Business Directory Lead Finder

Turn any business directory or listing page into an enriched, CRM-ready lead list — no API keys, no manual lookups.

Give it 1–50 URLs of business directory pages (Yelp-like listings, chamber-of-commerce pages, industry portals, Yellow Pages clones, event sponsor lists, vendor directories, etc.) and it returns a clean, deduplicated CSV of leads ready to import into your CRM.

What it does

1. Stage 1 — Directory parsing. A CheerioCrawler fetches each directory URL and auto-detects repeating listing blocks via three strategies: <table> rows, <ul>/<ol> items, and card-like <div> grids that share a common class. For each listing it extracts the business name, the external website link, any visible phone, and a basic address.
2. Stage 2 — Enrichment. For every business with a website, a second crawl visits the home page plus any /contact, /about, /impressum-style page it can find, and extracts email addresses, additional phones, and social profile URLs (Facebook, Instagram, LinkedIn, X, TikTok).
3. Deduplication. Leads are deduplicated by root domain across the whole run, so the same business appearing in two directories only shows up once.
4. Output. All leads are pushed to the Actor's Dataset as JSON. If you keep the default dataset_and_csv output format, a flat CSV is also written to the Key-Value Store as leads.csv for direct download.

Input

Field	Type	Default	Description
directoryUrls	array<string>	— (required)	1–50 URLs of business directory pages.
maxLeadsPerDirectory	integer	100	Cap on listings extracted per directory (1–5000).
enrichDepth	boolean	true	If false, skips Stage 2 and only returns what's visible on the directory page.
outputFormat	enum	dataset_and_csv	dataset_only or dataset_and_csv.

Example input

{
    "directoryUrls": [
        "https://www.yelp.com/search?find_desc=Restaurants&find_loc=New+York",
        "https://www.chamberofcommerce.com/united-states/new-york/restaurants"
    ],
    "maxLeadsPerDirectory": 50,
    "enrichDepth": true,
    "outputFormat": "dataset_and_csv"
}

Output

Each lead in the Dataset / CSV has the following fields:

Field	Notes
businessName	From the directory listing.
website	External URL of the business's own site.
domain	Normalized root domain (used for dedup).
phone	Best phone we found (Stage 1 or Stage 2).
address	Best-effort regex match from the listing.
email	First email found on the business site (Stage 2).
facebook / instagram / linkedin / x / tiktok	Social profile URLs (Stage 2). Multiple values are separated by ; in the CSV.
sourceDirectory	The directory URL the lead came from.
enriched	true if Stage 2 ran successfully for this lead.

Limitations & guarantees

• No external API keys. All crawling uses Crawlee's CheerioCrawler. No Google Maps, Hunter.io, Clearbit, or Apollo calls.
• No login walls, no CAPTCHAs. Pages that require authentication or anti-bot bypass are skipped and logged — never forced.
• No SMTP verification. Emails are format-validated only, not deliverability-checked.
• No sales claims. The Actor describes its technical behavior, not outcomes.

Pricing

Pay-per-result at $3 per 1,000 leads. You are only charged for the rows that actually appear in your final Dataset, after deduplication.

Local development

npm install
apify run --input "{\"directoryUrls\":[\"https://example.com/directory\"]}"