UK B2B Lead Finder

🔍 What does UK B2B Lead Finder do?

Type what you sell to and where, and get back businesses in that market with contact details you can actually use. Say "marketing agencies" and "Manchester", or "accountants" and "Leeds", and UK B2B Lead Finder finds the active companies in that market on the Companies House register, proves each company's own website, and reads the emails, phone numbers and social profiles the business publishes there.

The part that makes it a lead finder rather than a company list is the middle step. Every website in the output has passed an evidence gate: the page has to show the company's name and at least one signal tying it to the registered company, its postcode, its Companies House number, or its town. A domain that merely looks right is rejected, and every row records exactly what evidence proved the match, so you can filter on confidence.

And the pricing follows the product: you pay per delivered lead, and a lead always carries at least one contact. Companies that were found but could not be matched to a website, and websites that turned out to have no contact details, are counted in the run summary and cost you nothing.

📊 What data can I extract?

Field	What it holds
🏢 companyName	Registered company name from Companies House
🔢 companyNumber	Companies House company number
📅 incorporatedDate	Date the company was incorporated
🏷️ sicCodes	SIC 2007 codes the company registered under
📮 registeredAddress	Registered office address as published
🏙️ registeredTown	Town read off the registered address
📍 registeredPostcode	Postcode read off the registered address
🌐 domain	The company's own website domain, proven by the evidence gate
🔗 website	Full URL of the proven homepage
🧭 resolutionMethod	How the domain was found: slug-exact, slug-fuzzy or search
🧾 resolutionEvidence	What proved the match, for example name-legal+postcode+town
✉️ emails	Every address found, classified role, personal or unclassified, with free provider and on-domain flags
📞 phones	Phone numbers, tel links first, with their source
📱 socials	Social profile links by platform
📄 pagesCrawled	How many pages of the site were read
🗃️ source	Where the row was assembled from
⏱️ retrievedAt	When this run collected the row

🆚 What about the official route?

Companies House is open data and its advanced search is genuinely good: free, filterable by SIC code and registered address, with a CSV download. If all you want is a list of registered companies, use it, it costs nothing.

What the register does not hold is any way to contact those companies. No websites, no emails, no phone numbers. Turning "1,548 active advertising agencies registered in Manchester" into people you can reach means finding each company's own website and reading its contact pages, which is the work this Actor does for you, with the evidence for every match recorded.

Worth saying plainly: this data is assembled from Companies House and the open web, and nothing else. There is no directory subscription, no purchased contact database and no private source behind it. Every contact detail in the output was published by the business itself on its own website.

💡 Why use UK B2B Lead Finder for lead generation?

Because the leads come from the primary source. Bought lists decay and directories lag, but the Companies House register is updated by the companies themselves, and the contact details are read from each company's live website on the day you run it. You see when each company was incorporated, so you can target new businesses in their first year, and you see the evidence behind every website match instead of taking a database's word for it.

• Agencies prospecting a city. "Dentists in Leeds" or "restaurants in Brighton" as a fresh, sourced list with emails and phones.
• New business alerts. Filter the output by incorporatedDate and greet companies formed this quarter.
• Suppliers selling to a trade. Builders, electricians and plumbers in a postcode area, with the phone numbers they publish.
• Market sizing with reachable names. Not just how many marketing agencies Manchester has, but which ones you can talk to.

🚀 How do I use UK B2B Lead Finder?

1. Open the Actor and type an industry in plain words, for example marketing agencies, accountants, dentists, builders.
2. Type a location: a UK town, city or postcode area.
3. Set maxLeads to how many contactable leads you want.
4. Run it. The Actor reports its funnel as it goes: discovered, checked, proven, delivered.
5. Export the dataset as CSV, Excel or JSON, or read it over the API.

Around 65 industries are mapped to their official SIC codes. If your keywords are not in the mapping, the run stops immediately and names the nearest supported industries, so a typo never silently searches the whole register. You can also pass exact SIC codes yourself.

⬇️ Input

Field	Type	What it does
industry	string	The kind of business, in plain words. Mapped to official SIC codes
location	string	UK town, city or postcode area, matched on the registered office address
maxLeads	integer	How many contactable leads to deliver, 1 to 100. Default 10
maxCompaniesScanned	integer	Hard cap on how many discovered companies the run works through, up to 2,000. Default 10 times maxLeads
sicCodes	array	Exact SIC 2007 codes, instead of or as well as the keywords
audienceType	select	b2b or local-consumer. Both draw on the same register, this only orders industry suggestions
requireEmail	boolean	When on, a lead must carry an email. When off, an email or a phone qualifies
useSearchFallback	boolean	One polite web search per unresolved company, same evidence gate applied

{
  "industry": "Marketing agencies",
  "location": "Manchester",
  "maxLeads": 10,
  "requireEmail": false
}

⬆️ Output

Table view

One row per delivered lead, with the company, its proven domain, the contact details and the evidence side by side.

