Receipt — The delivery scoreboard: what the federal government reports about its own targets

A Receipt is a dated evidence summary for one product. It shows the question, the current answer, the sources, the method, the limits, and the next update.

Open the product this Receipt documents → · Portable JSON · Correction policy

The question	How many of the federal government's own performance targets does it report meeting — and how many does it report on at all?
Current reading	Federal organizations publish thousands of performance indicators about themselves, each with a target and a verdict on whether that target was met. The verdicts are their own, and they are the raw material of almost every public claim a department makes about delivering on its commitments. A reader who has only a headline share of targets met cannot tell whether it moved because more targets were met or because fewer indicators arrived without a verdict, cannot tell whether the indicators being compared are even the same ones, and cannot tell what a met target establishes. The register that answers those questions is published by the government itself, in one file, and is simply less often counted.
Evidence reviewed through	2026-07-10
Basis	The Treasury Board of Canada Secretariat's own published performance file for Departmental Plans and Departmental Results Reports, held here as one preserved edition under a content hash: 19,010 rows, of which 13,723 are Departmental Results Report rows covering fiscal 2020-21 through 2024-25 and 5,287 are Departmental Plan rows covering fiscal 2025-26 and 2026-27. Every figure on the product is a count over those rows, or a share derived from two such counts. Only Results Report rows are scored; Plan rows are counted and never scored. Nothing is copied from the issuer's own commentary, and no departmental narrative report is read into this product.
Method	Pinned, cited evidence; no scenario is computed — every figure restates a pinned, re-derivable value from the product's member Resolutions.
Confidence	The arithmetic is exact over a content-hashed file, and every count and share on the product reproduces from those bytes on one stated convention; the derivation refuses to run at all if the archived bytes do not hash to the pinned value, if the indicator key stops being unique, or if a plan row ever arrives carrying a verdict. What the product deliberately does not settle is whether any of these verdicts is correct. Each is the reporting organization's assessment of its own target, and the field that would let an outside reader check one against an independent series is documented by the issuer and not published.
Next trigger	The Treasury Board's next republication of the performance file. No publication schedule for it is recorded here; the republication that first carries fiscal 2025-26 results will also be the one that removes the fiscal 2025-26 plan rows this record holds.
Next evidence	The next preserved edition of the performance file: it adds a reported fiscal year to the scoreboard and, by the same act, deletes plan rows no public archive other than this one would hold.
Expiry	The reading holds for the March 2026 edition of the file; the next preserved edition either re-affirms it or changes it on the record, with this product's correction history intact.

Material definitions

A verdict is the reporting organization's own published status against its own target, taken exactly as published. It is not recomputed from the target and the result, not overridden, and not verified — by the Treasury Board or by Countability.
A decided verdict is one reported as met or as not met. An indicator reported as to be achieved or as not available has no decided verdict, and is counted in the second denominator only.
Two shares are published for every fiscal year: the share of decided verdicts reported met, and the share of all indicators reported that year recorded met. Quoting one without the other reports a movement in the basis as a movement in delivery.
A share is two integer counts divided, expressed as a percent and rounded half away from zero to one decimal — on the percent, never on a count.
An indicator is identified by the issuer's own indicator identifier within a fiscal year. That pairing is verified unique across all 13,723 Results Report rows on every build rather than assumed.
The four published verdicts are every value the status column takes on a Results Report row, so counting all four is a decomposition of the whole rather than a selection from it.
Organizations are compared on one variable, one fiscal year, one edition of the file, one basis and one evaluation method, computed identically for each and listed alphabetically. No score is composed, no measures are combined or weighted, and no organization is graded.

Decisive limitations

Every status on this product is self-reported by the organization it describes, against a target that organization set, on a methodology it states in its own words. Consistency across organizations is not established and cannot be established from this file.
A met target is not a measured outcome. Nothing here states, or can support, that a target worth meeting was met, that meeting it changed anything, or that an organization reporting a higher share performed better.
The organization table is a matched comparison of one measure and is not a ranking. It carries no rank column, no composite score and no better-organization verdict, and it supports none.
The issuer documents a field naming each indicator's data source and does not publish it, so no indicator in this file can be checked against an independent official series. This product counts verdicts; it verifies none of them.
An indicator identifier that stops appearing is an absence in the file. This product asserts nothing about why one stopped appearing, and nothing about why any status moved: the file records what was reported, never why.
The same identifier is not guaranteed to be the same indicator across years. Of the 1,822 identifiers present in all five reported fiscal years, 596 carry a published name that is not identical across them and 882 carry a recorded target that is not.
Departmental Plan rows and Departmental Results Report rows share no fiscal year, so no plan can be compared with the report that follows it. The plan rows for a fiscal year are deleted from this file when that year's results are published, which is why the file is preserved on capture.
One preserved edition, not a live feed: the issuer's March 2026 edition of the file, captured 2026-07-10, whose latest reported fiscal year is 2024-25.
Contains information licensed under the Open Government Licence – Canada. Source: Treasury Board of Canada Secretariat, GC InfoBase — Departmental Plans and Departmental Results Reports: performance information by program and by organization, 2026-03-19.

Key sources and licences

GC InfoBase — Departmental Plans and Departmental Results Reports: performance information by program and by organization — Treasury Board of Canada Secretariat Licence text read in full

Every figure this Receipt covers can be reproduced from the named public sources under Countability's published method. If a correction is ever needed, it produces a new dated version rather than editing this one.