Every data product has holes. Most vendors leave you to discover theirs. Here are ours, so you can decide whether they matter for what you are doing.
Coverage
- We do not cover every employer. Sources that restrict automated access are absent, and we do not estimate around the gap.
- Coverage is uneven by country and by employer size. Smaller employers who post only to a local board are underrepresented.
- A market where we have thin coverage will produce a plausible-looking median computed from too few postings. The counts are always shown so you can judge.
Salary
- Only postings that publish a range are included in any salary figure.
- Disclosure is not random: it skews to regulated markets and larger employers. Medians inherit that skew.
- We do not convert currencies. A cross-border comparison of our medians compares tax systems as much as pay.
Age and freshness
- Age is computed from the publication date, not from continuous observation.
- Republishing resets it, and we cannot always detect that.
- A listing being visible does not prove the role is open.
Titles and categories
- Grouping by title is a judgement call. Two people would draw the lines differently.
- Composite roles land in one bucket.
- New role names take time to appear in any grouping, including ours.
Intent
We cannot see whether an employer means to hire. Nothing we publish should be read as a claim about intent, and where our pages discuss stale listings they say "signal", not "verdict", deliberately.
Why publish this
Two reasons. A buyer who finds a limitation on their own concludes it was hidden; a buyer who reads it here concludes it was measured. And writing the list down keeps us honest about which gaps we have accepted and which we are still working on.