US New Business Leads — Newly Registered Companies

US New Business Leads — newly registered companies, licences and funding signals

US New Business Leads delivers new business leads — newly registered US companies and newly licensed local businesses — straight from official open-data sources: state Secretary-of-State registries and major-city business-license feeds, with no scraping of protected sites, no login, and no API key. Schedule it daily and get a de-duplicated new business leads list: one row per business with company name, normalized address, registration date, entity type, NAICS-based industry tag, a lead score, and a contact person — with a phone number or email wherever the registry publishes it (nearly every new Connecticut company comes with an email).

Every phone number and e-mail in the feed is then checked before you get it: the number against the North American numbering plan, the e-mail domain against a live MX lookup, a throwaway-domain list and a role-address list — so bounces and dead numbers are labelled instead of sold to you as leads.

A third layer adds funding signals: companies that just raised private capital, from SEC EDGAR Form D filings — amount raised, date of first sale, investors, executive officers and a phone number.

Built for B2B sales: reach a business the week it registers — before it has a bank, an accountant, insurance, a website, or a marketing agency — or the week it closes a round, when it hires, buys software and outsources. Export scraped data, run the monitor via API, schedule runs, or push new leads straight into your CRM.

What is in this README

• What is US New Business Leads?
• Supported sources
• What data does it extract?
• The company website, found and confirmed
• Phone and e-mail checks built into the feed
• How much does it cost to get new business leads?
• How to set up a daily new-business lead feed
• ⬇️ Input
• ⬆️ Output
• Use cases for new business data
• Integrations and new business leads API
• Troubleshooting and error rows
• 🤖 For AI Agents & LLM Apps
• Newly registered companies, days after they file
• SEC Form D: funded startups with a phone number
• ❓ FAQ
• Also known as

What is US New Business Leads?

US New Business Leads is a newly registered businesses feed that reads official government sources — state and city open-data portals, the public Secretary-of-State registries of Florida, Idaho and West Virginia, and SEC EDGAR. Pick a quick-start preset or your own states and cities, set a date, optionally filter by industry, city, ZIP or radius, and turn on monitor mode to receive only businesses you have not seen before.

Quick start — three lead feeds, one click

Pick a preset in the 🎯 Quick start field and click Start; add Monitor mode and a daily schedule to turn it into a feed.

• Three layers of coverage: state registries (new LLCs and corporations), municipal business licenses in the largest metros (newly opened businesses with owner contacts), and SEC EDGAR Form D funding rounds (startups and growth companies that just raised money).
• Monitor mode, not just a scraper: remembers delivered records and returns only new ones next run — a clean daily lead feed.
• Only real new businesses: formation filings in New York, newly issued licenses in Chicago; biennial statements, renewals, amendments, and dissolutions are dropped. Future-dated and mistyped records are filtered out.
• One row per business: a company that shows up twice — two licenses, or a state filing plus a city license — is merged into one record with combined contact details and every source listed.
• Clean, CRM-ready fields: phones in E.164 (+14079058827), lowercase validated emails, 5-digit ZIPs, one-line addresses split into street / city / state.
• The company's own website, and the contacts on it: each company's domain is guessed from its registered name, then proved — the registry's ZIP or city has to be on the page, or the page's phone number has to be in the state the company registered in. A match that only looks right is reported as websiteGuess and never allowed into contactEmail / contactPhone; a page with a phone number from another state is discarded. An empty website is a signal in itself: those are the companies with no web presence yet.
• Contact checks included: numbering-plan validation with state and time zone for every phone, MX-based deliverability plus throwaway, role-address and free-mailbox flags for every e-mail — then filter the feed on what the checks found.
• Lead score 0–100 on every record: freshness, phone, email, contact name, address, and entity type — sort by it or set a minimum.
• Industry tags out of the box: exact NAICS sector where the registry publishes a code (Connecticut, Los Angeles, Seattle, Baton Rouge), the registry's own category elsewhere, name-based inference as a fallback — Real Estate, Construction & Trades, Healthcare, Food & Beverage, Technology, Legal, Finance, Transportation, Beauty, and more.
• Contact person: beneficial owner (District of Columbia), officer or member (Florida, West Virginia), registered agent, process agent, or owner; business email for Connecticut and Orlando, phone for New York City, Seattle, Orlando, and New Orleans.
• Geo targeting by city or ZIP prefix.
• Official, legal source and pure HTTP — no browser, no proxies, low cost.

