Methodology
How we collect, process, and present federal lobbying data from 726,000+ Senate LDA filings.
Data Source
All data comes from the Senate Lobbying Disclosure Act (LDA) API, which provides machine-readable access to all lobbying registrations and quarterly activity reports filed since 1999.
Our dataset covers 2018–2025, comprising 726,000+ filings with a total of $37.7 billion in reported lobbying income/expenses.
Data Collection
We pull all filings from the Senate LDA API in annual batches. Each filing includes:
- Client name, state, and description
- Lobbying firm name
- Individual lobbyist names and former government positions
- Issue categories (79 LDA-defined codes)
- Reported income or expenses
- Foreign entity affiliations
- Filing type (registration or quarterly report)
Processing Pipeline
Raw filings are processed into aggregated datasets:
Influence Score
Our composite Influence Score combines five dimensions:
- Spending: Total lobbying expenditure
- Lobbyist count: Number of individual lobbyists deployed
- Revolving door: Former government officials on the team
- Issue breadth: Number of different issue categories lobbied on
- Longevity: Years of continuous lobbying activity
Each dimension is normalized using a power curve (exponent 0.35) against the maximum in the dataset, then averaged into a 0–100 score.
ROI Calculation
The Lobbying ROI Calculator cross-references lobbying spending with government contract data from USASpending.gov. ROI is calculated as: (Total Contracts Received) / (Total Lobbying Spent). This is correlational, not causal — lobbying doesn't guarantee contracts.
Limitations
- LDA filings are self-reported by lobbyists. Accuracy depends on filer compliance.
- Income/expense amounts are often rounded or estimated.
- Some filings report $0 income (amendment or termination filings).
- Our entity matching (for ROI) uses name similarity and may have false positives/negatives.
- This is journalism, not legal accusation. Being a top spender doesn't imply wrongdoing.
Open Data
All processed data is freely available on our Downloads page. We encourage journalists, researchers, and citizens to use it for their own analysis.
Entity Resolution
One of the biggest challenges in lobbying data is that the same entity can appear under different names across filings. "Alphabet Inc.", "Google LLC", and "Google Inc." might all refer to the same parent organization. Similarly, lobbying firms may change names after mergers.
We use a combination of exact matching, fuzzy string matching (Levenshtein distance), and manual review to link related entities. Our client database consolidates 46,000+ unique client names into normalized profiles, though some edge cases may remain.
AI-Generated Content
AI-generated summaries and analysis appear on client pages, industry overviews, and investigation articles. These are clearly labeled and are generated from the underlying data — not from external sources. All factual claims in AI summaries are verifiable against the raw filing data available on each page.
Our AI analysis uses Claude (Anthropic) to generate narrative context for lobbying trends, spending patterns, and revolving door connections. These summaries help make complex data accessible but should not be treated as legal or investigative conclusions.
Update Frequency
Lobbying filings are due quarterly: January 20 (Q4 of prior year), April 20 (Q1), July 20 (Q2), and October 20 (Q3). Filers often submit late — some filings trickle in weeks or months after the deadline.
We process new filings in batches as they become available through the Senate API. Our database typically reflects filings within 2-4 weeks of their publication. Historical data is periodically reprocessed to incorporate amendments and corrections filed by lobbyists.
Editorial Standards
OpenLobby is a data journalism project, not a legal or investigative authority. We present publicly available data in accessible formats with contextual analysis. Our investigations highlight patterns and correlations but do not accuse any individual or organization of wrongdoing.
All lobbying activity shown on this site is legal and disclosed under federal law. High spending does not imply corruption. Our goal is transparency — giving citizens the tools to understand who is trying to influence their government.
Explore the Data
Data Freshness and Update Schedule
Lobbying filings are due quarterly: January 20 (Q4 of prior year), April 20 (Q1), July 20 (Q2), and October 20 (Q3). OpenLobby ingests new filings within hours of their appearance in the Senate's system. Late filings and amendments are captured in subsequent processing runs. Our database reflects the most current available data at all times.
Federal contract data from USASpending.gov is updated monthly. ROI calculations on the lobbying vs. contracts page use the latest available contract data paired with cumulative lobbying disclosures.
Data Sources: Senate LDA Filings (lda.senate.gov) · USASpending.gov (contract data) · Lobbying Disclosure Act Reports
Last updated: February 2026
This site is an independent journalism project. Analysis and editorial content are not affiliated with or endorsed by any government agency.
Data Notes & Methodology
All data on this page is sourced from Senate Office of Public Records lobbying disclosure filings under the Lobbying Disclosure Act of 1995. Figures reflect reported spending as filed and may be subject to amendment. Quarterly totals are annualized where noted.
Industry classifications follow the Center for Responsive Politics methodology. Where companies operate across multiple sectors, spending is attributed to the primary business classification. Foreign entity designations follow FARA and LDA Section 4 definitions.
Year-over-year comparisons use inflation-adjusted figures (2026 dollars) unless otherwise noted. Historical data extends back to 1998 when electronic filing became mandatory.
For questions about our data or methodology, see our full methodology page or contact us.