Why am I losing tenders? A self-diagnosis guide for NZ contractors

Steve Parker
Jul 4
11 min read

Updated: 4 days ago

Another "we regret to advise" landed this morning, with no scores, no debrief, and no clue. You are pricing more than ever and winning less than ever, and the only advice anyone offers is to sharpen your pencil. Before you cut margin blind, work through this page — most lost tenders trace to one of seven diagnosable causes, and on government work you are legally entitled to be told which one applied to you.

By Steve Parker · Trueworks · NZ construction estimation · 11 min

Quick answer: NZ contractors lose tenders for seven diagnosable reasons: bidding the wrong jobs, non-conforming submissions, scattered pricing, misread scope, visible risk loading, weak presentation, and relationship or payment-risk pricing. On government tenders, Rule 33 of the Government Procurement Rules (5th edition) entitles you to a debrief covering why you lost and how you scored. On private tenders, nobody owes you an answer.

If this is your week

"We regret to advise your tender was unsuccessful" is the only feedback you have had in six months.
You suspect you are too high, but cutting margin blind is a race to the bottom.
The jobs you do win are the ones you least wanted — and they run hard.
You have started nudging rates on instinct, and the instinct has no data behind it.

This page is the diagnosis you were never given. It covers what a normal win rate looks like, the seven reasons tenders actually get lost, how NZ evaluation panels score you, and — the part almost nobody uses — how to extract real tender feedback from a government agency that is required to give it to you.

First, the numbers: what a normal win rate looks like

There is no published NZ-specific construction win-rate statistic, so here are the closest usable benchmarks. International industry data puts the average commercial contractor at roughly a 25% bid-win rate — two to three wins in every ten bids. Open public-sector tenders run harder, typically 10–20%. Negotiated and selective work converts at 30–50%, and consistently strong performers on repeat relationships reach 40–50% or better.

NZ industry commentary is consistent with the low end: on one NZ construction podcast, practitioners across several episodes put it at one win in every three to nine bids depending on trade and market heat, and estimated the industry burns something like 30% of project value on tendering costs across all the losing bidders. Treat those as practitioner claims rather than audited statistics — but the shape is right.

Two takeaways before you diagnose anything. If you are winning one in four open tenders, you are normal, not failing. And if you are winning most of what you price, that is not automatically good news — in a lowest-price market, a very high strike rate often means your mistakes are winning jobs. We have reviewed a roof quote that "won" at a margin of roughly minus 5 to minus 10% — a four-figure loss signed the moment the customer accepted.

The seven reasons contractors lose tenders

Work down the list in order. In our review work the cause is usually in the first four, and it is almost never the one the contractor assumed.

1. You are bidding the wrong jobs

Win rate starts at bid/no-bid, before any pricing happens. Chasing every tender dilutes the hours available for the ones you can actually win, and some packs should never be priced at all. We reviewed a "roofing tender" that was actually a whole-building reclad head contract — roofing was 15–25% of the value, with 20-plus sublet trades and a ten-year single-point warranty deed attached. A day spent reading the pack costs $500–1,000; pricing it anyway wastes $3,000–8,000 of estimating; winning it risks $50,000–150,000. Losing that one was the win.

2. Your submission never reached scoring

Evaluation panels check conformance before merit. Missing documents, unsigned forms, unmet pre-conditions, late lodgement, or tags and exclusions that breach the conditions of tendering can all see a bid set aside without a single rate being read. If your price was competitive and you still hear nothing, ask whether you were conforming — it is one of the specific things a government debrief must tell you. Tags are a particular trap: heads demand "no tags" while untagged risk is commercial suicide, so the skill is tagging in a way that stays conforming — priced tag options rather than blanket exclusions where the conditions allow it.

3. Your pricing has scatter

When every bidder estimates with some error, the lowest bid tends to be the biggest underestimate. If your own pricing has scatter, your mistakes win jobs and your good prices lose them — which is why the wins feel bad and the losses feel unfair. Scatter has mundane causes: stale rate libraries, inconsistent supplier pricing, and arithmetic method. The classic is the markup-versus-margin error — on one electrical tender we reviewed, multiplying the cost floor by the margin percentage instead of dividing by its complement would have quietly surrendered about $60,000 of intended margin. An audit of the rate calculator before tender is the cheapest fix in this list.

4. You priced a different job than the one in the documents

A price that quietly includes more scope than the next bidder's is not expensive — it is misread. The reverse loses differently: a scope gap that surfaces after award comes back as a back-charge. This failure mode is endemic because subbie bids are rarely comparable: in one head contractor's telling, fifteen electrical bids arrived in fifteen different formats, roughly half missing parts of the scope. When the evaluator cannot compare bids line by line, the default decision is cheapest number — and nobody can tell you what your price did or did not cover, because nobody worked it out.

