Company Name to Domain Finder — Logo, LinkedIn & Socials

Turn a list of company names into domains, and enrich every company from its own website, in one row:

• 🌐 Domain + match confidence — "Shopify" → shopify.com (0.99), acronym brands included, with the close alternatives listed so you can review edge cases.
• 🖼️ Logo and favicon — the logo the website itself publishes (its JSON-LD logo or header logo, else its app icon).
• 💼 LinkedIn company page — taken from the links on the company's own website (LinkedIn itself is never visited).
• 📣 Social profiles — X (Twitter), Facebook, Instagram, YouTube and TikTok.
• ✉️ Role mailboxes — info@, sales@, support@, hello@, press@ … on the company's own domain. Never the address of a named person.
• 🏛️ Company facts from Wikidata — headquarters country and city, founding year, industry, Wikidata id.

Misses are free: names that aren't found, matches below your confidence threshold, and websites that are blocked, parked or empty cost nothing. Already have domains? Give them instead of (or next to) names and only the website enrichment is charged.

Built for CRM clean-up, lead lists, Clay / n8n / Make workflows and AI agents that need "company name → website" resolution.

📦 What you get

One row per company found:

Field	Example
domain, matchConfidence, matchSource	shopify.com, 0.99, clearbit+wikidata
logoUrl, iconUrl	the site's own logo (or app icon) and its favicon
linkedinCompanyUrl	https://www.linkedin.com/company/shopify
twitterUrl, facebookUrl, instagramUrl, youtubeUrl, tiktokUrl	profile links
roleEmails, phones	["help@mercury.com", "press@mercury.com"], ["+18889975597"]
hqCountry, hqLocation, foundedYear, industry, wikidataId	Canada, Ottawa, 2006, ["e-commerce"], Q7501150
description, companyName, finalUrl	meta description, name, where the homepage landed
otherCandidates	up to 3 other domains the name could mean, with their scores
websiteStatus	ok, blocked, parked, unreachable, timeout, no_content …
fromWikidata	which fields came from Wikidata because the website didn't show them
billedEvents	exactly what this row was charged for

⚙️ Input examples

Names, domains, or both — one per line:

{ "companies": ["Shopify", "stripe.com", "HubSpot, hubspot.com", "Canva"] }

Names only, no website visit (fastest, cheapest):

{ "companies": ["Shopify", "Canva", "Wealthsimple"], "enrich": false }

A CRM import where a wrong domain is worse than none, with every miss kept as a row:

{ "companies": ["Delta", "Mercury", "Acme Plumbing"], "minConfidence": 0.8, "includeMisses": true }

Records from Clay / n8n / Make (any of these column names work: companyName, company, name, title, domain, website, url, country — a location / city column counts as a country hint only when it ends in a country name, like "Toronto, ON, Canada"):

{ "items": [{ "companyName": "Pizza Nova", "country": "Canada" }, { "title": "Mill Street Brewery", "website": "https://millstreetbrewery.com" }] }

A Google Sheet or CSV file, or the dataset of another Actor (e.g. Google Maps Scraper output, which has title + website):

{ "csvUrl": "https://docs.google.com/spreadsheets/d/<sheet-id>/edit#gid=0" }

{ "datasetId": "<dataset id>" }

Field	What it does
Companies	Names, domains / websites, or "Name, domain". Duplicates are looked up once.
Enrich from the company website	On (default): homepage (+ one about / contact page when needed) for logo, socials, LinkedIn, role mailboxes. Off: name → domain only.
Role mailboxes	On (default): collect info@ / sales@ / support@ … on the company's own domain.
Company facts from Wikidata	On (default): HQ country and city, founding year, industry; fills social links the site doesn't show. Free.
Minimum match confidence	0-1, default 0.6. Below it a name counts as not found (free).
Country hint	Favours the company headquartered in that country when several share a name. A country name or ISO code ("Canada", "CA", "CAN"); anything else is ignored.
Pages per website	1-3, default 2.
Include misses in the dataset	Also save a free row for each miss, with the reason in status.
Max companies	Look up at most this many (up to 50,000 per run).

💵 Pricing

Pay per result — misses are free.

Event	FREE / BRONZE	SILVER	GOLD / PLATINUM / DIAMOND
Domain found — a name matched at or above your minimum confidence	$0.0008	$0.0007	$0.0006
Company enriched — the website gave its own logo, its LinkedIn company page, a social profile that is clearly the company's, or a role mailbox	$0.0015	$0.0012	$0.001
Actor start — once per run	$0.001	$0.001	$0.001

A full row from a name costs $0.0023 ($0.0016 on GOLD and up); from a domain you already have, $0.0015 ($0.001). Not charged: names not found or below the minimum confidence, domains you supply (for the "domain found" event), websites that are blocked, parked, unreachable or empty, favicons, app icons and share images, social links that can't be tied to the company (a web designer's credit, a personal profile), Wikidata facts, phone numbers, miss rows.

❓ FAQ

How is the match confidence computed? Each candidate domain is scored on how closely its company name matches yours (legal forms like Inc., Ltd, GmbH are ignored), how well the domain fits the name (shopify.com for Shopify, ibm.com for International Business Machines), and its rank in the source. When Clearbit and Wikidata name the same domain the score goes up (up to 0.99); when two different domains score alike it goes down, and the runner-ups are listed in otherCandidates. A domain you give yourself is 1.

Where do the domains come from? Clearbit's public company autocomplete first, then Wikidata (the company's official website, property P856). Single-word or very common names ("Delta", "Mercury") can mean several companies: use a country hint, a higher minimum confidence, or give the domain.

I used the Clearbit Logo API / Clearbit autocomplete. Is this a replacement? For company name → domain and logos, yes: the logo comes from the company's own website (its published logo, else its app icon; the favicon is in iconUrl), and names resolve through Clearbit's autocomplete with Wikidata as a fallback. If Clearbit's autocomplete ever stops answering, runs keep working on Wikidata alone (fewer small-company matches) and the status message says so.

Do you scrape LinkedIn? No. LinkedIn is never visited. The LinkedIn company URL is the link the company's own website points to (or the LinkedIn id recorded on Wikidata). Only company pages (linkedin.com/company/…) are returned, never personal profiles.

What e-mail addresses do you return? Only role mailboxes on the company's own domain — info@, sales@, support@, hello@, press@, careers@ and similar. Addresses that could belong to a person (jane.doe@, jsmith@) and addresses on other domains (gmail.com etc.) are never collected.

Why is a website "blocked"? Some websites refuse automated visits from data centres (bot protection, or rate limits — many Shopify-hosted stores answer "Too many requests"). This Actor does not try to get around that — no retry from another IP address, no residential proxies — and the website enrichment is not charged. For a company you gave by name, the row is still saved (its domain was found) with websiteStatus: "blocked" and any social links Wikidata has. For a company you gave as a domain, there is nothing new to return, so it is a free miss: it appears in the dataset only with "Include misses in the dataset" on, and is always counted in the RUN_REPORT.

Can I send thousands of companies? Yes, up to 50,000 per run; ten companies are looked up in parallel. If a long stretch of the list finds nothing (100 in a row, or misses that would cost more than the run earns), the run stops early with a clear message instead of burning your time.

Is it legal? The Actor reads public pages that companies publish about themselves, Wikidata's open data and a public autocomplete endpoint, without logging in. You are responsible for how you use the data (for example, anti-spam laws such as CAN-SPAM, CASL and GDPR when e-mailing role mailboxes).