swan-gtm Hiring signal verification
"They're hiring" is one of the most-used signals in outbound — and one of the most misattributed. Research tools stitch together careers pages, job boards, and ATS tenants, and they regularly hand you another company's hiring spree under your target's name. The prospect knows their own company better than any research tool: one wrong "noticed you're opening a sales org" and your credibility is gone, or worse, your scoring model promotes an account on evidence that belongs to someone else. This skill is the gate every hiring claim passes through before it becomes a fact.
Template placeholders
Replace every {{...}} before enabling. See the setup checklist
for the full setup list.
{{WEB_RESEARCH_TOOL}}— Your AI web-research tool or agent whose prose summaries feed scoring and outreach (e.g. a deep-research agent, an enrichment provider's AI summary){{PEOPLE_SEARCH_TOOL}}— Your people/employee search tool, filterable by company domain and title (e.g. LinkedIn Sales Navigator, Apollo, Lusha)
When this applies
Any time a hiring or open-roles signal is about to be used as a fact — in an ACV/tier assessment, an MQL or scoring alert, an outreach observation/hook, or a CRM/account memory note. If you are about to write "they're hiring X" or "actively hiring for Y," this standard governs it. No exceptions.
The rule — verify at the company's OWN source
A hiring claim only counts when confirmed from the target company's own source:
- The company's own careers page (hosted on their domain), OR
- A LinkedIn Jobs listing (or other job board) whose posting is confirmed to belong to the target company — matching company name AND domain, not just a shared ATS or job-board slug.
{{WEB_RESEARCH_TOOL}} prose is NEVER sufficient on its own for a hiring claim. Web
research summaries routinely misattribute another company's roles to your target — same ATS
tenant, same job-board slug, or a similarly named company. Treat any hiring detail that came
from web research as unverified until it is confirmed at one of the two sources above.
To verify at source, use any of: the company's careers page directly, a LinkedIn Jobs (or
company-page jobs tab) check scoped to the exact company, or {{PEOPLE_SEARCH_TOOL}}
filtered by title at the exact domain.
Misattribution guard — discard when citations point elsewhere
When a {{WEB_RESEARCH_TOOL}} answer supplies hiring (or any firmographic) "facts,"
check the citations before using anything. If the answer's sources are dominated by a
domain that is NOT the target company's domain, DISCARD those derived facts entirely.
Do not launder another company's data into the target's record, score, alert, or outreach.
Re-verify from the target's own source, or drop the claim.
When a hiring signal fails verification
Drop it. Do not soften it ("they may be hiring…") or hedge it ("I noticed what looked like…"). If it did not verify at the company's own source, it does not appear in the score, the ACV assessment, the alert, the outreach, or the account record. A wrong hiring claim is worse than none — the prospect knows their own company better than any research tool.
Real failure this prevents
A scoring run's web research returned a full GTM hiring spree for a target account — VP Sales ($280–400K OTE), CMO, Director of RevOps, GTM Engineer, Demand Gen Lead, and more — and it landed in a high-priority MQL alert as a "KEY INSIGHT" driving a "massive GTM expansion" narrative. Every role was cited from a different company's careers page — an unrelated software vendor the research tool had conflated with the target. The target actually had one open role (an analyst position). Checking the target's own careers page or LinkedIn Jobs for the real domain would have caught it in one step.
What good looks like
Every hiring claim that reaches a score, an alert, an outreach draft, or an account note carries a source on the target company's own domain (or a name-AND-domain-matched job listing). Web-research prose alone never survives as a hiring fact. Signals that fail verification disappear — no hedged versions leak through. And when a research answer's citations point at the wrong domain, everything derived from it is discarded, not just the hiring lines.