RealAnalystJobs Analyst Job Data

For every live role:

• Title and employer, as the company wrote them.
• Location as stated, from which we resolve a country (word-anchored matching) and, where it names a major city, a city (GeoNames, population 200k+).
• Skills extracted from the posting text at collection time — matched as whole tokens, never substrings.
• Experience level as stated, normalised into five buckets (entry, mid, senior, lead/staff, not specified).
• Posting date from the applicant-tracking system where it provides one, and the date we first saw the role otherwise.
• Salary text when the employer states one — recorded verbatim, never inferred. A role counts as “disclosing pay” only when it states an unambiguous two-ended range with a currency.
• Work mode (remote / hybrid / on-site) only on roles employers post here directly; for scraped roles we detect remote from the location text and report nothing about hybrid vs on-site.
• Source — which careers platform the role came from — and the employer's own application link.

We do not collect anything about the people who view or apply to roles as part of this dataset.

• Company-first. We read a curated list of employers' own careers pages and applicant-tracking systems (Greenhouse, Lever, Ashby, Workday, SmartRecruiters and others — around 18 platforms). We do not scrape other job boards, so roles are not duplicated across aggregators.
• Analyst roles only. Listings are matched on title to business, data, product, BI and analytics-engineering roles; adjacent analytical titles (financial, marketing, operations analyst) are included and classified as such.
• One row per application link. The same role appearing twice at its source is counted once.
• Live only. A role that has come down from the company's careers page is removed on the next run, not left to age. Counts therefore go down as well as up.
• Nothing inferred. No salary is estimated, no country is guessed from a company's headquarters, no skill is assumed from a title.

The careers pages are re-read nightly and the site's copy of the data is rebuilt in the early hours (UTC). Every figure on the report and on this page is recomputed from that copy within an hour. Each role also carries the date it was first seen, which is how the “added in the last 7 days” figure is produced.

The data is well suited to:

• Sizing analyst hiring by country, city, role family and employer at a point in time.
• Measuring which skills employers actually list — for curriculum planning, course design, or a piece on what “analyst” means in practice.
• The share of roles that are remote, the seniority mix, and how many postings disclose pay.

It is not suited to, and we do not publish:

• Trends over time — closed roles are removed rather than archived, so there is no history yet. We are recording first-seen dates and will publish trends once a comparable history exists.
• Salary benchmarks — most postings state no pay and the ones that do use mixed currencies and periods; we publish the disclosure rate, not an average.
• A hybrid vs on-site split — work mode is only known for employer-posted roles.

Please link to the report page so readers can check the live figures, and keep the month and sample size with any statistic you quote — a number without its window becomes wrong the moment the board moves.

RealAnalystJobs, "Analyst Job Market Report", September 2026. Data analyzed: 2,783 live analyst roles. https://realanalystjobs.com/analyst-job-market

Short form: RealAnalystJobs, “Analyst Job Market Report”, 2026.

Download the current report as JSONView the report

Want a cut of the data the report doesn't show — a specific country, role family or skill — or a quote on what the numbers mean? Email hello@realanalystjobs.com. We answer with real counts from the live feed, and we will tell you when a question is one the data cannot answer.