Disclosure: I built the tool used in this article (US New Business Registrations & LLC Leads on Apify).

Every new company is a customer for somebody. In its first weeks a business opens a bank account, buys insurance, hires an accountant, orders a website and a sign. That is why "new business leads" is an industry — and why most of those lists are expensive, stale, or both.

The underlying data is public. When someone forms an LLC or a corporation, the state records it, and several states publish their whole register as open data, updated daily. The trouble is that "several states" means several datasets, each with its own column names, its own type codes and its own quirks.

This article shows what six states publish, what you have to fix to use it, and how to get it as one clean JSON feed.

Six registers, six shapes

State Published by Size Updated
New York Department of State 4.3 million active entities daily
Texas Comptroller 3.5 million weekly
Pennsylvania Department of State 4.2 million rows every 1–2 weeks
Colorado Secretary of State 3.1 million daily
Oregon Secretary of State 1.6 million rows daily
Connecticut Business Registry 1.3 million daily

Each of them needs something different before the rows mean "one business":

  • Oregon repeats every business once for its principal office, once for its mailing address, once for its registered agent. You want the principal office row.
  • Pennsylvania has one row per person linked to the business. You want one row per business, and you probably do not want the people.
  • Texas encodes the type in two letters: CL is a Texas LLC, CI is an out-of-state LLC, CT a Texas corporation, CF an out-of-state corporation. (I had two of those the wrong way round until I checked them against the company names.)
  • Colorado uses its own codes too: DLLC, FLLC, DPC, DNC…
  • New York publishes the address where legal papers are served, which can be a lawyer's office, not the shop.
  • Texas lists some entities with a formation date weeks in the future.

And one distinction matters more than any other for leads: a domestic entity was formed in that state — a genuinely new business. A foreign entity is an existing company from somewhere else registering to operate there. Mix them and half your "new businesses" are Delaware corporations that have existed for years.

One call

curl -X POST "https://api.apify.com/v2/acts/surefeed~us-new-business-registrations/run-sync-get-dataset-items" \ -H "Authorization: Bearer YOUR_APIFY_TOKEN" \ -H "Content-Type: application/json" \ -d '{"states": ["NY", "CO"], "registeredInLastDays": 7, "entityTypes": ["llc"], "origin": "domestic", "maxResults": 20}' 

The answer is a JSON array in one format whatever the state:

{ "state": "NY", "registryId": "8129776", "name": "WIRT & COMPANY LLC", "entityType": "llc", "entityTypeRaw": "DOMESTIC LIMITED LIABILITY COMPANY", "origin": "domestic", "registeredOn": "2026-11-11", "status": "Active", "street": "510 Washington St Apt 701,", "city": "Buffalo", "addressState": "NY", "zip": "14203", "addressType": "service of process", "county": "Erie", "source": "New York Department of State — Active Corporations", "sourceUrl": "https://data.ny.gov/d/n9v6-gdp6" } 

The same in Python (pip install apify-client, version 3 or later) — new LLCs and corporations formed in Texas and Connecticut in the last 30 days:

from collections import Counter from apify_client import ApifyClient client = ApifyClient("YOUR_APIFY_TOKEN") run = client.actor("surefeed/us-new-business-registrations").call(run_input={ "states": ["TX", "CT"], "registeredInLastDays": 30, "entityTypes": ["llc", "corporation"], "origin": "domestic", "maxResults": 40, }) rows = list(client.dataset(run.default_dataset_id).iterate_items()) print(len(rows), Counter(r["state"] for r in rows)) for r in rows[:5]: print(r["registeredOn"], r["state"], r["name"], "|", r["city"]) 

Run on November 16, 2026, it returned 40 businesses, 26 in Texas and 14 in Connecticut, the newest registered that same day. States take turns when there is a result limit, so a small run is not filled by the first state alone.

What you get, and what you do not

  • You get the register entry: name, type, formation date, status, an address, and — where the state publishes them — the county (New York, Pennsylvania), the industry as a NAICS code (Connecticut) and a link to the register page (Oregon).
  • You do not get phone numbers or email addresses. Registers do not hold them, and lists that include them took them from somewhere else.
  • Names of people are left out on purpose — registered agents, organizers, officers. The feed is about businesses.
  • Texas and Pennsylvania publish late. They suit a weekly list better than "yesterday's filings".
  • Six states, not fifty. California and Florida are not covered.

Daily leads without duplicates

{ "states": ["NY", "CO", "CT"], "registeredInLastDays": 10, "origin": "domestic", "onlyNewBusinesses": true, "maxResults": 0 } 

onlyNewBusinesses remembers what earlier runs returned and gives you only what was added since. Use a rolling window of ten days rather than one: states publish a registration a few days after it happens, and the overlap catches the late ones without returning anything twice. There is a ready n8n workflow that sends the result as a morning email.

Cost

Pay per result, no subscription: $4 per 1,000 businesses. Businesses left out by your filters, and ones already returned to the same list, are not charged. Apify's free plan includes monthly credit, enough to try it.

The data comes from the open-data portals of the six states, which are published to be reused. The tool is not affiliated with any state government.

Link: https://apify.com/surefeed/us-new-business-registrations