Recruitment Agency Database
A registry-backed recruitment agency database that discovers recruitment firms, verifies business status, enriches office contacts, and writes one normalized Excel row per company every day.
This downloadable system uses a scheduled Make.com scenario to maintain a searchable directory across New Zealand, Australia, the United Kingdom, the United States, Canada, and Ireland. It begins with New Zealand as the proof-of-concept connector, then applies the same canonical record model to the remaining five countries.
The scale justifies automation. The World Employment Confederation’s 2026 Industry Impact Report says private employment agencies placed 61 million people in jobs during 2024, while the American Staffing Association reports roughly 27,000 staffing and recruiting companies operating close to 54,000 U.S. offices. A static list quickly loses value when firms register, relocate, add branches, or cease trading.
What the Recruitment Agency Database Resolves
The tool replaces repeated registry searches, copied addresses, inconsistent phone formats, and spreadsheets containing several versions of the same firm. Each daily run searches eligible entities, classifies likely recruitment businesses, compares them with stored records, enriches public profiles, and applies an upsert or closure action.
Unlike candidate-focused database software for recruitment agencies, this product maps the agencies themselves. The resulting recruitment agencies database supports market research, partnership discovery, territory planning, supplier mapping, and directory operations without mixing candidate records into the dataset.
Registry Sources and Country Routing
Every country connector preserves its source identifier and retrieval timestamp. New Zealand records come from the Companies Office data APIs. Australia uses ASIC register data access, while UK companies are retrieved through the Companies House REST API.
The U.S. route resolves the relevant state registry rather than treating one federal directory as authoritative; the SBA state-registration guidance explains why entity registration is state-based. Canada starts with Canada’s Business Registries and federal corporation data, and Ireland uses CRO Open Services where available.
When the UK connector is enabled, the same record model becomes a UK recruitment agencies contact information database containing registered identity, office details, and verified public profile URLs.
Core Features
| Feature | Description |
|---|---|
| Registry-first agency discovery | Manual web searching misses legal-status changes. The system queries the appropriate registry, captures the entity identifier, and filters records against recruitment-related names and business classifications. |
| AI-assisted relevance classification | Broad registry results include unrelated companies. A structured classifier scores whether an entity operates as a recruitment, staffing, executive-search, or employment agency and stores the reason for review. |
| Head-office and branch normalization | Addresses arrive in different formats and branches can look like separate companies. The workflow standardizes country, city, address, and phone fields while retaining regional-office details separately. |
| Duplicate-safe daily upserts | Repeated runs can create duplicate rows. The agent matches registry ID first, then normalized company name, domain, phone, and address before inserting or updating the Excel record. |
| Closed-company reconciliation | Old directories keep inactive firms indefinitely. Each run checks registry status, marks confirmed closures, records the status date, and removes the company from the active output view. |
| Public profile enrichment | Registry records rarely contain discovery links. The tool resolves Google business locations, LinkedIn organization pages, and Facebook Pages without replacing registry evidence. |
| Structured Excel publishing | Copy-pasted spreadsheets drift in column order and formatting. The publisher writes a fixed business record to an Excel table through Microsoft Graph, one company per row. |
| Run ledger and exception queue | Silent failures make coverage impossible to judge. Every connector logs request status, retry count, source timestamp, classification decision, match action, and unresolved-field warnings. |
Recruitment Agency Database Matching Rules
A registry identifier is the strongest key. If that identifier is unavailable across an enrichment source, the matcher uses a weighted combination of normalized name, registered address, phone number, website domain, and city. Social links never decide legal identity on their own.
The best recruitment database for agency use is therefore not simply the longest list. It exposes where each field came from, when it was checked, why two records were merged, and whether the company remains active. Teams using a database for recruitment agency research can audit the record rather than trusting an unexplained row.
Recruitment Agency Database Tech Stack
| Component | Role in this build |
|---|---|
| Make.com | Runs the daily country routes, HTTP requests, retries, branching, and record upserts without requiring a separate scheduler. |
| Registry connectors | Retrieve legal name, status, registered office, jurisdiction, and stable company identifiers from authoritative sources. |
| Classification service | Evaluates borderline entities against a constrained recruitment-agency taxonomy and returns a score with its reason. |
| Normalization functions | Standardize phone numbers, addresses, country codes, URLs, and company-name suffixes before matching. |
| Microsoft Excel | Delivers the operational directory in a familiar table with filters, fixed columns, and one row per legal entity. |
| Audit store | Retains run summaries, source snapshots, errors, closure decisions, and manual-review items outside the presentation sheet. |
Daily Run Contract
The production schedule is once every 24 hours. Each accepted row contains country, city, company name, head-office address, head-office phone, regional-office address, regional-office phone, Google link, LinkedIn link, and Facebook link. Optional internal columns hold registry ID, source URL, status, confidence, first-seen date, and last-checked date.
