Help centre

How CSV export works

How to export CityRadar data to a spreadsheet: where the export buttons are, what the filters do to the file, every column in both exports, and the daily limit.

Where the export buttons are

Any list you can filter, you can export. The Export CSV button sits at the top-right of New businesses and of every category and community page; the Signals page has Export signals, which produces the signal-shaped file described below.

Exporting, step by step

  1. Filter the page first

    The export takes the filters you can see — category, community, look-back window, name search, and the numbered-company and chain-location toggles. Filter to the list you actually want, then export; there is no separate filter dialog to get wrong.

  2. Click Export CSV

    On your first export the button asks for an email address instead of downloading. That is the same one-time gate as the contact details, it is free while we are launching, and it is what the file is worth to us.

  3. Open the file

    The download starts as soon as the file is built. It is a UTF-8 CSV with a byte-order mark, which is what makes Excel read accented business names correctly instead of turning them into mojibake; Numbers, Sheets and LibreOffice open it as-is.

    Example filename

    cityradar-calgary-new-businesses-food-and-hospitality-90d-2026-08-10.csv

    The filters are in the filename on purpose, so five exports in a downloads folder are still tellable apart.

Screenshot — to be added

The Export CSV button on the new-businesses page, and the one-time email panel it opens.

/help/export-button.png

What is in the file

Two shapes. The businesses export is one row per business; the signals export is one row per event, so a business that opened and then pulled a permit appears twice, with the date and confidence of each event. Both carry the same seven contact and provenance columns, because a lead you cannot reach is not a lead.

Businesses export (17 columns)Signals export (16 columns)
  1. Business name
  2. Category
  3. Address
  4. Community
  5. Licence status
  6. First seen date
  7. Licence expiry date
  8. Website
  9. Social
  10. Phone
  11. Phone locality
  12. Email
  13. Contact source
  14. Contact fetched date
  15. Signal types
  16. Chain location
  17. Numbered company
  1. Signal type
  2. Business name
  3. Category
  4. Community
  5. Address
  6. Signal date
  7. Confidence
  8. Signal summary
  9. Relevant services
  10. Website
  11. Social
  12. Phone
  13. Phone locality
  14. Email
  15. Contact source
  16. Contact fetched date

Worth knowing about specific columns:

  • Contact source and Contact fetched date stay index-aligned: Google Places; business website pairs with 2026-07-03; 2026-08-08, in that order. They are the same per-value provenance the profile shows — how we source contacts.
  • Phone locality is spelled out (local, toll-free, out of province) rather than left blank for local numbers, so the column can be filtered on.
  • Dates are ISO (2026-08-10) so a spreadsheet sorts them correctly.
  • Chain location and Numbered company are yes/no columns. Rows hidden by the page’s default filters are excluded from the file too — tick the toggles first if you want them.
  • Relevant services (signals export) carries the fit score, e.g. Business insurance (82); Signage (60) — the number is why one row outranks another.
Column order is a contract: new columns are appended at the end, so a formula written against column N keeps working.

Limits

5 exports per 24 hours, counted across both file types together. Past that the button tells you rather than handing you a broken file, and the allowance frees up as the oldest export ages out. The limit is on taking the whole database, not on taking one flavour of it — if you have a genuine reason to need more, ask us.

A cell that begins with =, +, - or @ is prefixed with an apostrophe before it is written, so a business name that looks like a spreadsheet formula cannot execute when you open the file.