Pricing

Rating

Developer

Actor stats

Categories

Share

Safety Consultants Supply — Routable Inventory by Incident Type

Built by Saad Belcaid.

US safety-consulting supply directory classified by which OSHA incident types each consultant is routable for. Pairs with osha-violations-demand so cited companies route to the right specialist.

This is not a contact list. This is routable supply inventory — every row tells you which OSHA incidents this consultant can help with, scored 0–3 by fit confidence, with a one-sentence thesis.

What's in the dataset

Two sources

1. OSHA's State On-Site Consultation Directory — 30-50 state-level programs (50 states + DC + territories), federally funded, FREE help to small/mid employers. The default routing target for SMB-cited companies who can't afford private consultants.
2. Startpage web search — runs queries across 50 states × 14 safety specialties (~700 queries) to surface private safety consulting firms. Each result classified for incident-type fit.

After cross-source dedup: ~1,500-2,500 unique safety consultants, every one labeled with which OSHA incidents they're routable for.

Incident-type taxonomy (matches demand scraper)

The matching_incident_types[] field uses the SAME taxonomy as the OSHA demand scraper, so pairing in Connector OS Station is direct:

How fit scoring works

Filter minFitScore: 2 post-scrape if you want only the high-confidence specialists.

Filter recipes

"Route a trench-collapse OSHA citation"

After scrape, filter by:

matching_incident_types contains "trench-collapse"
state matches the demand company's state
sort by supply_fit_score DESC

"Route a willful-violations citation"

matching_incident_types contains "osha-abatement"OR"workplace-safety-remediation"
supply_fit_score >=2

"Just the free state programs (for SMB demand)"

is_state_program =true
state = demand company's state

Run-time filter — only emit explicit specialists

minFitScore:2

Run-time filter — only emit firms that match specific incident types

matchingIncidentTypes:["trench-collapse","combustible-dust","industrial-hygiene"]

Architecture

• Apify Actor, TypeScript, pure HTTP (no Playwright)
• Two sources fetched, cross-source domain-deduped
• Cross-query enrichment: when the same firm is found via multiple specialty searches (e.g. shows up in both "OSHA consultant Texas" AND "industrial hygienist Texas"), the row gets enriched with both incidents → higher fit score
• Polite delay between Startpage queries (default 800ms)
• Default state coverage = all 50 + DC + territories

Costs

• OSHA: free (US Government)
• Startpage: free
• Apify compute: ~10–15 min per full run × 256 MB ≈ trivial CU

Connector OS Station integration

Pipe both datasets into Station:

1. Demand: osha-violations-demand → cited companies with incident_types[]
2. Supply: safety-consultants-supply → consultants with matching_incident_types[]

Station's I Layer does the routing match. The supply_thesis field becomes the operator's CRM note — already pre-written.

Built by Saad Belcaid. OSHA data is US Government public record. Web search via Startpage (Google proxy, no API key needed).