Realtor License Roster, New Applicants & Broker Change Delta

Real-estate licensee lead feeds for brokerage recruiters, title companies, and proptech — built directly on official state licensing data. Six states, four feeds, one normalized schema:

Feed	What you get	Why it matters
new_applicants	TREC daily license applications — people who just applied for a Texas real-estate license	Future agents, pre-license. Reach them before they pick a brokerage. TX only.
new_licensees	Licences first issued in the last daysBack days (TX, FL, CT, CO, DE)	Brand-new agents in their day-one recruiting window, straight from the issue date — no baseline needed
delta → newly_licensed	Licences first issued since your previous delta run	Scheduled, de-duplicated new-agent alerts
delta → affiliation_change	An agent's sponsoring broker changed (with previous_broker)	THE recruiting/retention signal: who is switching brokerages, from whom, to whom
roster	Bounded slice of the current licensee roster	Bulk enrichment, territory mapping, CRM seeding

Coverage

State	Source (keyless, official)	Feeds	Freshness (measured 2026-09-25)	Notes
TX	TREC licence roster data.texas.gov/resource/s7ft-44qi (325,298 rows) + TREC applications bf5n-799f	new_applicants, new_licensees, roster, delta	Updated daily; newest licence issued 2026-09-24	Sponsoring broker on every sales agent; county often blank at source
FL	DBPR Real Estate weekly extract, regions 1-14 (RE_rgn{N}.csv, region 1 = 17,014 rows)	new_licensees, roster, delta	Weekly; newest issue date in region 1: 2026-09-25	Full mailing address; employer (sponsoring broker) name + licence on 11,126 of 17,014 region-1 rows; no application feed
CT	DCP eLicense "State Licenses and Credentials" data.ct.gov/resource/ngch-56tr, scoped to Real Estate Salesperson (97,478) + Broker (21,804)	new_licensees, roster, delta	Daily; newest issue date 2026-09-24	Street mailing address; broker affiliation joined from 6tja-6vdt (18,725 salesperson→broker links); no county column
CO	DORA Division of Real Estate data.colorado.gov/resource/4zse-6bnw, licence types containing "Real Estate"	new_licensees, roster, delta	Daily; newest first-issue date 2026-09-21	City/ZIP, link to DORA's licence page, disciplinary case flag; no sponsoring broker, no county
NY	DOS Active Real Estate Salespersons and Brokers data.ny.gov/resource/yg7h-zjbf (147,133 active)	roster, delta (affiliation_change)	Replaced in full daily (2026-09-25)	Business name + full business address; the brokerage a salesperson works under. Publishes no issue date, so NY never emits newly_licensed
DE	Division of Professional Regulation data.delaware.gov/resource/pjnv-eaih, profession "Real Estate" (31,007 rows)	new_licensees, roster, delta	Daily; newest issue date 2026-09-23	City/ZIP; no sponsoring broker, no county

Researched and rejected (evidence, 2026-09-25):

• Washington — data.wa.gov publishes only aggregate licence counts and transaction counts per county (ixni-jq78, a2n7-rij5), no licensee-level rows.
• Pennsylvania — data.pa.gov has no real-estate licensee dataset; its only professional-licence set (fwj2-whnj) is active-licence counts by county, last updated 2022-07-01.
• Oregon — no real-estate dataset on data.oregon.gov; the Real Estate Agency offers only an interactive lookup form.
• Illinois, Maryland, Virginia, New Jersey — no real-estate licensee dataset on their open-data portals.

Scope and cost — read this first

states is required. There is no default and no hidden nationwide mode, so a run always answers about the states you actually named.

{ "states": ["TX"] }

A call with no states is rejected before the run starts — HTTP 400, no compute, no rows, nothing billed. A rejected call is cheaper than a wrong answer.

You never get a failed run just because a state lacks a feed. Leave feed empty and each state serves its own default (TX new_applicants; FL/CT/CO/DE new_licensees; NY roster). Ask for a feed a state does not publish — e.g. new_applicants for Florida — and that state serves its closest honest feed instead, and every row says so in feed_requested, feed_resolution (as_requested / state_default / substituted) and feed_note.

