Telemarketing lists for home services: who compiles them, and what to insist on

“Telemarketing list” is an older phrase than the mobile phone, and the product it names still ships every day: a CSV of names, addresses and numbers, cut by ZIP and a homeowner flag, priced by the thousand. For a floor that sells roofs, heat pumps or solar, the phrase hides the question that decides the week. Two files can both be sold as “homeowner telemarketing lists for your ZIPs,” and one of them was assembled from a national consumer file while the other was built from the county recorder’s deed index — and the difference shows up as who picks up the phone.

This page is written for the person buying the file. It covers who compiles consumer telemarketing lists and how the same rows reach you under different logos, what a compiled record is and what a property-record file is instead, and the three things a home-services floor should refuse to dial without.

Disclosure: Scout Data is our product, and it sells the second kind of file described below. Treat that section as a maker’s pitch; the rest is the read we would give a floor manager comparing quotes.

Who compiles consumer telemarketing lists

Almost nobody who sells you one. The consumer telemarketing market is two tiers deep, and the tier a floor buys from is usually not the tier that built the file.

At the top are the national compilers: a small number of companies that maintain a master consumer file of the US population, assembled by matching public records, licensed sources and self-reported data onto a name and address, and license slices of it wholesale. Below them are the brokers and resellers — the sites that actually rank for this phrase. They buy or license slices, add a portal and some selects, and sell rows under their own brand. TelephoneLists.biz says its records are “compiled from public records and licensed data sources.” DatatoLeads describes itself as a “Data Lead & List Broker” that “aggregates information from a variety of sources,” and also sells a white-label program so that other people can resell its catalogue under a third name. Both are quoted as publicly advertised on 18 September 2026.

The consequence for a buyer is simple and worth saying plainly: the row you are quoted has usually been sold before, under another logo, to another floor in your ZIPs, and “exclusive” on a compiled slice means only that this reseller is not selling it to a second customer this month. The vendor map — brokers, skip-trace platforms and the property-record vendors that sit outside both — is in call center data providers.

What a compiled record is

The key of a compiled consumer record is a person at an address. Everything else on the row was appended to that key from a different source, at a different time, with a different level of confidence:

• The selects are attributes, and many are modeled. Age band, income and net-worth ranges, presence of children, length of residence. A “homeowner” flag on a compiled file is often inferred from tenure and mail history rather than read from a deed, which is why a homeowner list can carry renters and owners who sold last year. TelephoneLists.biz offers homeowner-status targeting; its consumer page does not say how the status is determined.
• The phone was associated, not matched. It is a number the file has seen alongside this name or this address — from directory data, ported-number files, licensed telecom sources, forms. It may belong to the person on the row, a previous occupant, or an adult child. The match-depth question is worked through in homeowner phone lists for call centers.
• Refresh is file-level. The compiler updates the master file on its schedule; the reseller passes the update on when it reloads. TelephoneLists.biz publishes that its US data is updated quarterly, and that bulk files refresh “about twice a year.”
• Accuracy is stated as a range, honestly. TelephoneLists.biz publishes “80 to 85% accuracy on consumer records” and adds that “individual records are not guaranteed.” That is the correct description of a compiled file and a more candid one than most of the category prints.

None of this makes the product bad. It makes it a commodity, priced like one, and built for the campaign it was invented for: a short script that any adult in a ZIP might respond to, dialed at very high volume, where a third of connects reaching the wrong person is a cost the arithmetic can absorb.

What a property-record file is instead

The other kind of file starts from a parcel rather than a person. The county deed index says who owns the house and since when. The permit record says when the roof, the furnace or the panels were last touched. Aerial imagery says whether there are panels on the roof today. Homes are chosen on those facts — a roof past its local median, a solar permit with no array, a sale in the last ninety days — and then, for each chosen home, a phone is matched by name to the owner on title and checked live before it ships.

That last step is where the two products part company, and it is worth quantifying rather than asserting. On an 18,309-parcel pool of single-family homes in suburban Dallas–Fort Worth that we measured on 18 August 2026, 79.7% of homes yielded a contact whose name matched the deed, and 54.6% yielded that owner with a live mobile. The other homes were dropped rather than filled with a relative or a landline. A compiled file would have shipped a row for nearly all of them, and the floor would have found out which ones on the phones.

Three things to insist on for home-services dialing

A roofing, HVAC or solar pitch is about the house, and the decision about the house belongs to one person. That constraint sets three requirements that a generic telemarketing file does not carry by default, and each is a question you can put to a vendor in writing before you buy.

