Our approach

How we turn filings into ownership maps

Every number on this platform traces back to a regulatory filing. We do not guess. We parse, clean, normalize, and publish. Here is exactly how that happens.
Institutional ownership data is messy. Filings arrive in different formats, use inconsistent identifiers, and report positions with varying lags. We built a five-step pipeline to transform that raw noise into structured, comparable, and verifiable ownership concentration metrics.
  • Parse every filing at the position level, not the summary level.
  • Normalize identifiers across filers, securities, and time periods.
  • Validate each data point against adjacent filings before publication.
See our data Meet the team
  • Live +12% Tracked positions $2.4T
    Filers 4,800
    Securities 8,200
  • Ownership data processing pipeline diagram
  • Quarterly $312M
    Average position change per filing period
  • Analyst performing data validation checks

From filing to insight

Our pipeline moves from raw regulatory disclosure to publishable ownership concentration score in five structured stages. Each stage is documented, reversible, and open to audit.

Ingest raw filings

We collect regulatory disclosures daily from primary sources. Each filing is stored in its original format with timestamp and source URL preserved for full traceability.

Covers 13F, N-PORT, and equivalent institutional reports.

1
2

Parse position data

We extract individual security positions from each filing, preserving filer identity, position size, security identifier, and reporting date at the most granular level available.

Handles XML, HTML, and plain text filing formats.

Normalize identifiers

Filers and securities are mapped to consistent internal identifiers. This resolves ticker changes, CUSIP updates, and filer name variations that would otherwise fragment the dataset.

Cross-references three external identifier databases.

3
4

Calculate concentration scores

We compute ownership concentration metrics relative to sector peers and historical baselines. Each score reflects the proportion of institutional ownership held by top filers.
Updated weekly with full cross-sectional coverage.

Validate and publish

Every metric passes automated range checks and manual spot reviews before publication. Anomalies are flagged, investigated, and either corrected or documented as known limitations.
Corrections published monthly in a public changelog.
5

What makes our pipeline reliable

Technical choices that keep the data honest and the analysis reproducible.

  1. Granular parsing

    We parse filings at the individual position level rather than relying on summary tables. This preserves filer-specific detail that aggregate reports obscure.

  2. Identifier normalization

    Our cross-reference engine resolves ticker changes, CUSIP updates, and filer name inconsistencies across the full historical dataset.

  3. Automated validation

    Every calculated metric passes range checks, historical consistency tests, and cross-filer verification before it reaches the publication layer.

  4. Versioned methodology

    All scoring algorithms, weighting schemes, and filter rules are versioned and published. Users can replicate any historical result using the documentation provided.

  5. Open documentation

    Our full analytical framework is published for external review. Researchers can critique, extend, or replicate our ownership concentration scores without proprietary barriers.

The research behind the signals

Academic literature on institutional ownership has long recognized that concentrated positions by large asset managers influence liquidity, volatility, and price discovery. We build on this foundation by applying cross-sectional methods to a continuously updated regulatory dataset.

Our concentration scores draw from established measures of ownership dispersion, adapted for real-time filing data. We weight positions by filer size and report relative to sector peers to ensure comparability across the equity universe.

The team reviews methodological choices quarterly against emerging research and user feedback. Major revisions are documented with backward compatibility notes so historical analyses remain reproducible.
We do not claim predictive power. Our contribution is measurement: making ownership concentration visible, consistent, and traceable to its regulatory source. What users do with that visibility is their own analytical work.
Academic research on institutional ownership dynamics

Experience the methodology firsthand

See how ownership concentration data looks in practice. Browse sample reports, explore the interactive ownership maps, or request a walkthrough of our analytical pipeline with a member of the research team.

Sample ownership reports

Download example concentration analyses with full methodology footnotes. See how we trace every metric back to its original regulatory filing before drawing conclusions.

Interactive ownership maps

Explore the live dataset through our visualization tools. Filter by sector, time period, and filer type to observe accumulation and dispersion patterns as they develop.

Research team walkthrough

Schedule a structured conversation with our analysts about methodological choices, data limitations, and how ownership dynamics research fits your analytical workflow.

Request demo

All demonstrations are informational and do not constitute financial advice or investment recommendations.

Cookie types

Essential cookies

Analytics cookies

Functional cookies

Read cookie policy

Cookie preferences

We use cookies to support website functionality, analyze how visitors engage with our research content, and remember your preferences. You can accept all cookies, reject non-essential ones, or read our full cookie policy for details.