What a run costs. Billing is pay-per-result at $10.00 per 1,000 rows (about $1.00 per 100 rows), and paid Apify plans get a graduated discount of 20-70% off that. Only rows actually written to the dataset are billed — a run that fails a pre-flight check emits nothing and bills nothing. maxResults is a hard ceiling; with several states it is split across them (budget one state does not use passes to the next), so the first state can never crowd out the rest.

How many rows to expect. The Texas new-applicant feed returns roughly 200 applicants for a 14-day window; Texas issues ~750 new licences in 14 days, Colorado ~130, Connecticut and Delaware a few dozen each, Florida region 1 ~40.

Example inputs

{ "states": ["TX"], "feed": "new_applicants", "daysBack": 14, "maxResults": 200 }

The daysBack window is anchored to the newest date actually present in the source, not to today — so the feed never returns empty just because the state paused publishing for a few days.

{ "states": ["FL", "CT", "CO", "DE"], "feed": "new_licensees", "daysBack": 7, "maxResults": 200 }

{ "states": ["TX", "NY", "CT"], "feed": "delta", "maxScan": 20000, "maxResults": 500 }

{ "states": ["FL"], "feed": "roster", "flRegions": ["1", "2"], "county": "Escambia", "maxResults": 200 }

Output fields

Every record uses the same normalized schema. A null means not published or not checked; it is never used to mean "no".

Field	Meaning
state	Licensing state (TX, FL, CT, CO, NY, DE)
feed	Feed that produced the row: new_applicants, new_licensees, roster, delta
licensee_name	Applicant, agent, broker or company name
license_type	Licence class as the state names it
license_number	Licence number (TX application ID for applicants; CO as PREFIX.NUMBER; CT as RES.0000000)
license_status	Current status (NY rows are all Active — that dataset is the active roster)
county	County on file (TX, FL, NY only)
city	City on file
mailing_address	Full mailing address (FL, CT)
sponsoring_broker_name	Sponsoring / employing broker (TX, FL, CT via join, NY brokerage for salespersons and associate brokers)
sponsoring_broker_license	Licence number of that broker (TX, FL, CT)
original_license_date	Date the licence was first issued (YYYY-MM-DD; NY does not publish it)
application_date	Application received date (TX new_applicants)
event_type	new_applicant, newly_licensed, affiliation_change, or roster
previous_broker	On affiliation_change: the broker the agent left
source / source_url	Government source name and dataset/file URL
feed_requested	Feed you asked for (null when you named none)
feed_resolution	as_requested, state_default, or substituted
feed_note	Why a default or substitute feed was served
license_class	Finer class where published (FL rank e.g. "SL Sales Associate"; CT entity type)
license_expiration_date	Licence expiration date
business_name	Business name / DBA (NY: the brokerage the licensee works for, or their own firm)
business_address	Business address (NY)
zip	ZIP / postal code
address_state	State of the address on file (many licensees live out of state)
license_detail_url	Link to the regulator's own licence page (CO)
discipline_on_record	true / false where the state publishes a usable discipline flag (CO case number); null elsewhere. DE is always null: its disciplinary_action column reads "N" on all 354,124 rows, including licensees in DE's own disciplinary-actions dataset, so it is not evidence of a clean record
sponsoring_broker_status	published_on_record, joined_from_affiliation_dataset (CT), affiliation_dataset_unavailable (not checked this run), or not_published_by_state — tells you whether a null broker means "none" or "unknown"
event_basis	Delta events: why the event fired
source_data_as_of	When the source dataset was last updated, as reported by the publisher (Socrata sources; null for FL)
run_legs_requested / run_legs_ok / run_legs_failed	Every state:feed leg this run planned / answered / did not answer
run_legs_failed_reason	Why each missing leg is missing — its rows are NOT in this dataset
run_complete	true only when every requested state answered

How the delta works

• The first delta run for a given scope seeds a baseline (licence number → status + sponsoring broker) in a named key-value store (realtor-license-roster-delta-baseline) and emits no events — that's honest delta semantics, not a bug. The baseline is scoped per state, filter and FL region set, so a filtered run never overwrites (or is compared against) an unfiltered one.
• Every later run compares the freshest maxScan records against the baseline and emits:
  • affiliation_change when the sponsoring broker differs;
  • newly_licensed only when the licence's original issue date is on/after the previous run (minus a 7-day slack). A licence that is merely new to the scan window — e.g. a 1988 licence renewed this week — is added to the baseline silently, never sold as new.
• Scanned records are not billed — only emitted events are.

