What is LedgerBase?

What LedgerBase is

LedgerBase reads normalized US public-company financials derived from SEC EDGAR filings. Every available metric carries source lineage back to the specific filing and XBRL fact it came from, so a returned number is never a bare value — it is a value plus a way to verify it.

It serves a database snapshot populated by scheduled SEC ingestion, not a live feed. See Status for what is currently implemented and what is still pending before production use.

Who it's for

Developers building financial tooling, dashboards, or research apps, and AI agents that need machine-readable financial context with verifiable provenance rather than a number with no way to check where it came from.

What problems it solves

Raw SEC filings and XBRL facts are not developer-ready: the same concept can appear under different tags across companies and periods, duplicate facts exist across dimensions, and fiscal-year/quarter identity is not always obvious from the raw data. LedgerBase normalizes this into a consistent metric layer while keeping the original filing and XBRL evidence attached to each value.

See SEC Company Facts vs. normalized metrics for the full explanation.

Data source and normalization

Data originates from SEC EDGAR company filings and their XBRL facts (10-K, 10-Q, and related forms). “Normalization” means mapping the XBRL concept a company actually filed under to a consistent LedgerBase metric key, selecting the correct form and period, and resolving unit/period differences — all without inventing values the filing does not support.

Source filing and XBRL lineage

“Lineage” is how LedgerBase answers “where did this number come from?” — every supported metric can be traced back to the exact SEC form, accession number, filing date, and XBRL concept it was read from.

See Trace a financial metric to its SEC filing for a worked example, or the Data Quality & Lineage topic in the API reference for the full glossary.

Direct, derived, missing, and ambiguous data

Not every value is a direct XBRL read: some are derived from other reported facts, and some are legitimately absent because a company did not file the underlying concept. LedgerBase reports these states explicitly (AVAILABLE, MISSING, LINEAGE_MISSING) rather than silently coercing a missing value to zero or approximating it.

See Handle missing SEC financial data safely for why this matters and how to handle it in application code or an AI agent.

How this differs from raw SEC data access

SEC EDGAR and its Company Facts API are the authoritative source, not a defective one — but they serve a different layer of the data stack: raw, per-company XBRL submissions with no cross-company normalization. LedgerBase sits on top as a developer-facing metric layer, trading direct access to every raw fact for consistency and retained lineage across a smaller, normalized metric set.

How to evaluate LedgerBase

Explore the live API reference (including a Try It panel that calls the real API), browse the Metric Catalog, or read Learn for guides, tutorials, and core concepts. A free, keyless public demo is available on the homepage for a small set of tickers, rate-limited, no API key required.

What LedgerBase is not

LedgerBase is deliberately not:

• A cryptocurrency ledger, blockchain product, or hardware wallet.
• A generic double-entry accounting ledger.
• A live stock-price API or real-time market-data feed.
• A trading platform or execution venue.
• An investment advisory product — LedgerBase does not produce buy/sell/hold recommendations, technical indicators, or price predictions.

“Ledger” in the name refers to financial statement line items (a general ledger of reported facts), not a blockchain or bookkeeping ledger product.

Source and community

LedgerBase’s public repositories are on GitHub.