Practical AI adoption · for New Zealand businesses
AI can improve a business, but it cannot fix the foundations for you
I help New Zealand businesses adopt AI as part of a wider improvement journey. That starts with safe personal productivity, but the larger opportunity is usually in understanding how work really happens, connecting the systems that should already be talking to each other and automating the parts of the process that need to be reliable. AI can then be added into the specific moments where interpretation, drafting or judgement creates value.
The position
AI is not the transformation
Tools such as ChatGPT, Claude and Microsoft Copilot can already help people write, summarise, research and think more quickly. Those gains matter, and helping a team use the tools safely is a sensible place to begin. However, personal productivity does not automatically improve the way the business operates.
If customer information is spread across several systems, approvals still move through email, people repeatedly enter the same data or critical processes rely on individual memory, AI will not remove the underlying friction. It may make one person faster inside a process that is still poorly designed. The more sustainable opportunity is to improve the process itself and then use AI where it adds something ordinary automation cannot.
The outcome is not an AI business. It is a better-run business that uses AI well.
The adoption journey
Adopt AI in the right order
Each stage creates value in its own right. Many businesses gain more from the middle stages than from any agent. You do not need to progress through all five.
Start by helping your people use AI safely and productively
Most businesses do not need to begin with a custom AI platform. They can create useful early value by giving people access to an approved tool, helping them understand what information can and cannot be shared, and developing the practical skills to use AI as a thinking and productivity partner. The objective is not a library of clever prompts. It is to help people recognise where the tool is useful, where human judgement still matters and how to check the quality of the output.
The larger opportunity is often in the processes between your systems
Small businesses frequently grow by adding people, applications and workarounds as each new need appears. Over time, information is copied between systems, spreadsheets become unofficial databases, and approvals depend on someone seeing an email. Before adding more intelligence, I help map how the work actually flows, identify the steps that do not add value, clarify where the right information should live and decide what should be integrated or automated. Some of the best improvements will have nothing to do with AI.
Automate the work that needs to be consistent
When the same input should always produce the same result, ordinary automation is usually the right answer. Moving information between systems, checking a defined rule, generating a standard document, routing an approval or creating a record should be handled by a reliable, predictable workflow rather than an AI model — it will be more consistent, easier to trust and easier to check. AI is better saved for the steps that involve language, interpretation or judgement.
Add AI where interpretation creates value
Once the process is clear and the surrounding workflow is reliable, AI becomes much more useful. Rather than asking a model to run the whole process, it can perform a focused task and then hand the work back to automation or a person.
Build agents only when the business can support them
More capable assistants and agents can create value when they have a clear purpose, appropriate access to reliable information, defined actions they are allowed to take and a process for handling uncertainty or exceptions. Without those foundations, an agent can simply make an unclear process move faster and less visibly. It should be pursued because the business case and process design support it, not because an agent is the most fashionable form of AI.
Only when readyBest fit
Add AI where interpretation creates value — with clear controls
In each example, AI contributes a specific capability while the process retains clear controls and accountability. The label shows which step is AI, which is automation and which is human.
Customer enquiry
Document processing
Meeting follow-through
How I can help
Start with the friction your team feels today
You do not need to decide in advance whether the answer is AI, automation or process improvement.
A useful starting point is to identify where your people are losing time, where work becomes inconsistent, where customers wait or where information repeatedly falls between systems. I can help you understand and prioritise those opportunities, map and redesign the process, define the system and automation changes, establish safe AI use and shape targeted AI-enabled workflows. Where specialist development or platform configuration is required, I can help brief, select and work alongside the appropriate technical provider so that the implementation remains connected to the business outcome.
AI productivity workshop
Safe, practical use of approved tools across everyday work.
Opportunity & readiness review
Find where AI, automation or process improvement will pay off most.
Process & automation sprint
Redesign one high-friction workflow and implement real improvements.
What a first step looks like
A sensible starting point, without a big commitment
You do not need a strategy, a budget for a platform, or a decision about AI before we talk. A short first piece of work is usually enough to see where the real value is.
A typical first engagement
- Starts with a conversation about where work feels slow, manual or inconsistent — no preparation required
- Looks at your actual processes and systems, not a generic AI maturity model
- Gives you a prioritised, plain-English view of where process, automation or AI will pay off
- Leaves you free to act on it yourself, with your existing providers, or with my help
- Does not commit you to buying any tool, platform or larger programme
Discuss where to start
The first useful step may not involve AI at all
That is not a failure of the AI conversation. It is how a business creates the foundations to use AI safely and sustainably. If you can describe where the work currently feels slow, manual, inconsistent or disconnected, we can begin there and work out what combination of process, integration, automation and AI will create the most value.