Data sources

How our data is collected, processed, and refreshed

Where the data comes from

Every consent record on Consented is sourced from the Christchurch City Council (CCC) Resource Management Act (RMA) consent register, which CCC publishes through their public open-data feeds. The register contains two record types:

  • Received, every consent application lodged with the council.
  • Issued, every consent decision granted or declined.

We ingest both in full. We do not cherry-pick records and we do not paraphrase the underlying data. The source-of-truth fields (application number, address, applicant, decision) come straight from the council feed.

How often it updates

  • Source cadence: CCC's register is publicly updated as new consents are filed and decisions made.
  • Our cadence: we pull the latest data daily at 7am NZT, run our processing pipeline, and refresh the dashboard and digest queue the same morning.
  • Digest emails: you choose the cadence in Settings. Daily, weekly, or monthly, delivered at 7am NZT.

There can be a short lag between a consent being lodged with the council and appearing in the public register. We do not claim real-time delivery, but if it is in CCC's public feed by 7am, it is in Consented by 8am.

Coverage

  • Geography: the Christchurch City Council district. Selwyn and Waimakariri district councils are not currently included.
  • History: January 2024 through today, refreshed daily. Over two years of continuous coverage.
  • Record types: RMA resource consents (subdivision, land use, etc). Both lodged-but-undecided and decided records are included.

How we process the data

Raw council data is useful as evidence but hard to use as intelligence. Between ingestion and what you see on the dashboard, we add three things:

  • Cleaning and normalisation: consistent column names, standardised dates, deduplication where the council file contains repeated entries, and validation of every row before it lands in our database.
  • Suburb and developer tagging: addresses are matched to Christchurch suburbs, and applicant strings are grouped so that the same developer trading under several legal entities shows up as a single firm.
  • Plain-English summaries: each consent gets a short human-readable description generated by an AI model from the council's own application text. The original council text is always preserved and viewable underneath the summary.

What this data is not

Consented is a powerful read on one specific dataset. It is not a complete picture of all construction activity in the city. In particular, the data does not include:

  • Building consents. Building consents are a separate council process and a separate dataset. Consented covers RMA resource consents only.
  • Private or off-market deals. If a project does not require a resource consent, it will not appear here.
  • Other councils. Selwyn, Waimakariri, and the wider Canterbury region are out of scope until we expand.
  • Instant lodgement notifications. We refresh from CCC's register daily, so there can be a short lag between a consent being lodged and it appearing in Consented.

Accuracy and limitations

We work hard on data quality, but the council record is what it is. If a row is missing a field at source, we cannot invent it. AI summaries are generated automatically and, while we monitor for quality, they should be treated as a guide, not as a substitute for reading the original council record. For the full legal position on accuracy, warranties, and AI-generated content, see our disclaimer.

Source attribution

Data sourced from Christchurch City Council RMA consent records and their public Power BI feed. Consented is not affiliated with, endorsed by, or supported by Christchurch City Council.