Datacircle

Free U.S. B2B leads dataset: 9.5M people, 1.75M companies

Datacircle gives away its 10M+ U.S. B2B dataset: 9,521,154 people and 1,748,139 companies, at U.S. businesses of 11 to 500 employees. Names, titles, LinkedIn profiles, work emails and mobile phones, joined to each company's website, industry, size and headquarters.

One download, free with a work email: a 2.3 GB zip of two Parquet files. No API calls, no pagination, no rate limits.

Download it

Sign up with your work email; the dataset is on your dashboard with a download button. Or from the API, with the key on that same dashboard: GET /files/ lists it as us_10m, and its download link is valid for one hour.

curl https://api.datacircle.dev/files/ -H "Authorization: Token $DATACIRCLE_API_KEY"
curl -X POST https://api.datacircle.dev/files/$FILE_ID/download-link/ \
  -H "Authorization: Token $DATACIRCLE_API_KEY"
import pandas as pd
companies = pd.read_parquet("us_smb_mid_market_companies.parquet")
people = pd.read_parquet("us_smb_mid_market_persons.parquet")
leads = people.merge(companies, left_on="CURRENT_JOB_COMPANY_LINKEDIN_ID", right_on="LINKEDIN_ID", suffixes=("", "_COMPANY"))

What the free plans of Apollo, Lusha, Hunter and others give a month, next to this file: free B2B database. A leads list cut by state, industry and title in one DuckDB query, with what it gives: build a free B2B leads list with SQL.

The people: 9,521,154 rows

Every column, with its fill rate in the October 2026 release. A field is filled only where we have it: nothing is guessed.

Field Type Nullable Fill rate Description
FIRST_NAME String No 100% First name (validated: no digits, 1–50 chars)
LAST_NAME String No 100% Last name (1–50 chars)
FULL_NAME String No 100% Full name
ABOUT String Yes 27.97% LinkedIn about/summary section
PROFILE_PIC_URL String No 100% URL to profile picture
LINKEDIN_URL String No 100% Full LinkedIn profile URL (https://www.linkedin.com/in/...)
LINKEDIN_ID String No 100% Unique LinkedIn numeric ID (all digits, 1–15 chars)
HEADLINE String No 100% LinkedIn headline
CONNECTIONS_COUNT UInt32 No 100% Number of LinkedIn connections (capped at 500)
MOBILE_PHONE String Yes 42.18% Mobile phone number (US format: starts with +1, 10+ digits)
CITY String No 100% City of residence
STATE_CODE String No 100% US state code (e.g. CA)
STATE_NAME String No 100% US state name (e.g. California)
CURRENT_JOB_TITLE String No 100% Current job title (1–200 chars, acronyms normalized: VP, SVP, EVP, RN, LPN, CNA, HVAC)
CURRENT_JOB_DESCRIPTION String Yes 10.35% Current job description
CURRENT_JOB_START_DATE Date32 Yes 58.91% When the current job started (validated: 1950–today)
CURRENT_JOB_CITY String Yes 23.14% City of current job
CURRENT_JOB_STATE String Yes 23.99% State name of current job
CURRENT_JOB_STATE_CODE String Yes 23.99% State code of current job
CURRENT_JOB_COMPANY_LINKEDIN_ID String No 100% LinkedIn ID of current employer (foreign key to company table)
EMAIL String Yes 29.31% Email address (validated format, lowercase, domain has dot)
EMAIL_VALIDATED_AT DateTime Yes 29.31% When the email was last verified
UPDATED_AT DateTime No 100% When the record was last refreshed

The companies: 1,748,139 rows

US companies with a claimed LinkedIn page and 11 to 500 employees (6,700,867 in the 50M+ dataset, every size). How they split by state, industry and size: the free list of US companies.

Field Type Nullable Fill rate Description
LINKEDIN_ID String No 100% Unique LinkedIn numeric ID for the company
LINKEDIN_URL String No 100% Full LinkedIn company page URL (https://www.linkedin.com/company/...)
NAME String No 100% Company name (acronyms normalized: LLC, PLLC, LLP, DBA, HVAC, USA)
HEADLINE String Yes 43.26% Company tagline/headline from LinkedIn
ABOUT String Yes 64.21% Company description/about section
FOUNDED UInt16 Yes 46.46% Year the company was founded (validated: 1600–current year)
LOGO_URL String No 100% URL to company logo image
LINKEDIN_FOLLOWERS UInt32 No 100% Number of LinkedIn followers
IS_CLAIMED_PAGE Bool No 100% Whether the LinkedIn page is claimed (always true - unclaimed pages are filtered out)
EMPLOYEE_COUNT_RANGE String No 100% Employee count bucket. 10M+ dataset values: 11-50, 51-200, 201-500
HQ_CITY String Yes 74.80% Headquarters city
HQ_STATE_CODE String Yes 73.14% Headquarters US state code (e.g. CA, NY)
HQ_STATE_NAME String No 72.94% Headquarters US state name (e.g. California, New York). Empty string when unknown.
HQ_POSTCODE String Yes 58.48% Headquarters postal/zip code
HQ_STREET String Yes 54.17% Headquarters street address
LINKEDIN_INDUSTRY String Yes 91.34% LinkedIn industry classification
TYPE String Yes 64.58% Entity type: Privately Held, Public Company, Partnership, Self-Owned
URL String Yes 79.47% Company website URL
UPDATED_AT DateTime No 100% When the record was last refreshed