5. Your risk is visible and everyone else's is hidden

Load contingency openly — say 5–15% for ground risk or design immaturity — and you lose to bidders who stripped it out and plan to recover through variations. Strip it yourself and you have bought a dispute. The same practitioners estimate NZ trade bids quietly carry something like a quarter of their value in smuggled waste and risk loading spread across rates, while visible, honest risk pricing gets punished at evaluation. The workable middle is a priced risk register attached to the tender, and hard exclusions on the genuinely unpriceable items — ground conditions being the canonical example.

6. Your document is losing, not your number

Format alone moves conversion. We compared quotes for one trade business that switched from plain itemised quotes to glossy lump-sum proposals — same work, same crews, same pricing policy — and conversion fell. Per-line pricing had let clients trim scope instead of saying no; the lump sum invited a yes/no decision on a bigger number wrapped in thirty defensive clauses. What your tags, exclusions, and layout signal to the person scoring you is part of the price they think they are reading.

7. The decision was never entirely about your bid

Price encodes relationships. One subcontractor's rule, recounted in the same series: double the price for a head contractor you distrust, halve it for one you would work for on a handshake — a four-times spread on identical scope, driven by payment risk, not estimating skill. If a competitor trusts the payer and you do not, they can sit under you with the same cost base. And some jobs are simply bought — priced below cost by a bidder who needs turnover. Those losses are not yours to fix, and cutting your margin to chase them is how construction ends up the biggest single contributor to NZ company liquidations — roughly a quarter of all of them, from an industry that is only about one in eight registered companies.

On government tenders, you are entitled to a debrief — use it

This is the most under-used right in NZ tendering. Rule 33 of the Government Procurement Rules (5th edition, in force since 1 December 2025 — it was numbered Rule 49 in the old 4th edition, so older citations are out of date) requires agencies to offer every unsuccessful supplier a debrief, and to deliver a requested debrief within 30 business days of contract signature or of your request, whichever is earlier.

The debrief must cover: the reasons your proposal was unsuccessful; how it performed against the evaluation criteria, including its relative strengths and weaknesses; and the relative advantages of the winning proposal. It must also address your questions and concerns. The agency may tell you where your price ranked, but not the winning price; the winner's identity is discussable once published on GETS. A debrief is not a forum to relitigate the decision — go in collecting data, not arguing.

Questions worth asking in the room:

Where did we rank on price, and where on non-price attributes?
Which criteria did we score weakest on, and what did the winner do differently there?
Was our submission fully conforming? Did any tag, exclusion, or omission cost us?
Was scope coverage a factor — did our price appear to cover more or less than others?
What would you want to see from us on the next similar tender?

A request needs two sentences:

Subject: Debrief request — [tender reference]
We were an unsuccessful tenderer on [reference] and would like to take up the debrief offered under Rule 33 of the Government Procurement Rules. We are happy with a phone call or meeting at your convenience — our questions concern our scoring against the evaluation criteria and the relative strengths of the successful proposal.

Do this on every government loss. It is fifteen minutes of the evaluator's time, agencies that debrief get fewer complaints and expect the request, and three debriefs give you a pattern no amount of guessing can.

On private tenders, nobody owes you anything

The Procurement Rules bind government agencies only. A private developer or head contractor has no obligation to debrief you, and most will not — partly policy, partly because, as the fifteen-formats story above shows, there is often no scoring record to reconstruct an answer from. You can still ask; the worst case is silence.

This asymmetry is where structured review earns its keep. When no feedback exists, the only diagnostic path left is your own tender file: your submitted rates against market bands, your scope coverage against the documents, your presentation against what an evaluator rewards. That is exactly what a post-loss review reconstructs — the debrief nobody will give you, built from evidence you already hold.

How NZ tenders are actually scored

Knowing the evaluation method changes how you should read a loss.

Lowest price conforming. Only the lowest bid's attributes are checked against pass/fail standards. If you lost under LPC, the answer is price or conformance — nothing else was scored.

Weighted attributes. Every bid is scored against weighted criteria and the highest total wins. Weightings vary by agency and package — on NZTA and Auckland Transport price-quality work the price weighting is typically held to around 20–30%, with the balance spread across methodology, track record, and health and safety — though practitioner sentiment is that scoring across the wider market often behaves closer to 60/40 price-heavy, with non-price attributes treated as a tick-box exercise. Under weighted attributes you can lose on methodology while being cheapest.

