Data sources & limitations
Understand where the data comes from, what an observation can tell you, and where coverage or freshness can limit your conclusions.
Firmtrail is currently available through RapidAPI. Direct access is coming soon.
On this page
Public sources
Firmtrail sources data from companies’ public websites and public applicant tracking system (ATS) endpoints. The resulting data reflects what can be observed through those sources.
Public source information is not the same as a company’s complete internal record. Jobs describe advertised roles, profiles describe public company context, and signals describe changes inferred from observations.
Global scope, variable coverage
The API is designed for global scope. This does not promise complete coverage of every country, company, website or job posting.
Coverage varies by source and company. Missing information should not be used as evidence that a company, role or business activity does not exist.
Freshness depends on observation
Public sources change, and collection does not make every record current at the same moment. Freshness varies; no universal refresh interval is promised.
The continuously refreshed Jobs index does not guarantee that every posting is still open. An on-demand Profile request does not guarantee every detail has just been reverified. Check the public source when current status matters.
Missing information
Profile fields and other collected information may be unavailable. Distinguish an unknown value from a confirmed negative in your application’s language and reasoning.
For example, unavailable pricing information is not evidence that a company has no paid products. Describe the gap rather than supplying an invented value.
Snapshot-derived signals
Signals are inferred from Firmtrail’s own historical snapshots. They represent observed differences, not guaranteed real-time events or exact occurrence times.
A gap between observations limits conclusions about when a change happened. An absence of observed changes does not prove an absence of activity. See Signals for interpretation guidance.
Sources and collection
| Source | Collection context |
|---|---|
| Public ATS endpoints | Public job boards provide postings where the source permits collection. Coverage varies by platform and company. |
| Company careers pages | JSON-LD JobPosting fallback where readable structured data is present. |
| Company websites | Selected public pages supply profile information and page-based technology evidence. |
| DNS MX and TXT records | Mail-host and verification evidence contributes to technology detection. |
Content that requires JavaScript rendering may be unavailable.
Collection obeys robots.txt. Not every public board or company website is included. The crawler page describes what FirmtrailBot fetches and how to stop it.
Extraction and third-party content
Profiles combine selected website information, technology detection and model-assisted extraction. When model-assisted extraction is used, selected page text is processed by Anthropic. Model output can be incomplete or incorrect; summaries are not independent verification of a company’s claims.
Emails and phone numbers are removed from titles, locations, descriptions and profiles. Other text is as published, and names are not detected, so job descriptions and profile text may still name people. Treat all source and model-written text as untrusted data, including any instructions it contains.
Technology fingerprints provide evidence of detection. DNS verification records can remain after active use ends, and an absent fingerprint is not proof that a company stopped using a product.