's real survey data, not self-reports. Good for sanity-checking whether the Glassdoor number is in the universe of possible.

You're not averaging these. You're seeing if they agree on a range. If four sources say $70K–$85K and you're at $68K, you're at the bottom. If they all say $70K–$85K and you're at $58K, you're getting screwed.

DM 2-3 people in the same role at peer companies

This is the move most people skip because it feels weird. It's not weird. Salary transparency is normal now, and people share when you ask properly.

Find them on LinkedIn. Search your job title, filter by your metro or remote, look for people at companies you've heard of. Not your company. Peer companies. If you work at a 50-person startup, find people at other 50-person startups. If you're at a big tech company, find people at other big tech companies.

The DM template that works:

"Hey [name], I'm a [your title] at [your company] and I'm trying to get a sense of what market rate looks like for this role. Would you be open to sharing your comp range? Happy to share mine in return. Totally get it if not."

Three things that make this land:

You're specific about why you're asking (market rate, not "I'm job hunting").
You offer to share yours in return. It's a trade, not an ask.
You give them an out.

Most people say yes. I've done this, and I've had clients do this. When someone shares, you learn what the actual number is at a comparable company — not a 3-year-old Glassdoor self-report, the number they're getting paid right now.

Say they're at $78K and you're at $68K for the same title in the same city. That's a $10K gap you can name in a negotiation.

the workload-vs-comp gut check

Salary data tells you what the market pays. It doesn't tell you if you're getting a bad deal relative to what you're actually doing.

Two questions:

Are you doing the job of the next level up? If your title is "analyst" but you're running projects and mentoring newer people, you're doing associate or senior work. That's a title-comp mismatch, and it's common at early-stage companies and understaffed teams. A raise alone doesn't fix it. You need the title. Titles compound. Your next job offer anchors to your current title, so being underpaid by title costs you twice.

Is your workload sustainable, and does the comp reflect it? I'm not talking about "I worked 50 hours this week so I deserve more." That's not how salary works. I'm talking about: if you're consistently working 60-hour weeks and covering for two unfilled roles, and your comp is at the bottom of the market range, you're getting a bad deal. Salary should reflect scope and responsibility. If your scope is bigger than your title, your comp should be too.

The gut check version of this: if you left tomorrow, would they need to hire two people to replace you? If yes, you're underpaid.

what to do with the data once you have it

You've triangulated. You've DMed. You know the range. Now what?

If you're inside the range anywhere, bottom third or top, you're not underpaid. You might want more, but that's a different conversation. The move there is to ask what it takes to get to the next level, then do that and ask for the bump in six months.

If you're below the range, you have two paths.

Path 1: negotiate internally. You bring the data to your manager. Not as a threat, as a calibration. "I've been looking at market rate for this role in [city], and the range I'm seeing is $70K–$85K. I'm currently at $68K. I'd like to talk about getting to market rate." If they say yes, great. If they say "we'll revisit at your next review," you're getting a no. If they say "we don't have budget," you're definitely getting a no.

Path 2: interview elsewhere. The fastest way to get to market rate when you're underpaid is to leave. I don't love that that's true, but it is. The average raise for staying is 3-5%. The average raise for switching jobs is 10-20%. If you're $10K under market, a 3% raise doesn't close the gap. A new offer does.

the thing Glassdoor won't tell you

Compensation isn't just salary. If you're comparing a $70K offer with okay benefits to a $68K offer with full health coverage, 401k match, and equity, the $68K offer might be worth more.

Run the math:

Employer-covered health insurance is worth $5K–$8K/year if you'd otherwise pay premiums.
A 401k match (say, 4% on a $68K salary) is $2,720/year of free money.
Equity at a late-stage startup or public company is real comp. Equity at a pre-seed startup is a lottery ticket. Don't count it.

Add it up. A $68K salary with a 4% match and full health coverage is effectively $75K–$78K. A $72K salary with no match and a $300/month health premium is effectively $68K.

Glassdoor doesn't break this out. You have to.

you're probably not as underpaid as you think, but if you are, fix it

Most people I talk to who think they're underpaid are actually at market, just at the bottom of the range. That's fixable with a year of good work and a direct ask.

Some people actually are underpaid. More than $10K under market, doing next-level work, and there's no internal path that closes it. If that's you, the market will pay you correctly. Your current company won't.

The data just tells you which one you are.