Apify Booking Host Leads Agent Skill

Booking.com host & operator lead finder

Turn a Booking.com destination search into a clean table of accommodation hosts / property managers with their real contact details.

The key insight

When you scrape Booking.com, the host's contact email is sometimes already in the data under traderInfo (a trader-transparency disclosure). But this is region-dependent: strong in the EU (Zurich apartments: 80%) and Australia (Melbourne apartments: 82% in a 50-property test), and increasingly present in Asia (Beijing apartments: 65% as of June 2026, typically a QQ, 163, or Gmail address plus a Chinese company name). It is historically sparser in the US and UK. A naive chained Google Maps email scraper found emails for only ~10% and often matched the wrong business.

So use a waterfall: try the cheapest source first, fall through to more expensive ones only for leads still missing a contact. Never start with a Google Maps email extractor.

| Tier | Source | Gets | Fires when |

|------|--------|------|-----------|

| 1 | Booking traderInfo | email, phone, company | always (free, already scraped) |

| 2 | Google Maps Email Extractor (by name) | email, phone, socials, website | lead still has no email |

| 3 | Google Search (by name) | a candidate website | lead still has no website |

| 4 | Website contact scraper (on the website) | email / phone / socials | lead has a website but still no email |

Prerequisites

• Apify account (sign up)
• Authentication via one of:
• apify login (OAuth, if using the Apify CLI)
• APIFY_TOKEN environment variable
• Token from Apify Console → Settings → Integrations

Workflow

• Collect inputs. Ask the user for: destination (city/area), property type (Apartments / Guest houses / Holiday homes / Villas surface the most independent operators; none = all), max properties, and output format (CSV / JSON / Google Sheet).
• Tier 1 - Scrape Booking.com. Run voyager/booking-scraper. Build a lead row per property from traderInfo and hostInfo (see Mapping below). Mark rows where traderInfo.email is present as contactSource: booking-trader.
• Tier 2 - Google Maps (only rows still missing an email). Run lukaskrivka/google-maps-with-contact-details with searchStringsArray = ["<name>, <destination>", ...]. Fill email/phone/socials and capture any website it returns. Mark filled rows google-maps.
• Tier 3 - Google Search (only rows still missing a website). Run apify/google-search-scraper with one query per lead ("<name> <city> official website"). Take the first non-OTA/non-social organicResults[].url as the candidate website.
• Tier 4 - scrape the website (only rows with a website but no email). Default: vdrmota/contact-info-scraper (deterministic, ~14s/site). Optional: apify/ai-web-scraper with a prompt for email/phone/contact-form URL - but in live testing it was slow and failed on multiple sites, so prefer the contact scraper. Mark filled rows website-contact (or website-ai). Cap the count to control cost.
• Deliver. Output one row per property. Report total rows, % with an email, and the contactSource breakdown. Rows still empty keep BookingURL (contact via Booking messaging).

Mapping (Booking item → lead row)

| Output column | Source field |

|---------------|--------------|

| name, type, address, city, country, rating, ratingLabel, stars, reviews, BookingURL | top-level Booking fields (address.full, address.city, address.country, rating, ratingLabel, reviews, url) |

| email, phone | traderInfo.email / traderInfo.phone → fallback Google Maps emails[0] / phones[0] |

| companyName, registrationNumber, traderAddress | traderInfo.companyName / .registrationNumber / .address |

| hostName, managedProperties | hostInfo.name / hostInfo.managedProperties (operator-size signal) |

| website, socials | Google Maps fallback only; drop OTA domains (booking.com, agoda, expedia, etc.) |

| contactSource | booking-trader \| google-maps \| none |

Put email and phone immediately after address in the output.

Always return BookingURL (link to the property), rating, ratingLabel (e.g. "Superb", "Exceptional"), and reviews (review count) for every lead. These give each row its own quality signal, so leads can be ranked and prioritized without re-opening Booking.

Actor routing

| Waterfall tier | Actor ID | Maintainer | Best for |

|----------------|----------|------------|----------|

| 1 - Booking + trader contacts | voyager/booking-scraper | community | Primary. traderInfo holds host email/phone/company |

| 2 - email/social/website by name | lukaskrivka/google-maps-with-contact-details | community | Emails Booking doesn't disclose; also returns a website |

| 3 - discover a website | apify/google-search-scraper | apify | Finding the host's official site when Maps has none |

| 4 - extract from the website (default) | vdrmota/contact-info-scraper | community | Deterministic email/phone/social extraction; fast and reliable |

| 4 - extract from the website (alt) | apify/ai-web-scraper | apify | LLM extraction via prompt; flexible but slow/unreliable in testing |

Prefer apify-maintained Actors where available.

Calling Actors — Apify CLI

Three flags on every call (--json, --user-agent, 2>/dev/null):

1. Scrape Booking.com

apify actors call "voyager/booking-scraper" \

-i '{"search":"Melbourne","maxItems":50,"propertyType":"Apartments","currency":"USD","language":"en-us","sortBy":"distance_from_search","rooms":1,"adults":2}' \

--json \

--user-agent apify-awesome-skills/apify-booking-host-leads \

2>/dev/null

Tier 2 - enrich names that have no traderInfo email

apify actors call "lukaskrivka/google-maps-with-contact-details" \

-i '{"searchStringsArray":["<name>, Melbourne"],"locationQuery":"Melbourne","maxCrawledPlacesPerSearch":1,"language":"en"}' \

--json \

--user-agent apify-awesome-skills/apify-booking-host-leads \

2>/dev/null

Tier 3 - discover a website for leads that still have none

apify actors call "apify/google-search-scraper" \

-i '{"queries":"<name> Melbourne official website","maxPagesPerQuery":1,"resultsPerPage":5,"countryCode":"us","languageCode":"en"}' \

--json \

--user-agent apify-awesome-skills/apify-booking-host-leads \

2>/dev/null

Tier 4 - AI-extract a contact email / form URL from the website

(confirm the prompt/instructions field name against the Actor's input schema)

apify actors call "apify/ai-web-scraper" \

-i '{"startUrls":[{"url":"<website>"}],"prompt":"Extract the contact email, phone, and contact-form URL. Return JSON keys: email, phone, contactFormUrl."}' \

--json \

--user-agent apify-awesome-skills/apify-booking-host-leads \

2>/dev/null

Inspect input schema / fetch results

apify actors info "voyager/booking-scraper" --input --json \

--user-agent apify-awesome-skills/apify-booking-host-leads 2>/dev/null

apify datasets get-items DATASET_ID --format json \

--user-agent apify-awesome-skills/apify-booking-host-leads 2>/dev/null

The Apify MCP server (<https://mcp.apify.com>) and any MCP client (e.g. mcpc) are equivalent alternatives.

Troubleshooting

• No emails returned → you are probably reading Google Maps output instead of traderInfo. Re-check the Booking item's traderInfo field first.
• Emails for the wrong business → that is the Google Maps fallback matching a generic place. Trust contactSource: booking-trader rows; treat google-maps rows as lower confidence.
• Many contactSource: none rows → those hosts disclosed no trader info and have no Google Business match. Use the BookingURL to contact via Booking, or accept the blank.
• Run cost climbs → skip the Google Maps fallback entirely; traderInfo alone covers ~80%.
• See references/gotchas.md for cost notes.