Every role that opened and every role that closed in Toronto across 1,210 company career boards, 2026-08-30 to 2026-09-04. A job count tells you who has the most openings in a place today; this tells you which direction that number is moving.
42.5% of the role changes in this window carry a location string this index recognises (4,577 of 10,782). Employers write that field by hand and the same city arrives in many spellings, so anything that did not match a known place exactly — a suburb, a region, a country on its own — is left out rather than guessed into the nearest city. Read the tables below as what the index could place, not as everything that happened.
Why these are hiring events and not crawler noise. Every opening is checked against the employer's own last-modified timestamp on the posting — the one number here that does not come from our fetch. 62.5% of openings carry a vendor timestamp within two days of the day we called them opened, median lag 1 day.
On its own that would prove nothing, so here is the control: requisitions already live on 2026-08-30, which are new by no definition, sit at 13.5% within two days against the same reference date, median lag 17 days. Openings measured the same way sit at 39.5%. A 2.9× separation.
35.9% of openings carry a vendor date more than a week old and are published as such rather than averaged away.
| Company | Net | Opened | Closed |
|---|---|---|---|
| MongoDB | +2 | 2 | 0 |
| 1Password | +1 | 1 | 0 |
| AvePoint | +1 | 1 | 0 |
| Avery Dennison | +1 | 1 | 0 |
| Block | +1 | 1 | 0 |
| Braze | +1 | 1 | 0 |
| BuildOps | +1 | 2 | 1 |
| Coursera | +1 | 1 | 0 |
| Kepler | +1 | 2 | 1 |
| Nubank | +1 | 1 | 0 |
| Phoenix | +1 | 1 | 0 |
| Reformation | +1 | 1 | 0 |
| Roku | +1 | 1 | 0 |
| StepStone | +1 | 1 | 0 |
The half nobody publishes: a ranking of open-role counts can only show you who is big in a place, never who is pulling out of it.
| Company | Net | Opened | Closed |
|---|---|---|---|
| Jobber | -6 | 0 | 6 |
| -4 | 0 | 4 | |
| Equinox | -3 | 0 | 3 |
| Okta | -3 | 0 | 3 |
| Lucid Motors | -2 | 0 | 2 |
| Toast | -2 | 0 | 2 |
| Astera Labs | -1 | 0 | 1 |
| Cohere | -1 | 2 | 3 |
| EXTREME NETWORKS | -1 | 0 | 1 |
| Harvey | -1 | 0 | 1 |
| Hut 8 | -1 | 0 | 1 |
| Lyft | -1 | 3 | 4 |
| Otter ai | -1 | 0 | 1 |
| Range | -1 | 1 | 2 |
| Samsara | -1 | 0 | 1 |
| Spotify | -1 | 0 | 1 |
| Stripe | -1 | 2 | 3 |
| Upwork | -1 | 0 | 1 |
| ZoomInfo Technologies LLC | -1 | 0 | 1 |
hiring-by-location.csv is every company in every place the index could place, with the window stamped on each row. All locations together.