SOURCESTATE REFERENCE
Understand the data behind the comparison.
How do company filings become financial data that can be compared across issuers and periods? Reference explains the choices involved, the detail worth preserving, and how to trace a value back to its source.
Financial data
Compare reported facts while preserving definitions, source context, and time.
History and infrastructure
How public disclosure became machine-readable, and why access still differs across markets.
How Financial Filings Became Data
The unlikely rise of XBRL, the different paths taken by regulators around the world, and why public company information remains harder to use than it ought to be.
From Japan's EDINET rehearsal to EDGAR, DART and Europe's emerging access point: a selective history of digital financial disclosure.
Read the referenceDefinitions and lineage
What a reported value means, how it is standardized, and how to trace it to evidence.
As-Reported vs Standardized Financial Data
As-reported data preserves how a company disclosed a financial fact. Standardized data makes selected facts comparable. Reliable analysis needs both—and a traceable path between them.
A practical guide to labels and XBRL concepts, dimensional detail, subtotals, filing vintages, and source lineage.
Read the referenceHow to Trace a Financial Number Back to an SEC Filing
A filing URL identifies a document. A complete evidence chain shows which fact, context, occurrence, and filing vintage support a number.
A practical guide to as-reported labels, XBRL concepts, dimensional context, source positions, derived components, and version lineage.
Read the referenceTime and filing vintages
Which version of a historical fact was known, filed, or available at a given point in time.
Restated vs Recast Financials
Why historical numbers change—and which version should you use?
Restatements, recasts, amendments, and filing vintages through the lens of latest-known and point-in-time research.
Read the referencePoint-in-Time Financial Data
What version of a financial fact—and what level of detail—was actually available to an investor at a specific moment?
Period ends, SEC acceptance, official filing dates, dissemination, and later filing vintages are different clocks.
Read the referenceDisclosures, changes, and operating metrics
The notes, revisions, and company-defined measures that add meaning beyond statement lines.
The Financial Statements Are Only the Summary
Why the most important detail often lives in the footnotes.
SVB, The We Company, CoreWeave, and other filings show how notes reveal valuation gaps, commitments, concentration, debt structure, and timing.
Read the referenceWhat Changed in the Filing?
How to separate real signals from disclosure noise.
CrowdStrike, NVIDIA, and Boeing show why a document diff is not yet an economic delta—and where the evidence and caveats belong.
Read the referenceARR Is Not ARR: Why Company KPIs Are Harder to Compare Than They Look
ARR, retention, and customer-count labels can hide different methods, populations, usage windows, and assumptions.
Compare KPI definitions without erasing the attributes and limits that determine whether the numbers are actually comparable.
Read the referenceBacklog Is Not Revenue: What RPO, Backlog, and Deal Value Actually Tell You
RPO, backlog, and deal value can all describe future business without measuring the same contractual or operating state.
A practical framework for interpreting optionality, accounting scope, delivery dependencies, cancellation rights, and recognition timing.
Read the referenceXBRL
How filing facts acquire their financial meaning.
XBRL building blocks
Understand the context, dimensional structure, and custom concepts behind tagged facts.
XBRL Contexts Explained
How entity, period, and dimensions determine what an XBRL fact actually represents—and why the same concept can legitimately appear many times in one filing.
A guide to quarterly versus year-to-date facts, dimensional breakdowns, filing vintage, and choosing the right observation.
Read the referenceXBRL Dimensions Explained
How axes and members represent segments, products, geographies, and other views of a company's reported facts.
Broadcom, Hilton, Disney, and CoreWeave examples show how dimensional detail stays meaningful without double counting.
Read the referenceCompany-Specific XBRL Extensions Explained
Why a custom tag can carry economically important detail—and when forcing it into a standard field would erase that meaning.
A sourced guide to custom concepts, axes, and members, using a conservative Map / Relate / Preserve framework.
Read the reference