Earnings dates

Who reports when — demonstrably accurate for Europe

34,020 dates, checked against the actual filings: median one day apart.

curl "$API/calendar/upcoming?days=14&category=earnings"

{
  "events": [
    { "isin": "FI0009000681", "name": "Nokia",
      "date": "2026-09-17", "category": "earnings",
      "confidence": 0.94, "sources": ["yf_info", "nasdaq", "marketscreener"] }
  ]
}

What you can do with it

Retrieve the next reporting datesWithin a time window, filtered by results, dividends or annual meetings.
For a whole watchlist in one call/calendar/for-identifiers takes tickers and ISINs mixed.
Check how reliable a date is/earnings/confidence/{id} shows which sources name it and how well they agree.

Why we built it this way

Dates without a quality measure are worthless — an estimated date looks exactly like a confirmed one. So every date carries a confidence value derived from agreement across sources.

For European securities the internationally common sources barely deliver. Measured: our calendar covers 88 % of Finnish and 85 % of Swedish securities, the scraped alternatives under 2 %. That is why we run a European chain of our own.

Historical dates stay put. A calendar that only looks forward loses past every day at the window edge — and with it any chance of ever measuring your own forecast quality.

Improvements

Considerably more securities have dates, because more than one source is asked
Dates are stored even when only one source knows them — flagged as such
Historical dates are retained instead of falling out of the window
Per-source hit rates are visible

Endpoints

GET	`/calendar/upcoming`	Next dates in a window
POST	`/calendar/for-identifiers`	Dates for a watchlist
GET	`/earnings/confidence/{identifier}`	Source situation for a date
GET	`/earnings/coverage`	Coverage and hit rates

Give feedback or ask a question

Something missing, something off, or an answer you don't understand? Write it here — it reaches us directly.