Consumer Risk & Recoveries

The reference universe refreshes continuously from source feeds, and every field carries a last-observed date, so you can tell a currently confirmed address from one that was accurate three years ago.

Phone numbers are handled differently again. Connectivity is checked live at the moment you query, not read from a stored status. A number that was connected when the record was built and has since been disconnected comes back as disconnected, which is the only answer worth having before a dialler campaign.

The information a recovery decision turns on:

• Current address, GNAF-verified, with address history in sequence
• Phone numbers with live connectivity status, and DNC status where relevant
• Email addresses with deliverability
• Property interests where held
• Deceased status, screened against more than 7 million records

Every field shows when it was last observed. That lets a reviewer weigh a recently confirmed address differently from a historical one, instead of treating an entire record as uniformly current because part of it is.

Every trace includes an Australian Death Check screen as standard, drawn from state and territory registry data, not inferred from returned mail or obituaries.

Matches are flagged before contact, so the account routes into your estates process instead of receiving a collection letter.

This is the highest-consequence check in recovery work. A demand letter addressed to someone who has died becomes a complaint, a regulator's attention, or a news story. Unlike most data errors, it cannot be walked back with an apology.

The checks give you current, right-party details. How you then make contact remains governed by your obligations under the ACCC and ASIC debt collection guidelines: contact frequency, permitted hours, conduct and hardship handling are yours to manage, and better data does not alter any of them.

What it does alter is the error rate. Most conduct breaches in recovery begin as wrong-party contact: pursuing someone who never held the account, or a person who happens to share a name with the debtor. Verifying the party before contact removes the most common cause instead of managing its consequences after the fact.

Property interests and other capacity signals also help separate the accounts that justify escalation from those that need a hardship conversation instead.

Both, and they suit different points in the cycle.

Caspar handles account-by-account work: the difficult individual files, where a reviewer needs address history, linked records and property interests together in one view.

The API re-verifies whole ledgers ahead of a campaign, returning per-record results so you can suppress, re-route or prioritise each account before the first contact attempt. Person History, Property Detail, address and phone endpoints all run in bulk.

Because the errors in a recovery file repeat every cycle, and each one costs money twice.

A letter to a superseded address is postage spent on a contact that cannot land, and it stays wrong next cycle unless something corrects it. A dialler working disconnected numbers burns agent time on calls that were never going to connect, then burns it again next quarter.

A single pass over the file before the campaign converts both into either a current contact point or a definite no-contact flag. The deceased screen is the clearest case of all: running it across the whole ledger costs less than handling one complaint that arises from missing it.

Yes, through WatchEye. Accounts stay under monitoring and you are alerted when something material changes: a new address, a deceased flag, a change in the underlying records.

The alternative is discovering those changes at the next campaign, which means a full cycle of contact attempts run against information that was already stale before the campaign started.