Finding a homeowner’s phone number is two problems that get discussed as one. The first is technical: given a property and a name on a public record, which telephone numbers belong to that person? The second is legal: having found one, are you permitted to dial it?
Most articles on this topic answer the first and skip the second. The second is the one that ends businesses, so it goes first here.
The boundary, before the method
This section describes the shape of the rules, not their application to your business. Telemarketing and consumer-data law is federal, state and fact-specific, and it changes. Take advice from your own counsel before building a calling operation. Nothing here is legal advice.
Four constraints exist independently of each other, and complying with one says nothing about the others:
- Do-not-call registration. National and state registries restrict marketing calls to registered numbers, with exemptions that are narrower than most people assume — an existing business relationship has a definition and a time limit, not a vibe.
- Consent for certain dialing technology. Rules attaching to automated dialing and to wireless numbers specifically are stricter than those for manually dialed landlines, which is why line type is a compliance field and not a nice-to-have.
- State law. Several states impose their own registration, timing, disclosure and consent requirements that go beyond federal rules, and they apply based on where the consumer is.
- Permitted purpose. Some categories of consumer data carry use restrictions — notably that they may not be used to make decisions about credit, employment, insurance or housing eligibility. Marketing is not screening, and the line is not one to approach casually.
The operational consequence: line type and scrub status are part of the record, not a step afterwards. A file of phone numbers with no wireless indicator and no scrub date is not a usable file regardless of how many numbers it contains. The registry side of this is covered in detail in DNC compliance for outbound solar.
How the match actually works
There is no directory of homeowners to phone numbers. What providers do is resolve an identity and then attach numbers to it, in roughly three steps.
Step one: anchor on the property. The public record gives an owner name and a situs address. Both are messy — names arrive as SMITH JOHN A & MARY, addresses as free text with inconsistent abbreviations. Standardising both is what makes the rest possible, and it is where a surprising share of failures originate.
Step two: resolve to a person. The provider searches its identity graph for a person matching that name at that address, across time. This is the hard part. A common surname at a common street address in a large metro can plausibly match several people, and the provider has to decide how much ambiguity to accept.
Step three: attach numbers. Once an identity is selected, associated numbers are returned, usually ranked, usually with a line type and sometimes a last-seen date.
Every error mode follows from step two. If the identity is wrong, the numbers are perfectly valid numbers belonging to the wrong household — and nothing about the data looks broken. That is why the wrong-party disposition is worth tracking separately from every other outcome on a floor.
What to give the matcher
Match quality is partly a function of what you supply. In rough order of value:
- A standardised property address. The anchor. Everything else is secondary.
- The full owner name as recorded, including middle initial and any second owner. Two names at one address is stronger evidence than one.
- The mailing address when it differs. For an owner who does not live at the property, the mailing address is where the person actually is, and matching on the situs address alone will find the tenant.
- The last sale date. It bounds when this person could have been associated with this address, which lets a matcher reject the previous owner.
That third point is the one most lists get wrong, and it is the entire reason absentee and out-of-state records need separate handling — see how to find absentee owners.
Match rate is the wrong number
Every provider quotes a match rate, and match rate measures whether something came back. It does not measure whether the number reaches the owner. The chain has four stages, and each one loses records:
| Stage | Question | Who controls it |
|---|---|---|
| Match | Did a number come back at all? | The provider’s matching tolerance |
| Live | Is the number still in service? | Data recency |
| Connect | Did someone answer? | Caller ID reputation and dialing practice |
| Right party | Was it the owner of that property? | Identity resolution quality |
A provider quoting only the first stage is quoting the one they control and the one that is easiest to inflate — returning four numbers per record raises match rate and lowers the quality of every dial. The full argument, including how to run a comparison that cannot be gamed, is in skip tracing accuracy. What it costs per usable contact rather than per record is in the skip tracing cost calculator. What a vendor means when it calls a mobile “verified”, and how to check that claim on a sample, is in verified mobile numbers for homeowners.
The single-lookup case
If the goal is one homeowner rather than a list, the honest answer is that the ordinary methods work and are cheap. Identify the owner from the county record — see how to find out who owns a house — and then search that name with the mailing address. For a single property, a consumer lookup service is usually adequate, and the compliance constraints above still apply the moment you dial.
The reason this becomes a serious data problem at all is volume. One lookup tolerates a 30% chance of being wrong because you will find out in ten seconds. Forty thousand lookups do not, because nobody finds out at all — the wrong ones simply become a slightly worse quarter. The two consumer routes people try first break in different ways at that scale: free lookup sites turn the cost into labour, and subscription people-search services stop at the export.
What good looks like
A phone append worth paying for returns fewer numbers than the alternatives, tells you the line type of each, tells you when it was last verified, tells you what evidence tied it to the owner, and has already been checked against the registries. Everything on that list reduces the headline match rate. That is the point.
The fourth item on that list — what evidence tied the number to the owner — is the one vendors are least likely to volunteer, and it is the difference between a phone attached to a building and a phone attached to a person. How to make a vendor state their definition before you buy, and why it outranks a two-cent difference in price, is in DataZapp alternatives.
Frequently asked questions
What does “skip tracing” mean for a homeowner list?
Matching a property owner to contact information — historically the practice of locating someone who had moved, now used loosely for any append of phones and emails to a record. For a homeowner list it means starting from the owner name and property address on a public record and finding phone numbers associated with that identity.
Is it legal to look up a homeowner’s phone number?
Identifying a number is a different question from calling it, and the second is the one with rules attached. Federal and state telemarketing law, do-not-call registries, wireless-specific consent requirements and state-level restrictions all govern the call rather than the lookup. Additionally, some data may not be used for credit, employment, insurance or tenant screening. Get advice from your own counsel on your specific use — nothing here is legal advice.
Why do two providers return different numbers for the same house?
Because they are solving an identity problem, not looking up an answer. Each builds its own view of which people are associated with which addresses over time, from different inputs and with different tolerance for ambiguity. A provider willing to accept weaker evidence returns more numbers and more wrong ones. Disagreement between vendors is expected and is not by itself evidence that either is bad.
Does a higher match rate mean better data?
Not on its own, and it is the easiest metric to inflate. A provider can raise match rate by returning any number loosely associated with the surname, or by returning several numbers per record. The number that matters is how often a dial reaches the actual owner of that property, which is a much smaller and much less flattering figure.