Price quality method. Common on infrastructure work (it is the method NZTA documents in its procurement manual). Quality scores convert into a Supplier Quality Premium deducted from each tenderer's price, and the lowest net figure wins — which is how the cheapest conforming bid can still lose to a dearer bid with a stronger quality score.

Quality-based selection. Attributes first, price negotiated after — rare in trade packages.

One recent change worth knowing: from December 2025, covered government procurements must include an "economic benefit to New Zealand" criterion at a minimum 10% weighting. The Rules' coverage thresholds are $100k for goods and services but $9 million for construction works, so trade packages will meet it mainly on larger public jobs — where, if you are NZ-owned, employ locally, and train apprentices, that is now scoreable material. Say so in the submission.

Classify every loss before you change anything

Every lost tender goes in one of four buckets: price (rates genuinely above market), scope reading (you priced a different job), presentation (the document lost it), or bought (the winner went below cost). Only the first three are yours to fix, and cutting margin only treats one of them.

The classification needs evidence, not memory. Three tenders is a working minimum; five or six give real signal. Include the wins — a book where the winners are your thinnest prices is telling you something the losses cannot. Across a set, the pattern is usually unmistakable within a week of looking.

Where Trueworks fits

We run this as a structured post-loss review: your last several tender packs and submissions, taken apart along the seven seams above — rates benchmarked item by item against market bands, scope coverage checked against each document set, presentation reviewed against what evaluators reward, margin structure mapped, and every loss classified. Findings in writing inside 5 business days for the first pass, then the same discipline applied pre-submission on your next tender, while changing the number costs nothing but time.

No charge for your first review packet. NDA available as standard. Files processed on NZ-hosted systems and deleted after 30 days unless you ask us to keep them, and your rates are never benchmarked across clients or reused in anyone else's review.

→ Email hello@trueworks.co.nz or start at trueworks.co.nz/contact

FAQ — losing tenders in NZ

Why was my tender unsuccessful?

Usually one of seven causes: wrong job to bid, non-conforming submission, pricing scatter, misread scope, visible risk loading, weak presentation, or a decision driven by relationships or a bought price. On government work, ask for the Rule 33 debrief and the agency must tell you which. On private work, a post-loss review of your own tender file is generally the only way to find out.

How do I get tender feedback in NZ?

Government tender: request a debrief — the agency must offer one and deliver it within 30 business days, covering your weaknesses against the criteria and the winner's relative advantages. Private tender: ask anyway, but expect little; the fallback is reconstructing the answer from your own submissions across several tenders.

What is a tender debrief and what must it cover?

A short session (typically 30–45 minutes, by phone, email, letter, or meeting) where the agency explains why you lost. Under Rule 33 it must give the reasons, your performance against the evaluation criteria including strengths and weaknesses, and the successful proposal's relative advantages, and it must address your questions. It is not an avenue to challenge the result.

What if the agency refuses a debrief, or it tells me nothing?

Put the request in writing and cite Rule 33 — the debrief is an obligation, not a courtesy. If it stalls past the 30-business-day window or the answers are non-answers, escalate to the agency's procurement lead, and beyond that through MBIE's government procurement supplier feedback channel. A refused debrief is itself worth reporting.

Am I entitled to know who won, and at what price?

The winner's identity, yes, once the award is published on GETS. The winning price, no — agencies may tell you where your price ranked, but not the number that beat you.

What is a good tender win rate?

Around 25% is a normal strike rate for open commercial bidding internationally — two to three wins in ten. Open public tenders run 10–20%; negotiated and repeat-relationship work 30–50% or better. There is no published NZ construction figure. If you win far more than that in a lowest-price market, check your margins before celebrating — underpriced wins are losses on a delay.

Is a post-loss review confidential? These are our submitted prices.

Yes. We work under NDA as standard, files stay on NZ-hosted systems, and they are deleted after 30 days unless you tell us to keep them. Your rates are never benchmarked across clients or reused in anyone else's review.

The short version

You cannot fix a loss you have not classified. Get the debrief on every government tender because it is your right; classify every private loss from your own file because nobody else will; and before you cut a single rate, find out whether price was even the problem. In our reviews, about as often as not, it was not.

Trueworks is Steve Parker — 20 years on the analytical side of NZ construction, now doing quote checks, tender pricing packs, and post-loss reviews for NZ trades and builders. If you want the seven-seam review run on your last few tenders, send them in for a post-loss review; the first packet costs you nothing.