Hotfrog Real-Time Data

Live Hotfrog business directory data — keyword + location search, profile scraping, and lead details — streamed to your dataset as structured JSON in real time.

Free trial: free Apify plan users get a 2-result sample per run. Upgrade to a paid Apify plan for full, unlimited data.

Who this is for

Agencies, local-SEO teams, sales-outreach operators, and CRM builders who need fresh business lists with contact data from the Hotfrog directory.

Use cases:

• Build targeted lead lists by keyword and city (e.g. plumbers in Austin, dentists in Chicago)
• Enrich business records with phones, addresses, websites, and emails for outreach
• Feed CRMs, Google Sheets, or pipelines via webhooks (JSON or Slack)
• Monitor local markets — track which businesses appear per keyword and location
• Power AI agents and chatbots with live local-business answers via Apify MCP
• Run national sweeps with the "US" location or drill into specific cities

Features

Checkbox	What it does
🔎 Business search (default on)	Keyword + location task table; each row yields name, address, phone, categories, and tags when listed
⚡ Simple search	One keyword across several semicolon-separated cities (used only when the task table is empty)
🔗 Scrape By URL	Paste Hotfrog business profile URLs directly for the same output
🎯 Lead details (default on)	Enriches every row with its full profile — verified phone, address, description, tags, website/email when publicly listed, plus geo coordinates. Enriches in place: no second row, no dropped records
🌐 Connection	Apify residential proxy on by default; switch country or custom proxies if needed
⚙️ Output & limits	Global caps, pacing, parallel lookups, and webhook delivery

Every discovered business is exported. Lead details enrich the same row (detailsFetched: true) and never filter results, so output counts — and budgeting — stay predictable.

Input reference

• searchTasks (array, default: plumbers → Austin, TX, roofing → Dallas, TX): one row per search. Each row: keyword (e.g. plumbers, roofing, restaurants, dentists), location ("Austin, TX" with state, or "US" for a national search), maxResults (per-task override).
• maxResultsPerTask (integer, default 10): default cap per search task.
• maxPagesPerTask (integer, default 5, max 50): page-depth safety cap per task.
• enableSimpleSearch (boolean, default false), simpleKeyword, simpleLocations ("Austin, TX; Dallas, TX" or "US"): alternative to the task table.
• country (US | UK | CA | AU, default US): which Hotfrog market to search.
• enableScrapeByUrl (boolean, default false), scrapeUrls (string list): Hotfrog /company/… profile URLs.
• enableLeadDetails (boolean, default true): full-profile enrichment per business (adds a little extra time per record).
• maxItems (integer, default 10000): global row cap. Also capped to 2 on the free tier.
• concurrency (1–8, default 4), delayBetweenRequestsMs (0–5000, default 1200).
• webhookUrl (string, default ""), webhookFormat (json | slack): additional real-time POST per record; dataset writes always happen.
• proxyConfiguration: Apify residential proxy (US) by default.

Empty input still runs a useful demo: business search with the prefilled tasks above.

Output field reference

One row per business. Key fields:

type, featureType (search | scrape_by_url), platform (hotfrog), name, profileUrl, businessId, phone, phones, website, email, emails, street, city, state, zipCode, location, country, category, categories, tags, description, imageUrl, reviewCount, rating, latitude, longitude, detailsFetched, searchKeyword, searchLocation, searchTaskIndex, searchTaskLabel, position, scrapedAt.

The run summary (OUTPUT in the key-value store) reports totalPushed, tasksRequested/Completed, leadDetailsEnabled, spendingLimitReached, and the paywall object (detected, isPaying, pricingTier, limited, blocked).

Webhooks

Set webhookUrl to also POST each record as it is collected — point it at your CRM, a Slack incoming webhook, Zapier, Make, or a Sheets bridge. Choose json for the full record or slack for a chat-friendly message. Delivery is fire-and-forget: failures are logged and never stop the run or dataset writes.

MCP / AI-agent usage

This Actor works with the Apify MCP server: connect it once and ask in plain language, for example:

"Find 10 roofing businesses in Dallas, TX on Hotfrog with phones and addresses, and save them to a dataset."

FAQ

• Which connection does it use? Apify residential proxy (US default) on the platform; the market country picker selects the Hotfrog market. No setup needed.
• How fresh is the data? Every run collects live directory data at run time — nothing is cached or pre-stored.
• What does the free tier include? A 2-result sample per run so you can verify the shape before upgrading. Paid plans are uncapped (subject to your own maxItems and run charge limits).
• How am I charged? One result event per exported row. Lead details enrich the same row — no double charge. When your max run charge is reached, collection stops gracefully and the summary flags spendingLimitReached.
• Why did a search return nothing? Keywords follow the directory's own taxonomy (e.g. plumbers vs plumbing can differ per city). Try the plural/category form, or broaden the location to "US".