Google Maps Service Area Business (SAB) Finder

Find mobile and home-service businesses on Google Maps — the "service area" leads other scrapers miss. This actor specifically isolates Service Area Businesses (SABs): contractors, cleaners, plumbers, electricians, and other providers that don't have a public storefront but instead serve named regions (neighborhoods, suburbs, cities). Returns a clean, flat 11-field row per SAB, including the named service-area list.

Built for lead-generation agencies, home-service marketplaces (Angi/Thumbtack-style), and local SEO auditors that need the contractor leads standard Google Maps scrapers skip.

---

✨ Why this scraper

• SAB-focused — filters out brick-and-mortar storefronts by default; keeps only businesses with hidden street addresses (the "Service area" / "Online business" Maps signal).
• Named service areas — extracts the list of neighborhoods, suburbs, or cities each business serves.
• Pay-Per-Event — one flat event per saved row. Filtered-out storefronts and duplicate placeIds are not charged.
• Maps-only, no website visits — V1 stays fast and cheap by skipping external website crawling.
• CSV-friendly output — flat structure, no nested objects, drops cleanly into Sheets / Excel / CRMs.

---

🚀 Quick start — sample inputs

Example 1 — minimal (PRD example)

{
    "queries": ["House Cleaners in Denver", "Electricians in Boulder"],
    "maxResults": 20,
    "onlySABs": true,
    "minRating": 4.5
}

Example 2 — broader run with location bias and custom proxy

{
    "queries": ["Plumbers in Austin", "HVAC contractors", "Roofing companies"],
    "maxResults": 200,
    "onlySABs": true,
    "minRating": 4.0,
    "locationBias": "30.2672,-97.7431",
    "proxyConfiguration": {
        "useApifyProxy": false,
        "proxyUrls": ["http://user:pass@proxy.iproyal.com:12321"]
    }
}

The actor blocks Apify Residential proxy. If you need residential routing, supply your own provider via proxyConfiguration.proxyUrls as shown — see 🚦 Proxy policy below.

---

📦 Output

The dataset has one view: Service area businesses — a flat 11-column table.

Service area businesses — table view.

Output fields (11)

placeId, title, is_sab, service_areas, category, phone, website, rating, reviews, is_verified, scraped_at.

Sample record — Service area businesses

{
    "placeId": "0x89c2598f4023dfaf:0x6a162028a92125ff",
    "title": "Parka Electric",
    "is_sab": true,
    "service_areas": [],
    "category": "Electrician",
    "phone": "+17189245237",
    "website": "https://parkaelectric.com/",
    "rating": 4.9,
    "reviews": 40,
    "is_verified": false,
    "scraped_at": "2026-05-10T04:44:24.190Z"
}

Why is service_areas an empty array here? Many businesses are tagged as service-area on Maps but don't publicly list their served regions, or list them in a non-standard way the parser doesn't catch. The is_sab flag is still set correctly. See the FAQ entry below for details.

---

⚙️ Inputs

Field	Type	Default	Purpose
queries	array	required	Search terms (e.g. "Plumbers in Austin"). One row per search.
maxResults	integer	10	Total business rows pushed to the dataset across all queries (hard cap 5000).
onlySABs	boolean	true	Drop physical-storefront listings; keep only Service Area Businesses.
minRating	number	0	Drop businesses below this Google rating.
locationBias	string	""	lat,lng (e.g. "39.7392,-104.9903") or city name to bias every search.
proxyConfiguration	object	{ "useApifyProxy": true }	Standard Apify proxy editor. Apify Residential is rejected — see below.

---

💰 Pricing

Pay-Per-Event. One flat event per saved row — final per-event price is configured on the Apify console:

Event	Charged when
business-result	Once per unique business row that passed all filters and was successfully written to the dataset (per placeId).

Not charged:

• Duplicates (same placeId across queries — Crawlee dedupes the request queue).
• Rows filtered out by onlySABs (physical storefronts when onlySABs is on).
• Rows filtered out by minRating.
• Rows where the detail panel never loaded.
• Anything after the user-configured per-run spending cap is reached.

The actor honors the user-configured per-run spending cap (Apify eventChargeLimitReached) and stops cleanly when reached.

---

🚦 Proxy policy

Use Apify Datacenter proxy (the default) or no proxy for normal runs — both work reliably for Google Maps search and place detail panels at this actor's concurrency (max=5).