Run summary

Alongside the dataset, every run writes a RUN_SUMMARY record to its key-value store with the whole funnel in numbers: companies discovered, scanned, websites proven, resolved but contactless, and leads delivered, plus a count of HTTP requests by stage. It is the honest account of the work behind the leads, and it is where to look when a market delivers fewer than you asked for.

JSON

{
  "companyName": "FULL SPHERE LTD",
  "companyNumber": "12669352",
  "incorporatedDate": "2020-06-13",
  "sicCodes": ["73110", "74100"],
  "registeredAddress": "17 St Anns Square Full Sphere Limited Manchester England M2 7PW",
  "registeredTown": "Manchester",
  "registeredPostcode": "M2 7PW",
  "domain": "fullsphere.co.uk",
  "website": "https://fullsphere.co.uk/",
  "resolutionMethod": "slug-exact",
  "resolutionEvidence": "name-legal+postcode+town",
  "emails": [
    { "email": "hello@fullsphere.co.uk", "type": "role", "isFreeProvider": false, "onSiteDomain": true }
  ],
  "phones": [{ "phone": "0330 133 2279", "source": "tel-link" }],
  "socials": { "linkedin": ["https://www.linkedin.com/company/fullsphere"] },
  "pagesCrawled": 3,
  "source": "Companies House advanced search + the company website",
  "retrievedAt": "2026-08-18T10:00:00.000Z"
}

🧭 How the evidence gate works

Matching a company name to a domain is a judgement, not a lookup, and the failure mode is expensive: a plausible wrong domain in an outreach list. So a candidate website is accepted only when the page shows the company name and at least one corroborating signal:

• the registered postcode appears on the page, or
• the Companies House number appears on the page, or
• the town from the registered address or your search appears, and only when the town is not part of the company's own name.

The measured case this design encodes: DAWLISH LIMITED, a London company, must not match dawlish.com, the Devon town website, however well the name fits. Name similarity alone is never enough.

Two further rules keep listing sites out of the output, because a directory's page about a company carries the company's name, town and often its number, which is exactly what corroboration looks for. A page that offers the reader other businesses is refused outright, and a domain that matches a second company in the same run is treated as a site that publishes company data rather than as anyone's own website, and blocked for the rest of the run. A domain found by search must also carry the company on its homepage, not only on one deep page.

Each row tells you how its domain was found (slug-exact from the full registered name, slug-fuzzy from a shortened form, search from one web query) and which evidence fired. Rows with postcode or company-number evidence are the strongest; if you want only those, filter resolutionEvidence in your export.

🎯 Honest limits

• Coverage follows web presence. A large share of registered small companies have no findable website at all. Those companies are counted in the run summary and never charged for. Expect the delivered leads to be a minority of the companies discovered, that is the nature of the register, not a fault in the search.
• The location filter matches the registered office. A company can trade in Manchester and be registered elsewhere, and vice versa. Registered office is how Companies House itself filters, and it is what this Actor uses. The register's filter is a text match against the whole registered address, so it can occasionally match the wrong part of one: a search for the M1 postcode area also returns a company at "Unit M1" in Amersham, and a town name can match a street named after the town.
• Very large markets are sampled from the top. The register serves at most 5,000 rows per filtered download and displays at most 10,000. Narrow the location or the industry to reach the rest.
• audienceType does not change the data source. B2B and local consumer businesses both come from the same register; the switch only orders industry suggestions.

⏱️ How long does a run take?

Measured on real runs, so you know what normal looks like and can tell it apart from a run that has stalled.

Leads delivered	Typical run time
5	25 seconds to 4 minutes
10	1 to 2 minutes
25	3 to 4 minutes

This is the one Actor here where run time genuinely does not track the output, and the reason is the evidence gate. The work is done per company scanned, not per lead delivered: every candidate from the Companies House register has its website found, read and checked for proof it belongs to that company, and a company that fails the gate has cost the same page loads as one that passes while adding no row. So a market with a high hit rate settles quickly and a thin one takes minutes to find the same number of leads. Five electricians in Kendal took 23 seconds. Five plumbers in Wigan took over four minutes.

maxCompaniesScanned is the setting that bounds this. It caps the work rather than the output, so it is the right lever when you want a run to finish inside a schedule window. requireEmail does not slow a run down, it just discards more of what was already fetched. The first few seconds of any run are the container starting rather than the work.

A run is never silently stuck. Every company is logged as it is scanned, with whether it passed the evidence gate and why, and if a run reaches its time limit it stops early, keeps every lead already found and says how many companies were still unscanned in the status message.