Supported sources

State registries (new company formations)

City business licenses (newly opened businesses)

Funding signals (companies that just raised capital)

Every US company that sells private securities — a seed round, a Series A, a real-estate syndication, a private placement — must file Form D with the SEC within 15 days of the first sale (investment funds and SPVs are filtered out by default). Companies raising from the crowd file Form C before the campaign opens — with their website and financials. Larger public-style offerings file Form 1-A. These are the most expensive leads in the feed: companies raising money hire, buy software, spend on marketing and switch vendors right after the round.

More states and cities are added on request — open an issue in the Issues tab.

What data does it extract?

The company website, found and confirmed

E-mail is the field this market charges most for. State registries hand out names and addresses and almost never an e-mail, so the only honest way to add one is to read it off the company's own website.

That is what this step does, and the hard part is not finding a site — it is proving the site is theirs:

1. Guess the domain from the registered name, with the entity suffix stripped: Additive Printworks LLC → additiveprintworks.com, .net, .co, additive-printworks.com. Names that reduce to one generic word are never looked up — 135 Properties LLC is not properties.com, and AD | CREATIVE | LTD. is not adcreative.com.
2. Confirm the page belongs to that company. The registry's 5-digit ZIP on the page, or its city, or every word of the brand name. Parked and for-sale domains are rejected.
3. Only geography proves it. A name match is a lead, never a proof. website is filled when the registry's ZIP or city is printed on the page, or when the phone number on the page is in the state the company registered in. A name match with nothing to back it up is reported in websiteGuess instead — and a page whose phone number belongs to a different state is thrown away, because it is somebody else's company.
4. Read what the company published — the homepage first, then /contact, /contact-us, /about. Only the site's own brand or an ordinary consumer mailbox counts: a page at miamiyacht.com printing contact@miamivipyachts.com is a different business. HTML and CSS comments and @font-face blocks are stripped first, so a webfont licence header cannot pass the type designer's address off as the company's. Theme placeholders (first.last@company.com), film numbers (555) and digit runs out of a stylesheet are dropped, and a mailbox that appears on two different companies' sites is discarded as part of the page template. A phone counts only when written like one — a tel: link or real separators.
5. Check it, with the same MX and numbering-plan checks as every other contact in the feed.

Measured on 150 fresh Florida companies, 23 October 2026 — a state whose registry publishes no e-mail and no phone at all: 3 websites proved, 9 more reported as guesses, 3 candidates thrown out because the phone on the page came from another state, and 2 e-mail addresses and 3 phone numbers added where the feed had none. 160 HTTP requests, 26 seconds. Those are deliberately small numbers: the step is built to be right rather than to look productive, and every row carries websiteChecked so an empty website means "looked for, none exists" — which is the row a web or marketing agency actually wants to buy.

Switch it off with 🌐 Find the company website if you only want the registry's own fields, and cap it with Companies to look up per run. There is no extra charge for this step.

Phone and e-mail checks built into the feed

Registries publish contacts exactly as the filer typed them. In a real sample of 1,189 fresh records this Actor found 6 throwaway addresses (yopmail.com), 3 domains that do not exist, 1 domain that accepts no mail at all, and a phone number that breaks the North American numbering plan. Every one of those would have been a bounce or a wasted dial.

Leave Check phone numbers and e-mail domains on and each record arrives already checked.

