A 10% economic-benefit weighting is an evidence problem.

The 5th edition Government Procurement Rules put economic benefit to New Zealand into every government procurement, with a minimum 10 percent evaluation weighting. Most suppliers are still answering it as a writing exercise.

The 5th edition of the Government Procurement Rules came into effect on 1 December 2025. It reduced the number of rules from 71 to 47, strengthened transparency requirements around publishing procurement policies and contract awards, tightened the rules for setting up and running supplier panels and secondary procurement, and introduced a new requirement for agencies to include economic benefits to New Zealand in all procurements — with a minimum 10 percent weighting on economic benefit criteria for evaluation purposes.

Ten percent sounds modest until you look at how close competitive tenders actually are.

Ten percent is frequently the whole margin

On a well-run procurement with three or four credible respondents, the gap between first and second is often a few points. A criterion carrying a guaranteed minimum of 10 percent is therefore rarely decorative. It is a criterion that can move the ranking on its own — and unlike price, it is scored on assertions the supplier chooses to make.

Assertions the supplier chooses to make are exactly the assertions nobody in the organisation is tracking.

What actually goes wrong

Economic-benefit responses tend to fail in four predictable ways, and none of them are writing failures.

1. The claim is generic. "We support local employment" scores poorly against a competitor who names the region, the roles, the number and the mechanism. Specificity requires somebody to go and find the real numbers, which takes longer than the bid programme usually allows.

2. The claim is unevidenced. A supplier states a local-spend percentage or an apprenticeship commitment with nothing behind it. Some evaluators will discount it. Some will ask at clarification, at the worst possible moment.

3. The claim is inconsistent with the rest of the submission. The economic-benefit section promises local subcontracting while the methodology section names an offshore supply chain, or the programme section shows a resourcing profile that cannot deliver the training commitment. Different authors, different sections, no reconciliation.

4. The claim becomes a contractual commitment nobody tracked. This is the expensive one. A commitment made to win the work is now a deliverable, and the people who have to honour it were not in the room when it was written.

The same claim appears in every bid, differently

Most organisations answer economic-benefit questions repeatedly, for different agencies, in slightly different words, with slightly different numbers — because the last answer is buried in a submission folder and it is faster to rewrite than to find.

That is how a supplier ends up with three different local-content figures in three live bids, none of which are wrong exactly, and none of which can be reconciled if a client asks.

The underlying discipline is unglamorous:

• A claim should exist once, with an owner.
• Each claim should carry its evidence and the date that evidence was last verified.
• Every use of the claim in a submission should point back to the same source, not a retyped variant.
• When a claim becomes a commitment in a won bid, it should be visible to delivery, not archived with the proposal.

This is a records problem before it is a bid problem

The 5th edition also makes agencies publish procurement policies and contract awards. Combined with a mandatory economic-benefit criterion, the direction of travel is clear: more of what a supplier says is going to be visible, comparable and durable.

Suppliers who treat each submission as a standalone document will keep re-deriving the same claims. Suppliers who maintain an evidence library — with owners, verification dates and prior usage — answer faster, more specifically, and consistently across concurrent bids.

That advantage compounds. It is the same argument as keeping addenda connected to the work they affect: the value is not in producing the document, it is in the organisation knowing what it has already said and what it can prove.

A practical starting point

Before the next government submission, an organisation can usefully ask itself three questions:

Can we list every economic-benefit claim we made in the last twelve months? If that requires opening submission files one by one, the claims are not under control.

For each claim, who owns it and what proves it? If the answer is a person's memory, the claim is a risk in both directions — understated in the bid, or unhonoured in delivery.

Which of those claims are now contractual commitments? If nobody outside the bid team knows, the organisation has bought obligations it is not managing.