1. The owner on title, by name

Not “a homeowner.” The name on the row should be the name on the deed, and the vendor should be able to tell you when ownership was last checked. The difference is not academic: the share of connects where the person who answered is not the owner on the row is the number that turns a cheap file into an expensive one, because every such connect costs a rep minute and produces nothing. Ask, in writing: is the name on this row the name on the deed, and what date was that read? Then ask for owner-occupied as a separate column, so absentee owners and entities are a segment you chose rather than a surprise.

2. Wireless, flagged, with the flag’s source

Among adults in homes they own or are buying, 74.0% lived in wireless-only households in the second half of 2024, per the federal health survey that has tracked telephone coverage since 2003. The homeowner’s phone is a mobile, and a landline on a row is usually the number a vendor supplied because the mobile could not be found. The wireless flag also decides which dialer mode may touch the number and which consent rules attach to it, so a wrong flag is a compliance problem before it is a connect-rate problem. Ask which of the three methods produced the flag — number-block assignment, a ported-number file, or a live carrier lookup on that exact number — and verify a hundred rows yourself, as laid out in wireless vs landline homeowner lists.

3. A do-not-call status with a date on it

A DNC flag is a claim; a scrub date is evidence. The FTC’s Telemarketing Sales Rule ties its safe harbor to having accessed the registry within 31 days of the call, and the registry moves every day. TelephoneLists.biz puts the honest version on its own site — “the registry changes daily, so we don’t guarantee compliance” — and that is the right way to read any vendor’s flag: as a starting point for your scrub, not a substitute for it. Insist on the scrub date per row, re-scrub anything that sat in a queue, and keep your internal suppression file ahead of every load. The operating routine is in DNC scrubbing for roofing and HVAC; the legal frame, with the caveat that none of it is legal advice, is in DNC compliance for outbound solar.

Two more belong on the checklist, though they are about the dialer rather than the law: one number per row, ranked, so a dialer’s attempt counts mean something; and the fields that set the pitch — last roofing or mechanical permit, panels on the roof, sale date — so the opener has a fact in it. What a finished row looks like, column by column, is in anatomy of a dialer-ready record.

What the lists cost, and the unit that ranks them

The compiled tier is cheap and publishes its prices. As advertised on 18 September 2026, TelephoneLists.biz sells portal plans from $99 a month for 3,000 leads to $999 for 500,000, and bulk state files at $549; DatatoLeads advertises lists “from $0.001/Lead” and daily-lead packages from $299 a month. Property-record files are quoted per record against a floor’s spec and cost more per row, and the honest way to say it is that they are asking to be judged on a different number.

Pricing on this page is as publicly advertised at the time of writing — confirm current pricing with each vendor, and with our sales team for Scout Data.

That number is cost per right-party connect: what the file cost, divided by the connected calls where the named owner answered. A three-cent row where a third of connects reach someone else, and a thirty-cent row where most connects reach the owner, can land within a factor of two of each other on it, and the ordering is not predictable from the sticker. The full spread of advertised prices and the three divisions that convert any of them into that unit are in homeowner data cost per record.

Which file for which campaign

• A demographic offer at very high volume — anything an adult in the ZIP might want, with a short script and a floor that can absorb wrong-party connects: a compiled file, scrubbed and deduplicated before load. Two of the resellers that rank for this phrase are read in TelephoneLists.biz alternatives and DatatoLeads alternatives.
• A pitch about the house — a roof, a system, an array, where the opener needs a fact and the answer has to come from the owner: a property-record file, judged on right-party connects. What each vertical’s list should be built on is in roofing call center leads, HVAC call center leads and solar call center leads.
• Both, stamped separately. Plenty of floors run a compiled base for volume and a property-record file for the segments where the pitch needs a fact. That works as long as each file is tagged at import so it can be judged — and, when the numbers say so, dropped — on its own connect and wrong-party rates.

Whichever you buy, buy it the same way: a sample to your spec, dialed against the incumbent file by the same reps in the same hours, with wrong-party logged as its own outcome. How to run that so the result holds is in free lead samples: how to test a vendor.

Telephone-coverage figures are from the National Health Interview Survey’s wireless-substitution early release for July–December 2024. Registry figures are from the FTC’s National Do Not Call Registry Data Book for fiscal year 2025. Yield figures are from our benchmarks page, each with its sample and month.

Frequently asked questions