Quid Pro Quo (QPQ) Tracker

Methodology

How we produce the figures on this site, and what they do and do not mean

What a tracked figure is

Every dollar amount on this site is labeled by what it measures. A figure marked tracked is the sum of the itemized filings that have been ingested into this database for that entity, deduplicated for amended filings. Coverage varies by entity and by cycle, so a tracked total is a floor, not a full-cycle figure. It will differ from totals published by the FEC or OpenSecrets. Where an official total is available, it appears alongside the tracked figure as a cross-reference.

Direct contributions, independent expenditures, and lobbying client spend are counted separately and are never added together into a single unlabeled number. Independent expenditures spent for a candidate and those spent against one are likewise separate figures. Money spent against a candidate is never counted as giving to or receipts of that candidate. Where a page shows one figure per row (the leaderboard, the sector cards on the home page), each column is labelled with the single kind of money it contains.

Tracked money is counted one hop at a time. When a firm gives to a super PAC and that super PAC spends, the sector total counts the firm's gift or the PAC's spending, not both. Party committees, candidate committees and joint-fundraising committees are never treated as members of an industry sector, whatever tags they inherited from their donors. Their outflows are transfers of other people's money.

Federal and state rows

The database holds FEC filings and itemized rows from a growing set of state campaign-finance systems (Texas, California, Illinois, Virginia, Georgia, Louisiana, Pennsylvania and others). A figure labelled federal uses FEC-sourced rows only. The home page's donor cards and its donor-concentration figure are federal. Leaderboard totals include state rows for the donors active in those states, and say so.

The donor-concentration figure (“share of tracked federal individual giving from the 25 largest donors”) combines duplicate person records of one donor by name before ranking (“GRIFFIN, KENNETH C.” and “Kenneth C. Griffin” are one donor). The records themselves are not merged until a human has confirmed the match.

Election outcomes

Win/loss tallies count the distinct 2024 general-election candidates a sector gave to or spent independent expenditures in support of. Candidates the sector only opposed are never included. Only results whose row names the source the result was taken from count. The tallies are not shown on the site at present. On 2026-10-23 no row of the election-results table recorded a source for its result (the won/lost values were load defaults). Thus every such tally is zero by this rule. When shown, the number of candidates with a sourced result is stated beside the tally.

Donor-position vote alignment

Legislator profiles show a figure labelled donor-position vote alignment (formerly “QPQ Score”). It is computed from three inputs. The first is the set of tracked direct contributions to that legislator, with the entities that made them. Independent expenditures, including those spent against the legislator, are excluded: money spent about a candidate is not money given to them. The second is the positions those donors took on specific bills, recorded as advocated_for or advocated_against relationships that cite a source. Positions the ingest inferred from the donations themselves are excluded, since scoring a vote against a position derived from the same donation would be circular. The third is the legislator's yea/nay roll-call votes on those bills. The figure is the share of such votes that went the way the legislator's donors advocated.

Two limits apply. The recorded positions are overwhelmingly “for” (about 99.7% of position rows on 2026-10-23). Thus the figure mostly measures how often a legislator voted yes on bills their donors supported. A legislator who votes yes on most bills will score high regardless of donors. And alignment is not quid pro quo. Donors tend to give to legislators who already agree with them, and the figure cannot distinguish that from influence. It describes the voting record beside the giving record. It does not show that the giving changed any vote.

Detailed notes

  • Chain attribution explains how money that passes through more than one PAC is apportioned from donor to candidate, and the difference between a modeled estimate and the smallest documented leg of a chain.
  • Policy events and spending explains how political spending is scoped to a specific rulemaking, vote, or enforcement action, and the limits of that method.
  • Data sources lists every source the ingestion pipelines draw on, with the terms and attribution each requires.

Sources and citations

Every relationship in the database carries either a link to the transactional record it was drawn from or an explicit editorial inference tag. A small residue of early seed rows has neither. Those rows are being re-sourced or retired. Wikipedia is not used as a source. Editorial descriptions cite filings, government records, or established financial and investigative press. Survey figures on the home page cite the Pew Research Center report they come from. Donor-population figures cite OpenSecrets. Where this site has done arithmetic on OpenSecrets' counts (the share of U.S. adults who gave $3,300 or more), the page says so.

Corrections

If a figure, relationship, or description on this site is wrong, write to kristofer@anticorruptionaction.org with the page address and, where possible, a link to the primary record. Corrections are made in the database rather than in the page text, so the fix appears everywhere the record is shown.