Phone. Validated against the numbering plan — area codes and exchanges are checked against the rules and against the fourteen exchange codes that are never assigned to subscribers — then formatted for humans and tagged with the state and time zone of its area code, and with whether the number is toll-free, premium rate or non-geographic. The area-code table comes from NANPA's own public database, so new codes appear as they enter service. Twenty-two area codes straddle a time-zone boundary; for those the Actor lists both zones rather than picking one. phoneMatchesState tells you whether the number belongs to the state the business is registered in — a mismatch usually means an answering service or a virtual office.

E-mail. Syntax, then a live MX lookup: does the domain actually accept mail? Plus the mailbox provider (Google Workspace, Microsoft 365, Zoho, GoDaddy…), whether the address is a role account (info@, sales@), whether it is a free consumer mailbox rather than a company domain, whether the domain is a known throwaway service, and a suggested correction when it looks mistyped.

What these checks are not. They are deliverability signals about the domain, not proof that one specific mailbox exists. Mailbox-level probing needs an SMTP conversation on port 25, which cloud platforms block outright — any tool claiming per-mailbox certainty from a serverless run is guessing. Nothing here is sent to a third-party service: it is DNS plus open reference data, so the checks cost nothing and leak nothing.

Then filter the feed on what the checks found with Keep only leads with: everything, a phone or an e-mail, a valid phone or a deliverable e-mail, a deliverable e-mail, or a deliverable e-mail addressed to a person rather than to info@. The lead score follows the same evidence, so a record whose only contact turned out to be undeliverable no longer outranks one you can actually reach.

How much does it cost to get new business leads?

Pricing on scrape.ly: the event prices below are what this Actor charges. Platform usage (compute, proxy, browser) is billed on top at your plan rates.

The Actor uses scrape.ly's pay-per-event pricing: you pay per lead returned, plus a small add-on for each lead whose phone and e-mail were checked. Error rows are never charged.

Because every source is an open JSON API with no browser or proxies, platform compute is negligible — thousands of records cost cents. Bronze, Silver and Gold Store tiers get a discount on top.

How to set up a daily new-business lead feed

1. Pick states and/or cities (or ALL), turn on Monitor mode (onlyNew), and optionally set industries, city, or ZIP starts with.
2. Click Start once to seed the memory; the first run returns everything since registeredSince.
3. Add a Schedule (e.g. every morning) and a webhook or integration — HubSpot, Google Sheets, Slack, n8n, Make, Zapier.
4. Every scheduled run delivers only businesses registered since the last run — no duplicates.

⬇️ Input

{
    "states": ["NY", "CO"],
    "cities": ["NYC", "CHI", "ORL"],
    "onlyNew": true,
    "industries": ["Real Estate", "Construction", "Food"],
    "maxResultsPerState": 2000
}

Build a territory list for one vertical

Set cities to ["NYC"], industries to ["Food"] and zipPrefix to "112" to get every newly licensed food business in Brooklyn — with phone numbers.

Get only freshly funded companies

Set states and cities to [], signals to ["EDGAR_FORM_D", "EDGAR_FORM_C", "EDGAR_FORM_1A"] and onlyNew to true. Every scheduled run returns the companies that filed a funding notice since the last run — with the amount, the executive officer, the phone number or the website.

Everything within 5 miles of your office

Set cities to ["NYC"], near to "40.7580,-73.9855" and radiusMiles to 5. Each record gets a distanceMiles field. Add excludeNames with your current customers so they never come back as leads.

⬆️ Output

One business per dataset item.