Honest notes / limitations

• Contact fields: public licensing files contain no email or phone. You get name + licence + location (full mailing address in FL/CT, business address in NY). This is a licensing-signal feed you enrich, not a contact list.
• Filters a state cannot apply: CT, CO and DE publish no county column, and NY publishes no status column. A run that combines such a filter with those states skips them and says so in run_legs_failed_reason, rather than returning unfiltered rows. If no requested state can apply the filter, the run fails before any download and bills nothing.
• If one state's source is down, the others still answer and every row carries run_complete: false plus the reason. If every requested state fails, the run fails and bills nothing.
• Each new-licensee window is checked for freshness: if a source's newest issue date is more than 45 days old, that leg fails loudly instead of selling old licences as new.
• FL region files are ~4-5 MB each; request only the regions you need via flRegions (region 1 when empty).

Use as an MCP tool

This actor is callable by AI agents (Claude, Cursor, etc.) via mcp.apify.com — the normalized field-level output schema makes it easy to chain into enrichment or CRM workflows.

Was this useful?

If a run answered your question, a short review on this Store page helps other buyers find the Actor; every published review is read. If a run gave a wrong or confusing answer, open an Issue on this page with the run ID and it will be treated as a defect, not feedback.

FAQ

How do I find newly licensed real estate agents? Run feed: "new_licensees" for TX, FL, CT, CO or DE — it reads the licence issue date directly, so it works on the first run. For scheduled de-duplicated alerts, use feed: "delta". For an even earlier signal in Texas, feed: "new_applicants" lists people who only just applied.

How do I know when an agent switches brokerages? Scheduled delta runs emit affiliation_change events with the new sponsoring broker and the previous_broker they left (TX, FL, CT, NY).

Does this include agent emails or phone numbers? No — state licensing files don't publish them.

Which states are covered? Texas, Florida, Connecticut, Colorado, New York and Delaware — see the Coverage table.

I asked for Florida and got new_licensees rows — why? Florida publishes no application feed. Rather than failing your run, the actor served Florida's closest feed and marked every row feed_resolution: "substituted".

Why was my first delta run empty? By design: it seeds the baseline. The second run onward emits events.

Pricing

Pay per result: you are billed only for the licensee records actually emitted (cap with maxResults). Records scanned to build a delta baseline are not billed. Runs that match nothing cost only the actor start.

Related actors

For brokerage recruiters, title companies and proptech teams working the agents this actor finds. The Where both cover column lists only the places both actors actually serve (see each actor's own Coverage table).

Actor	What it adds for you	Example workflow	Where both cover
License Verifier	A single agent's licence status, expiry and any board discipline, checked against the board record	Before making a recruiting offer to an agent who just changed brokerages, confirm the licence is current and look for disciplinary action	NY (NY DOS real-estate salespersons and associate brokers), CO, CT, DE
Parcel Owner Lookup	Owner of record and owner mailing address for any street address, from the assessor roll	Give newly licensed agents a way to confirm who owns a prospect's home before a listing appointment	NY (statewide incl. NYC), CT, CO (41 counties), Houston (Harris County, TX)
NYC Landlord Registry Lead List	Landlord and managing-agent contacts with business addresses, banded by portfolio size	Hand NYC agents you recruit a list of small landlords to prospect for rental and listing work	NY (New York City only)
Distressed Property Leads	Motivated-seller leads scored on violations, tax liens and evictions, with owner mailing address	Seller-side listing leads for New York agents	NY (New York City only)
Absentee Owner Lead List Builder	Entity-owned (LLC / corporation) properties from the assessor roll	Listing-side leads for new NY agents. In NYC the source publishes owner names but no mailing address, so these are name lists	NY (New York City only)
RIA Registration Delta Monitor	The same new-registration / affiliation-change pattern for SEC-registered investment advisers	Recruiters who track agent moves between brokerages can track adviser moves between firms the same way	Nationwide (SEC data)