5 Reasons Why AI Adoption Is Falling Short for NZ Businesses

This article is part 1 of the myITmanager Practical AI Series, sharing business-first insights to help New Zealand organisations move from uncertainty and experimentation to structured, confident technology adoption.

Executive Summary

Many New Zealand businesses are experimenting with AI, but few are seeing measurable productivity gains. The issue isn’t access to tools. It’s the absence of structure.

AI adoption typically falls short when organisations skip business discovery, lack a clear roadmap, fail to embed AI into everyday workflows, rely on training alone, or overlook governance. Sustainable value comes from a coordinated, business-led approach, not isolated experimentation.

AI is becoming part of everyday work, whether businesses plan for it or not

Across organisations, AI usage has grown quickly. Staff are using it to draft emails, summarise information, analyse data and speed up routine tasks. When applied well, these tools can improve consistency, reduce rework and support better decision-making.

But in many businesses, this use is happening quietly and independently.

Leadership teams often don’t know where AI is being used, what data is being shared, or whether any real productivity improvements are occurring. Without coordination, AI becomes something individuals experiment with rather than something the business can intentionally harness.

This is where many organisations begin to fall behind.

Reason #1: AI is adopted without understanding the business first

In many cases, AI adoption begins with tools or training rather than business priorities.

Few businesses start by examining how work actually flows, where bottlenecks exist, or which outcomes matter most. Instead, AI is introduced because it’s available, popular, or perceived as necessary to keep up.

The result is predictable - AI activity becomes disconnected from operational impact. Effort is being made, but its value remains unclear. Put simply, without first understanding processes and objectives, businesses struggle to identify where AI can genuinely improve performance.

Reason #2: There is no clear plan or roadmap

AI adoption often grows organically. One team trials a tool, another attends a session. Usage then expands without any structure.

What’s missing is a prioritised roadmap. While many organisations are experimenting with AI, far fewer are achieving consistent, business-wide impact, hence the gap is not technology - it’s coordination.

Without a clear plan, productivity gains are difficult to measure, initiatives compete for attention, and leaders lack defined success metrics. So, momentum builds, but direction is missing.

Over time, enthusiasm fades and AI becomes more sporadic rather than embedded in the business in a planned way.

Reason #3: AI is not embedded into real workflows

Even when AI tools are available, they are often layered on top of existing processes rather than integrated into them. This creates duplication instead of transformation.

If AI does not meaningfully reshape workflows by reducing steps, removing manual handling or automating decisions, it becomes “optional” - meaning usage varies between individuals, and benefits remain inconsistent.

For AI to deliver impact, it must change how work happens, not simply sit alongside it.

Reason #4: Training happens, but adoption does not follow

Training is valuable as it builds awareness and confidence, but training alone does not lead to adoption.

That's because knowing how to use a tool doesn’t answer bigger questions: where AI should be applied, what tasks should be automated, how outputs should be reviewed, or how all of this aligns to business goals.

Without integration and reinforcement, AI remains an add-on rather than part of everyday operations.

Reason #5: Governance and security are treated as an afterthought

Perhaps the most significant risk of ad-hoc AI adoption is what remains invisible.

Unmanaged AI use can result in sensitive information being entered into public tools, inconsistent outputs influencing decisions, or unclear accountability for AI-generated work. As business AI capabilities expand, so does the need for governance.

It is therefore essential that security, data management and responsible usage evolve alongside AI adoption rather than follow behind it.

A more strategic way forward

Businesses that achieve meaningful value from AI approach it differently. 

They: 

1. Begin by understanding how the business operates
2. Define priorities before deploying tools
3. Embed AI into workflows rather than layering it on top
4. Reinforce adoption through structured support
5. Govern AI use as part of their wider technology environment.

Now this is where AI delivers measurable ROI, as it becomes part of how the business operates, rather than a collection of apps used in an ad-hoc way.

Watch out for Part 2 – coming soon.