{
    "type": "filing",
    "leadScore": 70,
    "sourceType": "city_license",
    "source": "ORL",
    "sourceName": "Orlando, FL",
    "companyName": "ATTUNED CHILD CENTER",
    "entityCategory": "UNKNOWN",
    "industry": "Child Care Services",
    "industryTags": ["Childcare"],
    "industrySource": "category",
    "formationDate": "2026-09-01",
    "address": "2849 S ORANGE AVE STE 350",
    "city": "Orlando",
    "region": "FL",
    "contactName": "Aimee Miller",
    "contactRole": "Owner",
    "contactPhone": "+14079058827",
    "contactEmail": "owner@example.com",
    "registryNumber": "BUS-1104045",
    "sourceCount": 2,
    "sources": [
        { "source": "ORL", "registryNumber": "BUS-1104045", "licenseType": "Business Tax Receipt" },
        { "source": "ORL", "registryNumber": "BUS-1104112", "licenseType": "Business Tax Receipt" }
    ]
}

Funding signal (SEC EDGAR Form D)

{
    "type": "signal",
    "signal": "FUNDING_ROUND",
    "leadScore": 85,
    "sourceName": "SEC EDGAR Form D",
    "companyName": "D5Cash Inc.",
    "entityCategory": "CORPORATION",
    "yearOfIncorporation": "2024",
    "jurisdictionOfInc": "DELAWARE",
    "industry": "Other Technology",
    "industryTags": ["Technology & Software"],
    "formationDate": "2026-08-31",
    "dateOfFirstSale": "2026-07-27",
    "fundingAmountSold": 4000000,
    "fundingOfferingAmount": 4000000,
    "investorCount": 5,
    "securityTypes": ["Debt", "Option / warrant"],
    "revenueRange": "$1 - $1,000,000",
    "city": "NEWARK",
    "region": "DE",
    "postalCode": "19713",
    "contactName": "Aryn Chadha",
    "contactRole": "Executive Officer",
    "contactPhone": "+14083868873",
    "relatedPersons": [{ "name": "Aryn Chadha", "roles": ["Executive Officer"], "title": null }],
    "cik": "2075147",
    "sourceUrl": "https://www.sec.gov/Archives/edgar/data/2075147/000207514726000001/xslFormDX01/primary_doc.xml"
}

Use cases for new business data

SDR and sales teams

New companies need banking, insurance, payroll, accounting, software, and marketing. Be the first vendor they hear from — and call the owner directly where a phone is published.

Local agencies and vendors

Build territory lists by city or ZIP, filtered to your vertical.

SaaS, recruiters and service providers selling to funded startups

Turn on the EDGAR funding signal: companies that closed a round in the last two weeks, with the executive officer's name and the company phone — before the news hits Crunchbase.

Investors and researchers

Track business formation trends by state, city, and industry in near real time, and see who is raising private capital where.

CRM pipelines

A daily, de-duplicated sync into HubSpot or Google Sheets via webhook.

Integrations and new business leads API

Use the monitor as a new-business-leads API through the scrape.ly API and the JavaScript or Python client, or connect it without code to HubSpot, Google Sheets, Slack, n8n, Make, Zapier, and webhooks. It also works with the scrape.ly MCP server for AI sales agents.

One HTTP call, rows straight back

run-sync-get-dataset-items runs the Actor and returns the rows in the same request — this is the endpoint n8n, Make and AI agents use.

Lead score: what the numbers actually mean

Every business gets a leadScore out of 100 — points for being freshly registered, for a valid phone, a deliverable e-mail, a contact name, a company entity type, an address, a confirmed website and an industry tag.

Measured on 42 live filings in October 2026:

The median score is 65 — a fresh filing with an e-mail but no phone. The cliff between 65 and 70 is not a quality line: above 65 you are effectively requiring a published phone number, and state registries publish one for roughly 1 filing in 14. That is why 70 looks like "slightly stricter" and behaves like "almost nothing".

Use 0 for every lead, 50–60 to drop records with no contact at all, and 70+ only when you specifically want phone-ready leads.

When your filters leave nothing

A run that finds filings and then removes every one of them with your own filters no longer returns an empty table. It returns:

• a free notice row saying which filter removed them and, for the lead score, the exact value that would work (the best score among these filings is 95 — set it to 90 and this run would have returned 1 row);

They exist so an empty result is readable: you see what was available and what to change.

