An alliance shortlist is decided long before anyone prices the work.

Watercare has opened the Registration of Interest for the $1.7 billion Waikato Water Supply Programme. The commercial contest that matters happens in the next few months, and it is not a pricing contest.

On 2 September 2026 Watercare opened an ROI process seeking Alliance delivery partners for the programme. The scope is a new raw water intake, increased treatment capacity, roughly 50 kilometres of pipeline and a 50-megalitre reservoir at Pukekohe West, targeted for completion by the end of 2033.

The published timeline is unusually explicit about where the field narrows:

• September 2026 — ROI opens, through GETS.
• Late 2026 — up to three proponents shortlisted through an RFP process.
• Late 2027 — preferred Alliance partner selected.
• Early 2028 — construction expected to start.

Read that again from a bid team's perspective. Between now and the end of the year, the market goes from every credible water and civil contractor in the country to three. Everything after that is a year of development with a much smaller group.

Alliance selection asks for a different kind of proof

Watercare chief executive Jamie Sinclair said the Alliance model was selected because of the scale and complexity of the programme — integration with existing infrastructure, delivery timeframes, environmental considerations and the cultural significance of the area. The stated objectives include resilience, growth and drought protection, delivered through partnership with mana whenua and alignment with Te Ture Whaimana o Te Awa o Waikato.

None of that is answered with a rate. It is answered with evidence: who has actually done this, on what, with whom, and what happened.

A hard-dollar tender asks what you will charge. An alliance selection asks what you can prove about how you behave.

That shifts the bid team's problem from pricing accuracy to evidence assembly under time pressure. The questions a proponent has to answer well are the ones an organisation usually answers badly:

• Which of our completed projects genuinely demonstrate this scope, and where is the documentation?
• Which named people are we committing, and are they actually available across a delivery window that runs to 2033?
• What can we evidence about how we behaved when a previous alliance or partnership went badly?
• What do we claim about environmental and cultural outcomes, and can each claim be substantiated?
• Which capability do we not have, and who do we need alongside us to close it?

The last question is the one that creates the mess

Very few organisations will satisfy an intake, a treatment plant upgrade, 50 kilometres of pipeline and a reservoir on their own. Most credible responses will combine contractors, designers and specialists.

The moment more than one organisation is in a submission, the control problem multiplies. Requirements have to be mapped to whichever party can satisfy them. Evidence lives in someone else's document management system. Claims are drafted by people who do not report to the same manager. Commitments made in one partner's section have to be consistent with another partner's programme.

This is the same structural problem regional industrial alliances face when they try to act as a single front door — we wrote about that here. The unit of work is the opportunity, not the company.

Then the source starts moving

A multi-stage procurement running from an ROI in September 2026 to a preferred partner in late 2027 will not stand still. Documents get clarified. Requirements get sharpened between ROI and RFP. Interactive workshops change the buyer's expectations.

An organisation that treats each stage as a fresh document exercise loses the thread. The claims made at ROI, the people named, the evidence cited and the commitments implied all have to remain consistent — and defensible — a year later, in front of a client who has been reading the earlier submissions the whole time.

What good looks like here

Nothing in this requires new technology to state plainly. It requires that a bid organisation can answer four questions at any point in a long pursuit:

1. What exactly has the client asked for, and which version of the documents are we working from?

2. Which requirements can we satisfy ourselves, which need a partner, and where are the gaps?

3. For every material claim we are making, what is the evidence and who approved it?

4. What did we commit to in the last submission, and is it still true?

Most organisations can answer those questions eventually, by asking people. The cost is senior time, and the risk is that the answer arrives after the submission has gone.

The wider point

The Waikato programme sits inside a $15.2 billion Watercare capital programme of more than 1000 projects across nine major infrastructure programmes. It is one procurement among many, in a market where the national pipeline reached $290 billion in June 2026.

Organisations that pursue this kind of work repeatedly are not short of opportunities. They are short of the operating discipline to pursue several of them at once without losing control of requirements, evidence and commitments.