The run is considered complete after every enabled country route reaches a terminal state: inserted, updated, unchanged, closed, or queued for review. Rate-limit responses are retried with backoff rather than interpreted as missing companies.
Performance and Data Quality Benchmarks
| Check | Acceptance rule |
|---|---|
| Schedule | One completed reconciliation cycle per enabled country every 24 hours. |
| Row model | One active row per legal company, with all requested business fields represented as columns. |
| Duplicate control | Registry ID wins; fallback matches require enough agreement across name, address, phone, domain, and city. |
| Closure handling | A confirmed inactive company leaves the active worksheet during the next successful daily reconciliation. |
| Failure visibility | Rate limits and server errors retry with backoff; unresolved runs enter the exception queue rather than disappearing. |
Use Cases
- Build a current partner directory: Business-development teams can identify active recruitment firms by country and city while retaining the registry evidence behind each record.
- Map regional office coverage: Operations teams can separate headquarters from branches and see where an agency has a genuine local presence.
- Refresh market research daily: Analysts can detect additions, relocations, profile changes, and confirmed closures without rebuilding the spreadsheet.
- Prepare territory lists: Teams can filter the clean Excel table by country, city, status, or confidence before assigning research or outreach work.
- Audit directory quality: Data managers can trace duplicates, classification decisions, source timestamps, and unresolved fields through the run ledger.
Project Directory
recruitment-agency-database/
├── README.md
├── .env.example
├── make/
│ ├── daily-orchestrator.blueprint.json
│ ├── nz-proof-of-concept.blueprint.json
│ ├── enrichment-router.blueprint.json
│ └── closure-reconciliation.blueprint.json
├── config/
│ ├── countries.yml
│ ├── agency-taxonomy.yml
│ ├── excel-columns.yml
│ └── matching-thresholds.yml
├── src/
│ ├── connectors/
│ │ ├── nz_companies_office.py
│ │ ├── asic_registry.py
│ │ ├── companies_house.py
│ │ ├── us_state_registries.py
│ │ ├── canada_registries.py
│ │ └── ireland_cro.py
│ ├── classification/
│ │ ├── agency_classifier.py
│ │ └── prompt_schema.json
│ ├── enrichment/
│ │ ├── maps_profiles.py
│ │ ├── linkedin_profiles.py
│ │ └── facebook_profiles.py
│ ├── matching/
│ │ ├── normalize.py
│ │ ├── deduplicate.py
│ │ └── reconcile_status.py
│ ├── publishing/
│ │ ├── excel_writer.py
│ │ └── workbook_schema.py
│ └── audit/
│ ├── run_ledger.py
│ └── exception_queue.py
├── tests/
│ ├── fixtures/
│ ├── test_classification.py
│ ├── test_deduplication.py
│ ├── test_registry_status.py
│ └── test_excel_schema.py
└── output/
├── recruitment_agencies.xlsx
└── run_summary.json
How to Build a Global Agency Directory Using Recruitment Agency Database
Download & Set Up the Project
Download, set up, and install Recruitment Agency Database to get the project running. If you hit any difficulty, contact us here.
Open the Control Panel
Open the supplied dashboard, connect Make.com and Microsoft credentials, then confirm the workbook location and the New Zealand proof-of-concept route.
Choose Countries and Rules
Select enabled countries, recruitment classifications, closure handling, confidence threshold, and whether uncertain matches should enter the manual-review queue.
Run the Registry Scan
Select Run Registry Scan to fetch, classify, reconcile, and publish records; the dashboard returns row counts, exceptions, and the updated Excel workbook.
CogworkLabs provides AI agent development for connector changes, deployment, monitoring, and additions to the country or field model. Related workflow automation services can connect the directory to an existing CRM, data warehouse, or review process.
FAQs
How recruitment agencies maintain candidate databases?
Recruitment agencies usually maintain candidate databases inside an applicant tracking system or recruitment CRM, with consent controls, role-based access, deduplication, and retention rules. This tool serves a different purpose: it maintains a database of recruitment companies and their public business details, not candidate profiles.
Is a recruiting agency database proprietary?
It can be proprietary as a compiled dataset and workflow even when many source facts are public. This project preserves source provenance, but its classification rules, matching logic, enrichment decisions, review history, and normalized workbook are part of the delivered system.