Troubleshooting and error rows

Nothing is charged for an error row, and a run never fails because one source is down: the row lands in the dataset next to the good ones and names the source.

{
  "type": "error",
  "source": "sec-edgar",
  "url": "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&type=D",
  "error": "HTTP 403 — SEC rate-limited this address",
  "scrapedAt": "2026-09-11T03:12:44.180Z"
}

🤖 For AI Agents & LLM Apps

Compact reference for agents calling this Actor through the scrape.ly MCP server or the scrape.ly API (freshowl21/us-business-filings).

Purpose: returns businesses registered in the United States in the last few days — from ten state registries, nine city licence feeds and SEC EDGAR funding filings — as one de-duplicated row per business, with the phone and e-mail already checked. Use it to answer "who just opened a business here", "which companies in this industry registered this week" and "who just raised money".

Timing: most sources answer in seconds. Idaho and West Virginia are read from the states' live registries one filing at a time — 20 Idaho filings take about 30–60 seconds, longer with a city or ZIP filter — so call those through an asynchronous run rather than run-sync-get-dataset-items (5-minute cap).

Minimal input:

{ "preset": "local_leads_with_phones", "registeredSince": "2026-08-27", "maxResultsPerState": 200 }

Output: one dataset row per business — companyName, tradeName, formationDate, entityCategory, sourceName, sourceType, sources[], address, city, region, postalCode, latitude, longitude, contactName, contactRole, contactType, contactPhone, phoneStatus, phoneType, phoneState, phoneTimezone, phoneTimezoneOptions, phoneMatchesState, contactEmail, emailStatus, emailReason, emailIsRole, emailIsFree, emailIsDisposable, emailProvider, industryTags, naics, registryNumber, sourceUrl, leadScore (0–100) and summary (a ready sentence to quote back to a user). Funding records arrive in the same dataset with type: "signal" and add fundingAmountSold, fundingOfferingAmount, website, investorCount and cik.

Behaviors an agent should know:

• preset overrides states, cities, signals and requireContact. To control sources yourself, leave preset at custom.
• Always set maxResultsPerState (default 1000, per source). Runs stay well under five minutes at a few thousand rows.
• registeredSince is an ISO date and defaults to the last seven days under a preset. Registries publish with a lag of one to three days, so "today" is often empty.
• emailStatus is a signal about the domain, not proof that a mailbox exists: cloud platforms block outbound port 25, so no serverless tool can prove otherwise. Treat deliverable as "worth sending", risky as "check first", undeliverable as "do not send".
• To hand a user only reachable leads, set contactQuality to valid_contact or deliverable_email_person rather than filtering rows yourself.
• onlyNew: true remembers what was already returned in a named key-value store, so a scheduled agent never repeats a lead.
• Billing is pay per event: one event per returned lead, plus one per contact check when checks are on. No charge for rows filtered out before output.
• Every field is flat except sources[] and relatedPersons[]. Rows never contain HTML, so they can go straight into a prompt.

Newly registered companies, days after they file

State registries publish formation filings as open data, and this Actor reads them directly: new LLCs and corporations with the address, entity type, industry code and, where the state collects it, the founder's email. Business registration data with the renewals, amendments and dissolutions filtered out, so what is left is genuinely new.

SEC Form D: funded startups with a phone number

A company raising private capital files a notice within fifteen days of the first sale. That filing names the amount raised, the date, the investor count and the executive officers with a company phone. It is the highest-intent startup funding data that exists in public, and it is free.

Data sent outside scrape.ly

This Actor reads public sources directly and sends no input or results to any service outside scrape.ly.

❓ FAQ

Where does the data come from?

