Content layer
Evidence library
We do not publish invented company case studies. What we publish instead is every public number this site depends on, with its source and its date — plus the documented shadow-AI behaviour that motivated the whole exercise. Read the sourcing rules in methodology.
87% of NZ organisations use AI. 12% have scaled it.
The gap is not adoption, it is implementation. That gap is the product: moving a firm from "someone tried it" to "it runs here".
Source: embedAI — The State of AI in New Zealand Business · 2026-04
Metric on record: 87% → 12%
81% of staff bring their own AI tools to work; 13% of NZ SMEs have an AI policy.
Two thirds of your crew are already tools-down before you have bought anything. The policy is not a nice-to-have; it is the missing piece.
Source: Microsoft / EMA workforce survey · 2026
Metric on record: 81% BYOAI · 13% policy
SMEs using AI earned roughly NZD 400,000 more on average than those that did not.
This is the number to quote when the question is "why bother". It averages across all industries — use it as a conversation opener, not a promise.
Source: Deloitte Access Economics for 2degrees — Productivity Propelled · FY25
Metric on record: ≈ NZD 400k
Building small firms produce NZD 74.00 per hour, against a national small-business average of NZD 76.30.
Building is below the national average on output per hour while carrying the highest admin load per tender. Any hour saved is margin.
Source: Xero small business productivity benchmarks · 2026 Q1
Metric on record: NZD 74.00 / hr
59% of building decision-makers are worried about going bust or closing within 12 months; 35% name AI and digital as a growth contributor.
A business under that much pressure will buy the thing that pays for itself in one tender cycle. That is the sales argument.
Source: Kennards Hire — Construction Confidence Check 2026 · 2026
Metric on record: 59% at risk · 35% betting on AI
NZ construction firms are documented putting client names, project scopes and priced quotes into public AI tools.
Estimators paste the price; project managers past meeting recordings that contain variation and claim discussion. These are not hypotheticals, they are recorded.
Source: NSP — Shadow AI in Construction Firms in New Zealand · 2026
Metric on record: 3 recorded scenarios
AI Forum New Zealand runs an AEC working group; formalising AI policy is on its agenda for architecture, engineering and construction.
The industry body already agrees the problem is governance. That makes an industry policy template a mainstream ask, not a niche one.
Source: AI Forum New Zealand — AEC working group · 2026-05
Metric on record: industry working group
New Zealand Strategy for Artificial Intelligence and responsible-AI guidance for businesses are published by the Crown.
Cite the Crown, not us, when you need to justify a decision to a bank, an insurer or a client board.
Source: MBIE / New Zealand Government — AI strategy and responsible AI guidance · 2025-07-08
Metric on record: official guidance
Around 81,000 building and construction businesses in New Zealand — the single largest industry count in the country.
1% of that is 810 firms. That is the realistic first-year market for a paid diagnostic, and the whole thesis of going narrow.
Source: business.govt.nz — Data for business · latest
Metric on record: ≈ 81,000 firms
New Zealand AI readiness scoring exists, but not industry-specific, and not as a downloadable, dated report.
The competitor built the category of "NZ AI readiness". The open lane is the same score for one industry, with the compliance pack attached.
Source: embedAI (NZ AI status reporting) · 2026
Metric on record: generic, not vertical