Signal Radar — AI Account Intelligence & ICP Scoring

Turn a list of company domains into ICP-scored, signal-tagged account intelligence — built entirely on public, legal, company-level sources. It's the AI judgment layer that raw scrapers don't give you: feed it domains plus a plain-English description of your ideal customer, and it returns a fit score, the reasons behind it, and the concrete buying signals worth acting on.

This is the "signal → enrich → score" loop (think Clay) as a self-serve Apify actor.

🔗 Chain it after your scraper

Already pulling company lists with Google Maps Scraper (compass/crawler-google-places), Google Maps Email Extractor (lukaskrivka/google-maps-with-contact-details) or LinkedIn Jobs Scraper (curious_coder/linkedin-jobs-scraper)? Pipe those companies straight in — paste the domains (or the websites column from that run) and get each account ICP-scored and signal-tagged. Works with the output of any Maps, LinkedIn or company-list scraper.

Quick start

1. Fill in the input — set the fields in the input tab (prefilled with a working example).
2. Click Start — the Actor runs and writes one record per item to the dataset.
3. Get your data — export to JSON, CSV or Excel, or pull it via the API.

What it does

For every company domain it gathers public, logged-out signals:

• Firmographics — name, industry, description, location, size — from the homepage and about page
• Hiring signals — open roles from the company's public job board (Greenhouse, Lever, Ashby)
• Tech stack — detected from the homepage (commerce, analytics, CRM, payments, frameworks…)
• Recent news — headlines via Google News

Then OpenAI fuses everything into:

• A 0–100 ICP-fit score with an icpTier (strong / medium / weak / none) and written reasoning
• A suggested outreach angle
• The qualified buying signals it can justify from the evidence
• An optional role-based contact email (sales@, info@ — never personal data)

Recurring monitoring

Pass a previous run's dataset as baselineDatasetId and Signal Radar reports what changed — hiring surges, newly adopted tools, fresh news — turning it into a scheduled account monitor. Signals change constantly and B2B data decays ~25%/year, so there's always a reason to re-run.

Legally clean by design

Only logged-out, public, company-level sources (registries, public ATS job boards, RSS, homepages). No LinkedIn personal profiles, no logins, role-based inboxes only — GDPR-minimized.

Input

Field	Type	Description
domains	array	Company domains/URLs/emails (domain is extracted). Required.
icpDescription	string	Your ideal customer, in plain English. The AI scores fit against this.
monitorSignals	array	Which signal types to keep (empty = all).
baselineDatasetId	string	A previous run's dataset ID → report changes since then.
findRoleEmail	boolean	Surface a role-based inbox (default false).
maxConcurrency	integer	Companies analyzed in parallel (default 5).

{
  "domains": ["stripe.com", "linear.app", "notion.so"],
  "icpDescription": "Mid-market B2B SaaS (50-500 employees) in fintech or devtools, actively hiring engineers, on a modern cloud stack.",
  "findRoleEmail": true
}

Output

One record per company:

{
  "domain": "linear.app",
  "name": "Linear",
  "industry": "B2B SaaS",
  "icpFitScore": 85,
  "icpTier": "strong",
  "icpReasoning": "Cloud-based product-development tool aligned with devtools ICP…",
  "outreachAngle": "Reference their push into AI-assisted development…",
  "hiring": { "ats": "ashby", "openRoles": 30, "sampleTitles": ["Senior Fullstack Engineer", "…"] },
  "techStack": ["Next.js", "Zendesk"],
  "signals": [{ "type": "hiring", "detail": "Hiring 30 roles incl. multiple engineering…", "strength": "high" }],
  "roleEmail": "hello@linear.app",
  "citations": ["https://linear.app", "ashby:job-board", "…"]
}

Pricing (pay-per-event)

Event	When	Price
company-enriched	per company (public-signal gather)	$0.15
ai-icp-scored	per company (AI fusion + score)	$0.30
signal-detected	per qualified buying signal	$0.20

Who it's for

Sales teams, RevOps, agencies and GTM engineers running signal-based outbound. Signal-based outreach converts several times better than static lists.

Notes

• Requires an OpenAI API key (set as the OPENAI_API_KEY environment variable).
• Schedule it against your watchlist (weekly/monthly) with a baselineDatasetId to catch changes automatically.

Related actors

• Competitor Pulse — AI Competitor Intelligence Monitor — Watch competitors holistically — pricing, website, hiring, tech, news and
• Argus — Autonomous Intelligence Agent — Describe a mission in plain English; Argus plans what to watch, gathers from public
• AI Company Enrichment — Turn a list of domains into structured company profiles — firmographics, contacts