Running pattern detector against the corpus
The dataset shows departments where a single supplier or small group receives a disproportionate share of contract spend — incumbency replacing competition.
Aggregated from the matches below. Severity bands are calibrated per pattern; counts and dollar totals reflect this snapshot only.
Glassbox surfaces correlations and observations, not findings of misconduct. Severity is a calibrated read on how strongly the dataset matches the pattern definition — a starting point for review, not a verdict.
The match meets the pattern threshold but the signal is the lower band. Normal review queue.
The match exceeds the routine threshold. Worth a closer look — pull the source rows, walk the citation.
The match is in the highest band of the calibrated scale. Recommend: prioritise for active review.
Severity is computed from the detection signal · HHI ≥ 1500 (Σ recipient share² × 100²) over departmental spend with ≥ $100M total. Bands: ≥5000 extreme, ≥2500 highly concentrated, ≥1500 moderately concentrated.
The dataset shows Veterans Affairs Canada | Anciens Combattants Canada disbursing $24.7B across 1,326 recipients. HHI 9812 (extreme); top three recipients (batch report│rapport en lots, Centre of Excellence on Post Traumatic Stress Disorders and Other Related Mental Health Conditions, Métis National Council) account for 99% of total spend.
Once a match is identified, the dataset suggests these review actions. Glassbox does not prescribe outcomes — these are calibrated options for funders, auditors, and program officers, framed as correlations to investigate.
Concentration can reflect a few large contracts (volume) or many small ones (incumbency). Each implies different responses — the former may be one-off necessity, the latter suggests competitive design issues.
Some functions are inherently concentrated (specialty equipment, niche expertise). Compare with peer departments delivering similar mandates to identify outliers.
If concentration persists across multiple FYs, the procurement strategy itself may merit redesign — open tenders, longer notice periods, or explicit competition mandates.
Annual re-scoring lets the funder track whether concentration is stable, increasing, or decreasing — and whether interventions are working.
These are calibrated suggestions, not directives. The dataset shows the pattern; the funder decides the response.