Most people figure out what a job pays after they've already invested hours in the application and interviews. That's backwards. Knowing the realistic range up front tells you whether the role is worth pursuing and gives you a number to anchor on when the salary conversation arrives. It takes about twenty minutes per role, and it changes how you negotiate.
Why the research happens before you apply
There are two payoffs. First, you filter. If a job pays well under what you need and the posting is firm about budget, you've saved yourself a week of effort. Second, you anchor. When a recruiter asks for your expectations, a researched range lets you answer with a number that sounds informed instead of hopeful. People who name a specific, well-sourced figure tend to land closer to the top of the band than people who say "I'm flexible."
There's a confidence effect too. Walking into a salary discussion knowing the market rate changes your posture. You're not guessing, you're stating.
Where to actually find the numbers
No single source is reliable on its own, so triangulate across a few:
- Salary aggregators. Glassdoor, Levels.fyi for tech, Payscale, and Comparably give self-reported ranges. Treat them as rough, since the samples skew and titles vary, but they set a ballpark.
- Government and official data. In the US, the Bureau of Labor Statistics publishes median wages by occupation and region. Many countries have an equivalent. It's less granular but more trustworthy for a baseline.
- The job postings themselves. A growing number of postings list ranges, partly because pay-transparency laws now require it in several US states and across the EU. Search five or six similar roles and you'll see a real band form.
- People. The most accurate source is someone doing the job. If you have a contact in the field, a direct question about ranges is more useful than any website. An informational conversation is a good place to ask.
When these sources agree within a range, trust it. When they don't, weight the official data and the live postings over the self-reported sites.
Adjust for the things that move the number
A raw median hides a lot. Three factors shift it enough to matter:
- Location. The same title pays very differently in San Francisco, Austin, and a remote role priced to a national average. Adjust for cost of living and for whether the company pays by location or by role.
- Company size and stage. A funded startup may pay below market in cash and make it up in equity. A large enterprise pays steadier cash. These aren't comparable on base salary alone.
- Your specific level. "Marketing manager" spans a huge range depending on team size, budget owned, and years in. Place yourself honestly inside the band, not at the top by default.
Turn the research into a number you can say out loud
After twenty minutes you should have a range with a floor, a midpoint, and a stretch. Your stated expectation usually sits around the midpoint to upper third, assuming you meet the requirements. Keep the sources in a note so that if a recruiter pushes back, you can say where the figure comes from.
Doing this for every role you target adds up, so keep it lightweight. The same prep that sharpens your salary expectations also sharpens the questions you ask in interviews, since you'll understand the level the role really sits at. Pair this with a CV that matches that level, and you walk in aligned. Some people draft that CV straight from their LinkedIn profile with a tool like Postulit so the seniority signals line up with the band they're targeting.
Pick the next role on your list and run the twenty-minute version now. You'll know whether to apply, and you'll have a number ready before anyone asks for it.