auscyclopedia
Read on the site.
Findings and research drawn from the public record.
Long-form reads from Australia's public data. Every claim cited, every figure dated, every method shown.
Making Australia's public data usable.
Australia's data is scattered and disconnected. We're connecting the dots to make it useful for all.
What connecting changes
Use the data to go deeper than before.
Which federal contractors won contracts in the same year they donated to the party in government?
Federal contractors, ranked by what they donated to the governing party the year they won. Some have a clean reason. The list tells you which ones to ask about.
One question, pulled across contracts and donations, returns the most informed answer.
Which ‘Australian-made’ trademarks are owned by companies registered overseas?
The label is what’s printed on the front of the packet. The ownership is what’s filed in the registers. They don’t always agree.
One question, pulled across the label and the ownership, returns the most informed answer.
Which charities pay their leadership more than they spend on the cause they fundraise for?
Where the dollar you donate actually lands. The split between cause, admin, and executive pay, charity by charity. Some run lean. Some don’t.
One question, pulled across the mailer and the accounts, returns the most informed answer.
Which businesses claim millions in R&D tax incentives but have never filed a single patent?
Companies claiming millions in R&D tax credit and filing no patents to show for it. Not illegal. Not innovation either.
One question, pulled across the tax claim and the patent register, returns the most informed answer.
Two ways in
How we make this data accessible.
Query the data.
Go deeper than the site shows.
Direct query access for AI agents and assistants. Same provenance, same sources, no scraping.
What's in the index
What that lets you do
- 01 Ask one question across all 23 sources at once. The same answer would mean 23 lookups against raw APIs.
- 02 Compare across places, time periods, and programs. No single dataset can show those cuts on its own.
- 03 Track facts as they change. Every figure carries its source and the date it was true.
- 04 Surface things hiding between datasets. Patterns that only appear when records are joined.
Updates
A weekly read on the country
Findings, patterns, things worth knowing. No marketing.