See what the source says. Keep what it does not say visible.
A useful contact record carries more than a name and email. It preserves the project role, the source, the attribution basis and the checks that have or have not been done.
Discovery
A value can be found by a model, a deterministic extractor or a manual observation. The method is not the same thing as verification.
Attribution
A named person, firm, role and channel are linked only to the extent the evidence supports. A possible relationship stays a candidate.
Verification
Channel validity, ownership and organisation identifiers have separate verification states. An ABN is an identifier, not a contact channel.
Two public sources, one record
The state planning register publishes every development application, complying development certificate, construction certificate and occupation certificate in the state, so we can follow a project from its application to the certificates that come after it. Each council’s own document register holds the paperwork behind an application: the statement of environmental effects, the specialist reports, the plans and the notice of determination with its conditions. We read those documents through a multi-stage process. Site hazards come from the state’s own maps: bush fire prone land, flood, acid sulfate soils, mine subsidence, heritage and contamination notices, each attributed to its custodian. Nothing is drawn from a private register or a bought contact list.
What the record does not say
For many applications, the council’s decision is not separately recorded: the register shows the application was decided, but not whether it was granted or refused. We close part of that gap from the certificates. A construction or occupation certificate cannot lawfully issue without a development consent in force, so where a later certificate exists, the consent must have been in force, and we show that as known from the certificates, never as the council’s own recorded decision. Where there is no way to know, we say so, and we never guess which side of the ledger an application belongs on.
Tested, not promised
A multi-stage process reads each statement of environmental effects and draws out the trades a project will need, a plain-language project brief, the project value where the document states one, and the business contacts it names. How well that reading performs is tested against real applications read independently, by people who never see what our own reading produced. Only a way of reading that clears the bar reaches you, and one that slips below it later stops reaching you on its own. Where our reading of a kind of document is weaker, we label it as weaker rather than hide it.
Changes to a consent
A consent’s later life, its modifications and reviews, matters to anyone following a project. Many modification applications carry no reference linking them back to the original, so some links are matched by other means. Where a link is not certain, we mark it as probable rather than certain, and the changes that follow it are shown as an estimate rather than an exact count. The same mark travels with the record wherever it appears.
What counts as covered
We hold more than 220,000 applications from the councils whose document registers we read, and their council records list more than 470,000 documents. Not every listed document has been downloaded or read, and a field read from a document is shown only where we have read its text. An address with no usable street reference is shown as location unknown rather than placed on a nearby point, and it is left out of hazard flags and nearby searches rather than counted as covered. Where a council’s documents are not part of that set yet, the fields read from a document stay blank there. A blank is a limit on what we have read, not a claim that the answer is no.
Who you are shown
A contact is shown with where it was found, whether a stated contact block, a signature block, a title page or a certificate itself. Where nothing has been published, we say nobody is on file rather than guess one.