Company Name to Domain Finder - Bulk Website & Domain Lookup

Turn a list of company names into official website domains. Paste names (or a CSV export from your CRM) and get the best official domain per company with a 0-100 confidence score, a verified live homepage, the company's LinkedIn URL, logo, favicon, homepage title & description, redirects and alternative candidates.

A cheap, accurate Clearbit Name-to-Domain API alternative (Clearbit's free name-to-domain access was sunset in 2025). Built for lead generation, CRM cleanup, data enrichment, Clay / n8n / Make / Zapier workflows and AI agents.

• ✅ Pay only for resolved companies - companies without a confident match are free
• ✅ Confidence score on every result, plus alternatives so you can review edge cases
• ✅ Verified: the domain resolves, the homepage loads and mentions the company; parked and for-sale domains are rejected
• ✅ Never returns directories: LinkedIn, Crunchbase, Facebook, Wikipedia, Yelp, Bloomberg, ZoomInfo, etc. are filtered out
• ✅ Hints for ambiguous names: Delta | airline, Bolt | Estonia, Joe's Pizza | Brooklyn
• ✅ Fast and cheap: runs on 256 MB memory and resolves ~1,500 companies per hour by default (raise concurrency for more)

How it works

For every company the Actor combines several independent signals and only returns a domain when they agree:

1. Candidate sources - free public company autocomplete (Clearbit), the official website recorded on Wikidata (property P856), and domain guesses from the name (e.g. Warby Parker → warbyparker.com).
2. Web search fallback - only when the free sources are not confident (Bing / DuckDuckGo via Apify Proxy, or Google via Apify's SERP proxy).
3. Verification - DNS lookup, homepage fetch with redirects, and fuzzy matching of the company name against the page title, og:site_name, schema.org Organization name and copyright line. Parked/for-sale pages, dead domains and aggregator sites are penalized; your country / city / industry hints are checked against the domain TLD and page text.
4. Confidence score - the signals are combined into a 0-100 score. Results below your minConfidence (default 60) are returned as low_confidence / not_found and are not charged.

No LinkedIn scraping and no login-protected data - only public sources and the companies' own homepages.

Accuracy

Measured on hand-labelled test sets of company names with known domains (big brands, mid-size SaaS, non-US companies, ambiguous one-word names with a hint, small local businesses, and fake names that must return nothing):

Test set	Companies	Returned at confidence ≥ 60	Precision (correct / returned)
Development set	68	66 (97%)	100%
Hold-out set	52	50 (96%)	100%
Unseen final set	41	39 (95%)	97.4%
All	161	155 (96%)	99.4%

At minConfidence: 80 precision was 100% (152 of 161 returned). Measured on the Apify platform with default settings (Bing/DuckDuckGo fallback). The "not returned" rows are fake names (which must return nothing) and a few highly ambiguous one-word names.

Fake company names correctly returned nothing. Precision = share of returned domains that are the correct official domain. Raise minConfidence to 80 if you prefer fewer results with even fewer mistakes.

Input

Field	Description
companies	One company per line, optional hint after `
records	JSON objects like {"name": "Bolt", "country": "Estonia", "industry": "ride-hailing", "id": "crm-1"} - returned in inputRecord
csvData	Paste CSV with a name/company column (plus optional country, city, industry) or headerless rows name, hint...
country	Default country for all rows, e.g. Germany or DE
minConfidence	0-100, default 60
searchFallback	auto (default), always, never
searchEngine	bing (Bing/DuckDuckGo, default) or google (Apify Google SERP proxy - best for small local businesses)

Example input:

{
    "companies": ["Stripe", "Spotify", "Delta | airline", "Mercury | banking for startups", "Franklin Barbecue | Austin, TX"],
    "records": [{ "name": "Bolt", "country": "Estonia", "industry": "ride-hailing", "id": "crm-123" }],
    "minConfidence": 60
}

Output

One row per company. Example:

