Investigation Tools

Access is restricted to organisations with a permitted purpose (debt recovery, fraud investigation, legal process and similar), and that purpose is confirmed during account setup, not asserted at query time.

Use is logged. Marketing is excluded outright: this data is licensed for investigation and verification, and it cannot be repurposed into a contact list.

The restriction is the reason the data is available at this depth. Address history, associates and property interests are exactly the fields that would be most damaging in a general-purpose marketing tool, and keeping the permitted-purpose boundary intact is what keeps them accessible for legitimate investigative work.

Less than most people expect. A search can start from a name, an address, a phone number or an email address, and partial details are enough to begin.

More inputs narrow the result faster. A name alone across a universe of more than 2 billion records will return many candidates; a name plus a last known address, a date of birth or an old phone number usually converges quickly on one.

That is the normal starting position for investigative work: the file has something, it is just out of date. The trace is designed to work forward from stale information, not to require current information you do not have.

Candidates are scored on how many independent records agree with each other: name, date of birth, address sequence and linked phone numbers.

A 91% match means those records converge strongly on a single person. A low score means the records are consistent with several people, or with one person about whom little is known.

The distinction matters for how a result should be used. High-confidence matches are answers. Low-confidence matches are presented as candidates, not conclusions, so an investigator can see what the evidence supports instead of being handed a single name with unearned authority behind it.

Everything needed to locate a person and understand their circumstances:

• Address history in sequence, so you can see the movement pattern instead of a single current address
• Phone numbers and email addresses, each with recency
• Relatives and known associates
• Property interests where held

Court records can be added to the same view where your purpose covers them. History runs to decades of address and contact data, which is what makes it possible to trace someone whose file went cold years ago.

In two quite different ways.

For locating someone, associates are often the route to a current address when the subject's own records have gone quiet. People who have deliberately dropped off the grid usually remain connected to people who have not.

For fraud investigation, the connections are the finding, not the means to it. Shared addresses, shared phone numbers and common associates link applications that were constructed to appear unrelated. A single fraudulent application is hard to distinguish from an unusual customer; a cluster of them sharing contact points is not.

The reference universe refreshes continuously from source feeds, and each field shows when it was last observed.

That last part does most of the work. An address with a recent observation date and an address last seen in 2014 are very different pieces of evidence, and a profile that presented both as simply "known addresses" would be actively misleading.

Phone connectivity is checked live at query time, not read from a stored status, so a number that has been disconnected since the record was built returns as disconnected.

Yes, and for service of process that is usually the point. A legal document has to reach the right person at a current address, and you may need to demonstrate afterwards that reasonable steps were taken to locate them.

Each result carries its source and the date the information was last observed, and the confidence score records how strongly the underlying records converged. That gives you a defensible account of how the address was arrived at, not an unsourced assertion that this is where the person lives.

Yes, at volume. Four endpoint groups cover the same ground as the portal:

• Caspar — person search and Autotrace
• Person — history and enhance record
• Phone — connectivity and contact lookup
• Address — validation and property detail

The usual split is to run bulk traces through the API to triage a caseload, then work the difficult individual files in Caspar where a person needs the full profile in one view.