Irish Public Procurement Evidence Programme

The programme

In July 2026, Amata Consulting set out to answer a question no one could answer from published data: can the Irish State's procurement spend actually be traced to the suppliers who receive it?

We built the instruments first. A roster of 318 public bodies, 230 of them under weekly automated monitoring. A purchase-order dataset of 10,215 records totalling €1.02 billion across 8 bodies, every row carrying its provenance. Entity resolution across bodies, a retention ledger watching for records that quietly disappear, and preregistered predictions filed before the observations that would test them.

While the instruments ran, we published. A sequence of long-form pieces established the evidence in public — each one audited after publication against its ratified text, with a standing corrections log and right of reply.

Then the timing play: seventeen days before the government published Procuring for our Future: National Public Procurement Strategy 2026–2030 on 5 August, we completed our baseline measurement. When the Strategy landed, we read all 51 of its actions against that baseline and published a ten-page response briefing within days — the source text frozen and hash-pinned, the PDF built by a script that refuses any other input, every figure validated before posting.

The briefing's finding was deliberately balanced: supplier naming on newly published records had reached 95.2%, and the Strategy's transparency pillar is substantive. One gap remained, sitting directly on the value-for-money agenda: the Strategy never uses the word "identifier" — and without a common, resolvable supplier identifier, the analytics platform it commits to will aggregate spend it cannot resolve. Our ask was a decision, not a contract — specify the identifier in 2026, before the data-sharing agreements are drafted.

We mapped the access route, registered where the law required, and delivered a four-artefact pack to the Minister for Public Expenditure: the published briefing, a one-page decision brief, a vehicle-neutral technical specification his officials could adopt without attribution, and an exhibit of our parliamentary-questions analysis. To keep the evidence clean, Amata committed publicly not to tender for the Strategy's analytics platform or any work arising from its transparency actions, through 2030.

In September 2026, the Minister's office acknowledged the submission, welcomed the contribution of evidence-based external analysis, and referred the materials to the Office of Government Procurement — the body that owns the identifier decision — for consideration and direct follow-up.

The measurement continues on a published schedule: quarterly re-observations against preregistered predictions, annually to 2030.

The flagship briefing

Progression

What it achieved

Measurement before the measured event. The baseline was completed 17 days ahead of the Strategy's publication — the response briefing could not be accused of hindsight.

A number now on the record: 95.2% supplier naming on newly published awarded records, reproducible from the on-request methods package.

A verified negative on the record: the Strategy's published text never uses "identifier" or "purchase order" — checked across both PDF editions.

A specific decision put in front of the decision-maker, with a specification officials can adopt without attribution — and a documented referral to the office that owns it.

Independence engineered, not claimed: a public no-tender undertaking through 2030, and findings that credit the system's genuine improvements.

A reusable apparatus: roster, monitor, dataset, preregistrations and audit protocol all persist, producing the next instalments on schedule.

The published artefacts

Delivered to the Minister's office

Described, not linked: these were built for the decision-maker, not for traffic.

The verification apparatus

As-published audits: the live render of each piece re-verified against its ratified text after posting, deltas logged. Publication treated as an observable event, not an endpoint.

Preregistrations P1–P6: predictions filed before the October re-observations that test them. Removes the ability to quietly move the goalposts.

A running, numbered corrections log and right-of-reply standing on everything published — the answer to "why should anyone trust an interested publisher."

Outcome

On 10 September 2026, the office of the Minister for Public Expenditure, Infrastructure, Public Service Reform and Digitalisation acknowledged Amata's submission, welcomed the contribution of evidence-based external analysis to procurement policy discussion, and referred the materials to the Office of Government Procurement for consideration and direct follow-up. The OGP is the body within which the supplier-identifier decision sits.

Dated evidence