Set the run timeout to suit the size of the ask. This Actor's default is 3600 seconds, which is comfortably more than the largest run in the table above (50 leads takes about 2 minutes). You are charged per delivered result rather than per minute, so a generous timeout costs you nothing and a tight one risks losing the run's work. Lower it only if you want a hard ceiling on how long a scheduled run may sit.

💰 How much does it cost?

You are charged per delivered lead, and only for leads. A lead is a company with a proven website and at least one contact detail, an email or a phone number, or an email specifically if you set requireEmail. Companies discovered but not resolved, and websites with no contact details, cost nothing however much work they took. maxLeads is therefore also a hard cap on what a run can cost you. The current price per lead is on this page's pricing tab.

🔌 Integrations

UK B2B Lead Finder runs on Apify, so you can schedule it (say, first Monday of the month for each city you sell into), push results to Google Sheets, HubSpot, Salesforce or anywhere else through Apify integrations, trigger it from Make or Zapier, and set webhooks on run completion.

It is also callable by AI agents through the Apify MCP server. An agent asked to "find me twenty dentists in Leeds with emails" can discover and run spookyweb/uk-b2b-lead-finder directly, read the structured output, and chain it into whatever it is building. The output schema is declared, so agents can consume the dataset without guessing its shape.

🔗 Using UK B2B Lead Finder with the Apify API

curl "https://api.apify.com/v2/acts/spookyweb~uk-b2b-lead-finder/run-sync-get-dataset-items?token=YOUR_TOKEN" \
  -X POST \
  -H "Content-Type: application/json" \
  -d '{ "industry": "Accountants", "location": "Leeds", "maxLeads": 20 }'

The run's dataset is plain JSON, one object per lead, and the Apify API gives you the same data as CSV or Excel with a query parameter.

❓ FAQ

Which industries are supported?

Around 65, each mapped to its official SIC 2007 codes from the Companies House condensed list: agencies and professional services (marketing, PR, software, IT, accountancy, legal, recruitment, architecture), trades (builders, electricians, plumbers, roofers, joiners, landscapers), health (dentists, GPs, physios, opticians, vets, care homes), hospitality (restaurants, cafes, takeaways, pubs, hotels) and more. An unmapped industry is rejected with the nearest supported ones named, and you can always pass SIC codes directly.

What exactly am I charged for?

One charge per delivered lead, nothing else. If a run discovers 400 companies and delivers 12 leads, you pay for 12. Empty and unresolvable companies are reported in the run summary, not the invoice.

Why did my run deliver fewer leads than maxLeads?

Because the market ran out of provable, contactable companies before the target was reached, or the run hit its maxCompaniesScanned cap first. The status message shows the funnel: how many companies were discovered, how many were checked, how many websites were proven and how many carried contacts. Try a wider location, a broader industry, or a higher maxCompaniesScanned.

How accurate are the website matches?

Every match carries its evidence. slug-exact with postcode or company-number corroboration is as strong as this gets without a human check. slug-fuzzy with only town corroboration is weaker, and the resolutionEvidence field exists precisely so you can drop those if your outreach demands certainty.

Are Gmail and other free provider addresses included?

Yes, with isFreeProvider set true so you can filter them. Small businesses often publish a free mailbox as their only contact, and throwing it away would lose real leads.

Can I search by postcode area instead of a town?

Yes, checked against the live register: a search for M1 with one SIC code returned 85 companies and LS1 returned 83, almost all at addresses in that postcode area, and a full postcode works too. The register matches the location as text against the whole registered address, so a postcode area search can pick up the odd stray, such as a company whose address contains "Unit M1", and those strays still pass or fail the evidence gate on their own merits.

Does it use a browser?

No. Everything is plain HTTP: the register pages, the candidate homepages and the contact pages. That keeps runs fast and cheap, at the cost of missing the rare site that renders its contact details only in JavaScript.

⚖️ Is it legal to collect these leads?

The company data comes from Companies House, which publishes it as open government data precisely so the public can know who they are dealing with. The contact details are read only from each business's own public website, where the business published them so customers could get in touch, which makes them business contact information.

Where a detail identifies a person, for example a named sole trader's mailbox, UK GDPR applies and you are the data controller for what you do next. You need a lawful basis, typically legitimate interest for B2B outreach, you must honour opt outs, and PECR governs how you may approach them. Apify's ethical scraping guide covers the wider picture. Nothing here reads anything behind a login.

👍 Your feedback

Found a bug, want an industry added to the mapping, or need a field that is not here? Open an issue on the Actor's Issues tab. Industry mapping requests are quick to ship, and problems get fixed fast.

🔎 You might also like

• Website Contact Scraper for contact details when you already have the websites
• Company Email Finder to find work email patterns for a specific domain
• Company Domain Finder to resolve a known company name to its domain
• UK Disqualified Directors to screen who you are about to do business with