Local Business Lead Enrichment Pack

Production-ready Node.js Apify Actor for bounded local business website enrichment.

This actor is strongest when you provide startUrls, domains, or businesses with a known website. keywordLocationPairs are supported only as deterministic seed records when public search discovery is unavailable, so Maps/search-style inputs are optional and best-effort.

What It Does

• Accepts startUrls, domains, businesses, keywordLocationPairs, maxPagesPerSite, and requestTimeoutSecs
• Enriches each supplied business/site with same-origin bounded crawling
• Fetches the homepage plus likely contact/about pages first
• Extracts:
  • business identity
  • website/homepage/contact page URL
  • visible phones and emails
  • social links
  • schema.org JSON-LD
  • title and meta description
  • schema rating and review count when available
• Computes flags:
  • noWebsite
  • noHttps
  • missingContactPage
  • missingSocials
  • noVisibleEmail
  • weakTitleMeta
  • outdatedCopyright
• Writes one output record per input business/site to the default dataset
• Writes aggregate counts to the SUMMARY key in the default key-value store

Files

• .actor/actor.json
• .actor/input_schema.json
• src/main.js
• src/extractor.js
• examples/sample-input.json
• examples/sample-output.json
• tests/smoke.test.js

examples/sample-output.json is a representative record set produced by the bundled offline fixture used in the smoke test.

Input

Example:

{
  "startUrls": [{ "url": "https://example.com" }],
  "domains": ["example.org"],
  "businesses": [
    {
      "name": "Acme Dental",
      "domain": "acmedental.test",
      "location": "Springfield, IL"
    }
  ],
  "keywordLocationPairs": [
    { "keyword": "dentist", "location": "Springfield, IL" }
  ],
  "maxPagesPerSite": 4,
  "requestTimeoutSecs": 15
}

Input notes:

• startUrls: best option when you already know the website
• domains: actor tries https:// first and falls back to http://
• businesses: accepts string names or objects with name, url, domain, location
• keywordLocationPairs: produces deterministic seed records if no public search discovery is available
• maxPagesPerSite: hard cap on same-origin pages fetched per site
• requestTimeoutSecs: per-request timeout

Output

Each dataset item is one business/site record. Representative fields:

{
  "recordType": "website",
  "sourceType": "startUrl",
  "businessName": "Acme Dental",
  "website": "https://acme.example",
  "homepageUrl": "https://acme.example",
  "contactPageUrl": "https://acme.example/contact",
  "phones": ["+1 555 222 3333"],
  "emails": ["hello@acme.example"],
  "socialLinks": ["https://www.facebook.com/acme"],
  "title": "Acme Dental | Family Dentist in Springfield",
  "metaDescription": "Acme Dental provides family dentistry in Springfield.",
  "schemaOrg": [],
  "ratingValue": 4.8,
  "reviewCount": 127,
  "flags": {
    "noWebsite": false,
    "noHttps": false,
    "missingContactPage": false,
    "missingSocials": false,
    "noVisibleEmail": false,
    "weakTitleMeta": false,
    "outdatedCopyright": false
  },
  "suggestedOutreachAngle": "Lead with a concise website improvement pitch: ..."
}

The actor also stores a SUMMARY object in the key-value store with totals and flag counts.

Local Usage

Install:

$npm install

Run tests:

$npm test

Run smoke test only:

$npm run smoke

Run actor locally with Apify local storage:

$APIFY_LOCAL_STORAGE_DIR=./storage npm start -- --input=examples/sample-input.json

If you prefer, you can also place the input in storage/key_value_stores/default/INPUT.json and run npm start.

Publish Command

When ready to publish to Apify:

$apify push

Limitations

• Website/domain inputs are the robust path
• keywordLocationPairs do not use Google login, Maps login, or brittle authenticated scraping
• Pages outside the original origin are ignored
• The actor is intentionally bounded and may miss deep contact details hidden behind complex JS apps or forms
• outdatedCopyright is heuristic and based on visible page text