US B2B contacts by state

The 50M+ dataset's 43,598,308 people by the state on their profile, with how many have a work email and a mobile phone in its October 2026 release: 27.09% and 42.88% overall. The free 10M+ dataset holds 9,521,154 people at companies of 11 to 500 employees; all 43,598,308 unlock once 3 people you invited sign up. Counted from the file itself.

US B2B contacts by state, with a work email and a mobile phone, October 2026
StatePeopleWith a work emailWith a mobile
California5,021,0531,337,0742,070,036
Texas4,038,3241,071,0811,671,818
Florida2,907,942734,4621,233,349
New York2,852,255837,7841,146,991
Illinois1,726,922514,369778,344
Pennsylvania1,691,305477,087753,227
Ohio1,605,350432,568725,825
North Carolina1,489,572411,503643,860
Georgia1,480,858372,902613,394
Michigan1,330,136380,085609,760
Virginia1,188,994285,684505,004
Washington1,084,802307,910463,686
Massachusetts1,081,975367,095475,716
New Jersey974,662257,889405,926
Arizona968,323250,399406,623
Colorado950,798287,446408,532
Tennessee882,158223,848372,131
Indiana845,767223,684380,913
Maryland823,160202,292364,514
Minnesota787,528250,981389,880
Missouri784,076210,496344,727
Wisconsin766,436238,727371,030
South Carolina637,186155,977267,942
Oregon550,236168,411246,672
Alabama532,507114,615222,057
Kansas505,493101,136194,023
Utah504,200152,095219,292
Louisiana474,672115,604207,286
Kentucky459,254106,984192,302
Oklahoma455,226108,220194,430
District of Columbia452,736153,914195,930
Connecticut450,886127,912206,125
Nevada409,52782,809165,640
Iowa352,818100,732162,300
Arkansas297,37773,085127,921
Nebraska266,67875,558123,014
Mississippi237,85845,90895,128
Idaho213,92758,45895,295
New Mexico196,65443,42780,351
New Hampshire165,69848,90974,219
Maine153,69847,15168,138
Rhode Island148,64141,77166,910
West Virginia138,68828,20354,270
Hawaii123,58426,45349,196
Delaware121,16231,85453,592
Montana103,29528,52544,208
South Dakota82,66423,67437,162
Alaska75,55716,59230,153
North Dakota75,48021,68932,305
Vermont69,84821,52130,586
Wyoming60,36214,41024,904

The companies the same way, by state, industry and size: the free list of US companies.

Ready-made lists

Questions

Is there a free B2B database I can download?

This one: a US B2B database of 9.5M people and 1.75M companies, free to download whole with a work email, as two Parquet files you can load into DuckDB, pandas, Postgres or a warehouse. No API calls, no export limits.

Is the US B2B leads dataset really free?

Yes, for everyone who signs up with a work email: the whole dataset, in one download, no API calls, no pagination, no rate limits. Freemail addresses are refused at sign-up.

What format is it in?

A 2.3 GB zip of two Parquet files, us_smb_mid_market_companies.parquet and us_smb_mid_market_persons.parquet. Join people to companies on CURRENT_JOB_COMPANY_LINKEDIN_ID = LINKEDIN_ID. pandas, DuckDB, Spark and most warehouses read Parquet directly.

Can I download a free list of US companies?

Yes: the companies table holds 1,748,139 US companies of 11 to 500 employees, with their LinkedIn page, website, industry, headquarters address, founding year and employee range.

How many have an email or a phone number?

29.31% of the people have an email (validated format, with the date it was last verified) and 42.18% a US mobile phone, in the October 2026 release. A field is filled only where we have it: nothing is guessed.

Is there a bigger dataset?

The 50M+ US dataset: 43.6M people and 6.7M companies, a 9.6 GB zip. It unlocks once 3 people you invited sign up.

Sign up with your work email: your API key and a $5 credit that never expires are on your dashboard, with the free 10M+ U.S. B2B leads dataset.

Sign up