Apify Residential proxy is not supported. The actor will fail at startup if proxyConfiguration.apifyProxyGroups includes RESIDENTIAL. Reason: in pay-per-event actors, residential bandwidth (~$/GB) is billed to the developer, not the run user, so a single bandwidth-heavy run could exceed the per-result event revenue.

If you genuinely need residential routing, supply your own residential provider via the proxy editor's Custom proxy URLs field — that traffic goes through your provider, not Apify, and is unaffected:

http://user:pass@proxy.iproyal.com:12321
http://user:pass@proxy.brightdata.com:22225
http://user:pass@proxy.oxylabs.io:7777

---

📊 Run summary

After each run, a RUN_SUMMARY entry is written to the key-value store:

{
    "total_scanned": 47,
    "sabs_found": 31,
    "filtered_out": 19,
    "charged_events": 20,
    "runtime_seconds": 184,
    "scraped_at": "2026-05-09T15:00:00.000Z"
}

• total_scanned — place detail panels successfully opened.
• sabs_found — of those, how many were SABs (hidden address).
• filtered_out — rows dropped by onlySABs or minRating.
• charged_events — rows pushed to the dataset (= billable events).

---

⚙️ Filters

Filter	Stage	Effect
onlySABs	Pre-charge	Drop physical-storefront listings (default ON).
minRating	Post-extraction	Drop rows with rating < minRating.

Filters are applied before dataset push or event charges. Duplicate placeIds are deduped by Crawlee's request queue (each placeId is fetched at most once).

---

🚧 Limitations (V1)

• No external website visits — emails, social profiles, and on-site contact data are out of scope for V1 (see PRD §18 "Future Features").
• No service-area polygon coordinates — only the named area list (neighborhoods/cities). Polygon extraction is deferred to V2.
• SAB / verified detection is heuristic — Google Maps DOM evolves; we use stable data-item-id attributes and aria-label patterns where possible. Edge cases will exist.
• Per-run hard cap is 5,000 rows to bound runtime and cost.

---

❓ FAQ

What's a "Service Area Business" (SAB)? A business that doesn't operate from a public physical storefront — instead, on Google Maps, it shows a "Service area" label and a list of regions (neighborhoods, suburbs, cities) it serves. Plumbers, mobile cleaners, electricians, locksmiths, and most home-service contractors fall into this category.

Why is service_areas empty on some SABs? Some businesses are tagged as service-area on Maps but don't list their served regions, or list them in a non-standard way the heuristic doesn't catch. The is_sab flag is still set; only the service_areas array may be [].

Can I use Apify Residential proxy? No — the actor rejects Apify Residential at startup. Apify Datacenter, no proxy, and user-supplied custom proxy URLs all work fine. See 🚦 Proxy policy above.

How am I billed for runs that produce few results? Only for rows actually written to the dataset. A run that finds 100 candidates but filters 90 of them out (storefronts dropped by onlySABs) charges only 10 events.

Can I export to CSV? Yes — every field is flat (the service_areas array exports as a JSON-encoded string in CSV). Use Apify's CSV / Excel export from the dataset page, or call the dataset API with format=csv.

---

🛠️ Technical notes

• Stack: Node.js 18+ · Apify SDK 3 · Crawlee · Puppeteer (headless Chrome).
• Pipeline: single PuppeteerCrawler with two route handlers (search → place detail).
• Concurrency: min=1, max=5.
• Memory: 2 GB min · 4 GB default · 8 GB max (more memory ⇒ more Apify CPU shares ⇒ faster JS execution per page).
• Navigation: waitUntil: 'domcontentloaded'; navigation timeout 30 s; per-handler timeout 120 s.
• Resource blocking: aborts image, media, font resource types plus heavy URL fragments (/maps/vt/, lh3-6.googleusercontent.com, googletagmanager, fonts.gstatic.com, etc.) at the network layer to cut per-page CPU and bandwidth.
• Anti-blocking: explicit desktop Chrome User-Agent, synthetic mouse-wheel scroll on the search feed (real wheel events trigger Maps' lazy-load more reliably than programmatic scrollTop), --disable-blink-features=AutomationControlled Chrome flag.
• Diagnostics: On the first failed Maps render (no feed, no panel, or zero cards), the actor saves the page HTML + URL to the key-value store as debug-no-feed-html / debug-no-panel-html / debug-zero-cards-html.