Official US government sources — state open-data portals (data.ny.gov, data.colorado.gov, data.pa.gov, data.oregon.gov, data.ct.gov, data.texas.gov, opendata.dc.gov), the Secretary-of-State registries of Florida (the state's public daily Sunbiz file), Idaho (sosbiz.idaho.gov) and West Virginia (apps.sos.wv.gov), city license feeds (data.cityofchicago.org, data.cityofnewyork.us, data.lacity.org, data.sfgov.org, cos-data.seattle.gov, data.cityoforlando.net, data.norfolk.gov, data.nola.gov, data.brla.gov) and SEC EDGAR. All are public records; the Actor reads them politely, follows each site's robots.txt and does not bypass any access controls.

Is it legal to use new business filing data?

It is public government open data intended for reuse. Review each portal's terms and consult a lawyer for your use case, especially for outreach compliance (e.g. TCPA for phone calls).

How fresh is the data?

Most sources update daily; New York City and Los Angeles weekly; Pennsylvania monthly. Monitor mode returns only what is new.

How does monitor mode avoid duplicates?

Delivered records and business keys are stored in a named key-value store and skipped next run, with a 2-day overlap window so late-published records are not missed. A business that gets a second license later is not delivered again.

Why is one company shown once when it has several filings?

The Actor merges records that describe the same business — same name (legal suffixes ignored), same state, same ZIP — into one row. Contact details are combined and every underlying record is listed in sources. Turn mergeDuplicates off to get every raw record.

What is the lead score?

A 0–100 number that says how actionable the record is: up to 30 points for freshness (registered this week), 25 for a phone, 15 for an email, 10 for a named contact, 10 for an LLC or corporation, 5 each for a street address and an industry tag. Set minLeadScore to 50 to keep only businesses you can actually reach.

Can I get emails or phone numbers?

Yes where the registry publishes them: email for almost every new Connecticut company and in Orlando; phone in New York City, Seattle, Orlando, and New Orleans. Other state registries publish an address and a contact person — in the District of Columbia the beneficial owner by name, in Florida and West Virginia every officer or member. Phones are normalized to E.164 and emails validated, so they drop straight into a dialer or CRM.

How accurate are industry tags?

Where the source publishes a NAICS code the tag is the exact sector (industrySource: "naics"); a registry category gives a keyword-based tag ("category"); otherwise the tag is inferred from the company name ("name"), and about half of such companies receive one. Untagged businesses are still returned unless you filter by industry.

Can I use it with the scrape.ly API or an MCP server?

Yes — see the API tab for snippets, and connect the scrape.ly MCP server for AI agents.

What is a funding signal and where does it come from?

SEC EDGAR filings: Form D — the notice every US issuer files within 15 days of selling private securities (Regulation D); Form C — the offering statement for equity crowdfunding campaigns (Regulation CF); Form 1-A — the offering statement for Regulation A+ raises. The Actor reads the daily EDGAR index and each filing's XML from sec.gov (official, free, no key, within the SEC's 10-requests-per-second rule). Investment funds, VC/PE vehicles and SPVs are dropped unless includeFunds is on; amendments are not returned as new rounds.

Is the contact the owner or a registered agent?

Check contactType: owner (city licenses, San Francisco, Orlando, Norfolk, New Orleans), officer (Pennsylvania, EDGAR signals), registered_agent (New York, Colorado — a law firm or agent service, not the founder), filer, or contact. addressType says whether the address is the business's own (principal), a mailing address, or the agent's. The lead score counts a named contact regardless of type, so filter on contactType when you need the owner.

Can I filter by distance from a point?

Yes — set near to "latitude,longitude" and radiusMiles. Coordinates come from the portals that geocode their records (Chicago, New York City, Los Angeles, San Francisco, Norfolk, New Orleans, Pennsylvania); records without coordinates are skipped when near is set.

Which states or cities will you add next?

Any with an open-data registry or license feed — tell us which one you need in the Issues tab.

Also known as

People look for this Actor as new business leads, newly registered companies, business registration data, a business licence feed, SEC Form D data, funded startups, new LLC filings, company formation data, and a B2B lead generation source.