Pragmatic AI strategy, from someone who's seen waves before.

I've been inside NZ business for 26 years. ANZ in 2000, BNZ from 2004 to 2014, then founded my own FAP-licensed financial advisory firm (Wealth Health) in Tauranga in 2014, which I still run. That's three different vantage points on how NZ businesses adopt technology.

I've watched waves before. Internet banking in the early 2000s. KiwiSaver in 2007. Cloud accounting around 2012. Open data after the FMA reforms. Each one followed the same pattern: an early period of confusion and hype, a handful of firms moving early and well, a much larger group buying the wrong tools or none at all, and a tail of late movers paying twice to catch up.

AI in 2026 is the same pattern, faster. The difference is the pace. Where cloud accounting took five years to settle out, AI tooling is moving on a quarterly cycle. The headline tool from January is rarely the headline tool by October. Firms that paid for the January tool are mid-contract, locked in, often barely using it, and watching better tools arrive monthly. The cost of getting it wrong is higher and the window to course-correct is shorter.

That's why this practice exists. The market is full of consultants happy to confirm that AI is the answer to questions you haven't asked yet. There aren't enough people willing to look at your specific business, run honest diagnostics, and tell you straight where AI moves the needle, where it doesn't yet, and what to fix first.

I started my finance career at ANZ in 2000, just out of Massey University. Four years inside ANZ taught me banking operations at scale, in a period when the whole industry was moving from paper-and-branch into digital-and-call-centre. It was unglamorous, and it was the best foundation I could have asked for.

From 2004 to 2014 I was at BNZ, working in commercial banking, then private banking, then financial advisory. Ten years across the Bay of Plenty and Waikato regions, sitting across the table from owner-operators in trades, professional services, primary industries, and tourism. That's where I learned how NZ SMBs actually work: how owners think about risk, how teams adopt or resist new tools, how a board approves spend, how an accountant sanity-checks a number.

In 2014 I founded Wealth Health, a Tauranga-based financial advisory firm. We took the FAP licensing route when it came in, completed CFP certification, and built a practice that now advises hundreds of NZ households on retirement, KiwiSaver, insurance, and investment. Wealth Health is still operating and still my day job in financial advice.

The shift toward AI consulting started organically in 2024. Wealth Health clients (and other business owners in the network) started asking the same questions on repeat: which AI tool should I buy, is it safe to use ChatGPT with client data, is my accountant's AI feature actually doing what they claim, do I need this or am I being sold to. I started answering those questions semi-formally for a handful of firms.

By late 2025 it was clear two things were true. First, the questions were real and the gap in the market was real. Second, AI consulting and financial advice are different practices with different ethics, different regulatory frames, and different deliverables, and they deserved separate brands. Coupland Consulting launched in 2026 to be that separate practice.

Financial advice in NZ is regulated. AI consulting isn't. So when you're choosing an AI advisor, the question isn't "do they have a licence." Anyone can put up a website and call themselves an AI consultant tomorrow. The question is what evidence they bring that they can do the work.

Here's what I think actually matters, and what I bring:

• Discipline from a regulated field. CFP certification and FAP licensing aren't just titles. They're a way of working. Document the basis for advice. Disclose the limits of what you know. Treat client data with the seriousness it deserves. Get a peer review before you publish. None of that is required in AI consulting, all of it should be.
• Time inside NZ business. 26 years of sitting across the table from NZ owners and boards. I know how an NZ accounting firm actually works, not just how the brochure says it works. I know how a mid-sized law practice debates a software purchase. I know what a typical Bay of Plenty SMB's IT setup looks like in real life.
• Plain English about complicated things. 10+ years of explaining KiwiSaver, FIF tax, gearing, sequence-of-returns risk, and FAP conduct rules to people who didn't go to finance school. That practice translates directly to explaining what a large language model does, what it doesn't, where it's likely to be wrong, and what guardrails matter.
• Hands-on with the tooling. I use the major AI systems daily, inside both Wealth Health (with appropriate compliance) and Coupland Consulting. Claude, ChatGPT, Copilot, Gemini, Perplexity, Cursor, and a long tail of specialist tools. Not because I love AI, but because you can't advise on tools you haven't lived inside.
• No vendor money. No commissions, no referral fees, no kickbacks from AI vendors. Ever. This isn't a moral position, it's a structural one: if my income depends on selling you a particular tool, my advice is worth less. So we built the model to remove that.

Things I'm explicitly not: a software engineer, a data scientist, an ML researcher. I have working knowledge of all three fields and partner with specialists when an engagement needs them. What I bring is the bridge between what those specialists can build and what an NZ business owner actually needs.

Two practices, one person, separated. Here's why.

Wealth Health is a FAP-licensed financial advice provider, regulated by the FMA. It gives personalised financial advice to NZ households and small businesses on retirement, KiwiSaver, insurance, and investment. Every interaction is governed by the FAP code of conduct, with documented advice records, suitability assessments, and conflict-of-interest disclosures. That's what financial advice requires, and it should.

Coupland Consulting is an AI strategy practice. It advises NZ businesses on AI readiness, tool selection, workflow integration, and risk management. AI consulting is not financial advice. We don't recommend investments, we don't manage money, we don't advise on insurance. So we don't operate under the FAP code for this work, because the FAP code wasn't written for AI consulting and applying it badly would be worse than not applying it at all.

Both practices share Craig as the principal, and both practices share the same standards on disclosure, documentation, and conflict-of-interest handling. The difference is the regulatory frame applied to each, which matches the actual work being done. If a Wealth Health client asks about AI, they get pointed to Coupland Consulting. If a Coupland Consulting client asks about KiwiSaver, they get pointed to Wealth Health. Clean separation, no conflict of interest, no awkward overlap.

The practical implication for AI consulting clients: you get a senior practitioner who treats your engagement with the discipline of someone who runs a regulated practice in parallel, not someone who'll cut corners because nobody's checking.

I genuinely think generative AI is the biggest shift in how knowledge work gets done since the spreadsheet. I also think it's being oversold in 90% of public commentary, and undersold in the 10% that matters.

What's exciting to me: the gap between what a small NZ professional services firm can do today and what an offshore mid-market firm can do is collapsing. Tools that used to cost six figures a year are now $20 a seat. A four-person Bay of Plenty practice can credibly compete on output quality with a 40-person Auckland firm, if they pick the right tools and build the right workflows. The playing field is genuinely flatter than at any point in my career.

What's overhyped: "AI agents will replace your team." Almost never the right framing for the next 24 months. The real wins are smaller and more specific. A research workflow that used to take three hours now takes 40 minutes. A first-draft proposal that used to need a senior partner's time now drafts itself overnight for review in the morning. Multiply that by a team of 12 across 200 working days a year. That's the real revolution, and it's quiet.

What's underhyped: the gap between firms that figure this out now and firms that wait until 2028. Same pattern as cloud accounting. The firms that moved early didn't just save costs, they restructured their offer and won market share. By the time the laggards caught up, the leaders had already moved to the next thing. I expect a similar pattern in professional services over the next 24 months.

That's what Coupland Consulting is built for. If you're an NZ business owner who wants to be in the moving-early-and-well group, but doesn't want to be sold to by every vendor with a deck, that's exactly the conversation we're set up for.

Craig.