{
    "input": "Mercury | banking for startups",
    "companyName": "Mercury",
    "status": "found",
    "domain": "mercury.com",
    "confidence": 97,
    "website": "https://mercury.com/",
    "companyNameFound": "Mercury",
    "title": "Online Business Banking For Startups, Small Businesses & Enterprises | Mercury",
    "description": "Mercury is banking built for startups ...",
    "linkedinUrl": "https://www.linkedin.com/company/mercuryhq",
    "logoUrl": "https://www.datocms-assets.com/.../mercury-logo.svg",
    "faviconUrl": "https://mercury.com/favicon.ico",
    "finalUrl": "https://mercury.com/",
    "redirectedFrom": null,
    "isLive": true,
    "httpStatus": 200,
    "sources": ["clearbit", "guess", "search"],
    "wikidataId": null,
    "alternatives": [
        { "domain": "mercuryinsurance.com", "confidence": 41, "sources": ["clearbit"], "title": "Mercury Insurance ...", "isLive": true, "isParked": false }
    ],
    "hint": { "text": "banking for startups" },
    "inputRecord": null,
    "error": null,
    "checkedAt": "2026-10-08T08:00:00Z"
}

status is one of:

• found - domain returned with confidence ≥ minConfidence (charged)
• low_confidence - a best guess exists but is below the threshold (see bestGuess and alternatives, free)
• not_found - nothing credible (free)
• error - the row failed after retries (free)

You can download the dataset as JSON, CSV, Excel or use the API. A run summary (counts, requests) is stored in the key-value store under SUMMARY.

Pricing

This Actor uses pay-per-event pricing – you pay for successfully resolved companies, plus optional Google search charges when that fallback is enabled.

Event	Price
Company resolved (company-result)	$1.50 per 1,000 companies ($0.0015 per company)
Actor start (actor-start)	$0.0001 per run
Google search (google-search)	$0.003 per search

Companies that are not found, below your confidence threshold, or failed are free. Google search is billed only when you set the search fallback to Google (typically for less than half of companies). Platform usage (compute) is included. Set Maximum cost per run in the run options to cap your spend – the Actor stops cleanly at the limit.

Tips

• Add hints for short or common names. Square | payments, Front | customer service software, Joe's Pizza | New York. A country hint also prefers that country's domain (.de, .co.uk).
• Use records/CSV to keep your IDs. Every output row carries your original record in inputRecord.
• Small local businesses: choose searchEngine: google for the best coverage.
• Strict mode: minConfidence: 80 for near-zero false positives; 0 to get a best guess for everything.

Related tools

Lead enrichment chain: company name → domain → contacts → tech stack → authority and traffic.

• Website Contact Scraper: get emails, phone numbers and social links for each domain
• Tech Stack Detector: see which CMS, ecommerce, analytics and CRM tools a site uses
• Domain Authority & Traffic Checker: bulk domain authority, backlinks and organic traffic
• ATS Jobs API (Greenhouse, Lever, Ashby, Workday): pull every open job straight from company career sites

All of these are maintained by the same developer and export flat JSON, CSV or Excel, so you can feed one tool's dataset into the next.

FAQ

Is this a Clearbit Name to Domain / Autocomplete replacement? Yes - same job (company name in, domain out), plus a confidence score, verification of the live homepage, alternatives, LinkedIn URL and logo, and you can feed thousands of names in one run.

How accurate is it? On our test sets 99.4% of returned domains were correct at the default threshold, and ~96% of companies were resolved. Big brands, funded startups and well-known local businesses resolve best; very small businesses with generic names need a city or industry hint.

Do I pay for companies that are not found? No. Only rows with status: found are charged.

Does it scrape LinkedIn? No. The LinkedIn URL is taken only from links on the company's own homepage.

What about parked domains or companies that changed their domain? Parked/for-sale pages are rejected. Redirects are followed and the final domain is returned, with the original candidate in redirectedFrom.

Can I call it from Clay, n8n, Make or Zapier? Yes - use the Apify integration or the Run Actor synchronously API endpoint and read the dataset items.

Why is a well-known company returned with confidence below 100? Some corporate sites block automated visitors, so the homepage cannot be read. The domain is still returned when the other sources agree, with a slightly lower score.