Overview

Pre-flight cost estimate for any /v1/* data query. Tells you, before you commit credits, how much a given request would cost based on cardinality and date range.

No API key required — this is the only data-pricing surface that's fully public, so you can show prices in your pricing-tier UI or do a sanity check before kicking off a large query.

⚠️

Estimates are not exact, especially with ticker_count. The number is a statistical projection from a sampled population — real cost for your specific request can land meaningfully above or below the quoted figure. Treat it as a budget guide, not a guarantee. Pass ticker (an actual list of symbols) instead of ticker_count whenever you can — the estimate is much tighter when it knows the exact tickers.

When to use

• Pricing-tier comparison pages.
• Pre-flight cost confirmation before a heavy batch query.
• Cost-modeling tools that loop over (endpoint, ticker_count, date_range) combinations.

Key parameters

• endpoint (required) — the target data endpoint. Accepts any of these equivalent forms: /v1/dilution-data, v1/dilution-data, or just dilution-data. Both v1/ and v2/ prefixes are stripped before lookup.
• Exactly one of ticker or ticker_count must be provided:
  • ticker — comma-separated symbol list (e.g. "AAPL,TSLA,MSFT"). Up to 100 tickers per call. Yields a tight estimate because the sampler measures real bytes per ticker for the names you actually care about.
  • ticker_count — integer in [1, 10000]. Estimates the average cost across a randomly-sampled population of this many tickers. Lower precision: if your real workload happens to skew toward heavy or light tickers, the estimate will be off in proportion.
• date_from / date_to — date range (YYYY-MM-DD), only relevant for endpoints that accept a date range. Wider ranges generally compound the uncertainty above.

Response highlights

• Cost in credits, projected row count, and estimated payload size.
• The values come from real recent data — not synthetic estimates — so they track actual query behavior closely, but they're still a projection, not a measurement. See Dilution Data, News, or any other data endpoint for the surface this is pricing.

Notes

• 422 cases: passing neither or both of ticker / ticker_count; passing more than 100 tickers; passing an unknown endpoint. Each returns a validation_error (or unknown_endpoint) with a specific message.
• Precision recap:
  • ticker=AAPL,TSLA,MSFT → tight; the estimate samples those exact tickers.
  • ticker_count=50 → directional; assumes a representative population. Variance grows as your real workload diverges from the population's average.
  • Add date_from / date_to (where applicable) to narrow the range the estimator projects over — that helps both modes.