Methodology

How lotbook turns municipal open data and planning documents into a canonical, receipted record of development activity.

Where the data comes from

Toronto applications are ingested from the City's open data (the Development Applications dataset on CKAN) and enriched from the public Application Information Centre. Each source artifact is stored before it is parsed, so any figure can be traced back to the exact document it was read from.

How we track change

Every ingest run compares against the prior state and records what changed as an event — a new application, a status change, a new document. We never overwrite history; the event stream is the product.

Extracted facts and provenance

Figures like unit counts, storeys, and gross floor area are extracted from staff reports and planning rationales by a language model. Every extracted field carries its own provenance: the model used, a confidence score, and the page and quote it was read from. Low-confidence extractions are escalated to a stronger model before they are shown.

The free tier runs one week behind

The public map is deliberately delayed by seven days. Current status and newer events are available on Pro. This delay is how the free map stays free.

What this is not

lotbook is an independent product built from public records. It is not affiliated with or endorsed by the City of Toronto, and it is not a substitute for the official record. Always confirm against the municipality for legal or transactional decisions.