Refresh companies
Refresh a CSV already saved in data/companies/. Subfolders are not listed.
A matching id and slug is updated. If neither exists, a new company is created.
What a refresh writes
- Header must include id, slug, name, size, website, industry, locality, linkedin_url, region.
- Update sets name, size, domain (taken from website), website, industry, locality, linkedin URL, region, and country.
- The first letter of each word in the name is capitalized. The rest of the word is left as it is.
- New rows use platform
genericand are inactive. Crawl fields are not changed on update. - If the id belongs to a different slug, or the slug belongs to a different id, that row is skipped.
- Country is inferred from locality and region. When it cannot be inferred, an existing country is kept.
- After the file is applied,
industry_verticalsis updated from the industry dictionary. A value likesoftware development / nicheuses the vertical for the part before the slash.&andandare treated as the same word. - Database writes require
DB_TRANSACTION_READ_ONLY=Falsein.env.
| File | Size | Modified (UTC) | |
|---|---|---|---|
finalized_companies.csv |
1.5 MB | 2026-09-28 19:40 |
Processing…
Refresh complete
- Total rows
- —
- Created
- —
- Updated
- —
- Skipped (invalid)
- —
- Conflicts
- —
- Domain kept
- —
- Country unresolved
- —
- Verticals inserted
- —
- Verticals updated
- —
Fields written
| Field | Written | Left blank |
|---|