Methodology

Every data-backed number published on Remote First Jobs comes from the job-postings corpus this site runs on. This page explains where that data comes from and the rules every published statistic follows. How articles themselves are produced is covered in our editorial policy.

Corpus Size and Collection

The system scans 21,000+ company career pages directly, every few hours, and has detected and verified over 1 million remote job postings since launching in 2023 — currently analyzing 100,000+ postings per month. Postings are collected from company career sites and the applicant-tracking systems behind them, at the source, not scraped from other job boards.

Normalization

Raw postings are messy, so before anything is counted the corpus is normalized: duplicate and reposted listings are collapsed, third-party agency posts are filtered out, and job attributes such as location and remote policy are mapped to a common structure so that postings from different companies can be compared and counted consistently.

The n ≥ 30 Suppression Rule

We do not publish statistics computed from small samples. Any figure that would be based on fewer than 30 postings (n < 30) is suppressed rather than published. Published numbers state their sample size so you can judge them.

What as_of Means

Every corpus-derived number or chart carries an as_of date. That is the date of the corpus snapshot the figure was computed from — the number reflects the postings we had collected as of that date, and it does not update live after publication. If the same figure is recomputed later, it gets a new as_of date.

The Keyword-Match Caveat

Counts based on keyword matching are a demand proxy, not a census. When we report how many postings match a term, a posting counts because its text matches that term — a relevant job described in different words is not counted, and an unrelated job that happens to mention the term can be. Term-matched counts are useful for comparing relative demand, not for measuring the absolute size of a market.