This morning we sat down with a Gold Coast manufacturing business and did not open a single product demo. No tools, no pricing pages, no feature comparisons. We just talked. We mapped out which roles in their business are most likely to change as AI becomes more embedded in their operations, what their team is actually worried about, and what those people need to understand before anything gets switched on. That conversation took a few hours. Most consultancies never have it at all.
The Step Most AI Rollouts Skip
The standard AI adoption playbook goes something like this, identify a problem, find a tool that claims to solve it, buy a licence, send a training video, and hope the team figures it out. It is fast. It looks decisive. And it fails more often than it succeeds.
The research is consistent on this. Around 70% of AI adoption failures come down to human and process challenges, not technical ones. The tools generally work. The people side is where things fall apart. And yet the default approach in most businesses is still to sort the technology first and trust that the people will catch up.
They usually do not. At least not without help.
What we did this morning was the step that comes before any of that. Not because it is a nice thing to do, but because it is the thing that determines whether everything that follows actually sticks.
What the Conversation Actually Covers
When we sit down with a business before any tool is on the table, we are working through three things.
First, which roles are about to change. Not in a vague, theoretical sense. Specifically. In a manufacturing context that might be quality inspection workflows, scheduling coordination, supplier communications, or documentation. AI is already touching all of those. The question is not whether it will affect those roles. It is how much, how fast, and who needs to be ready.
Second, what the team is actually worried about. This is the part most businesses skip because it feels uncomfortable. People in operational roles often have real, grounded fears about AI. Some worry their job will be automated away. Some worry they will be left behind because they are not technical. Some worry their experience and judgment will suddenly count for less than a system they do not understand. Those fears are not irrational. They are reasonable responses to genuine uncertainty. If you do not surface them early, they become the friction that stalls your implementation later.
Third, what people need to learn before anything gets implemented. Not a generic AI literacy course. Something specific to their role and their workflow. A machine operator and a production planner need different preparation. A one size fits all approach to upskilling is one of the most common ways AI rollouts lose their people along the way.
Why Manufacturing Businesses Need This More Than Most
Manufacturing in Australia is under real pressure right now. Labour costs, supply chain complexity, and global competition are all pushing businesses to look for efficiency gains wherever they can find them. AI offers genuine ones. Predictive maintenance, demand forecasting, process optimisation, and documentation automation are all areas where the technology is mature enough to deliver real value.
But manufacturing workforces also tend to have a high proportion of people in roles that have been stable for a long time. People who are very good at what they do, who have built their expertise over years, and who have not necessarily needed to engage with rapidly changing technology before. When AI arrives in that environment without preparation, it does not just create technical challenges. It creates trust challenges. People who feel blindsided by change do not become enthusiastic early adopters. They become quiet resisters, and quiet resistance kills implementation momentum faster than any integration problem.
The businesses getting this right in manufacturing are the ones that gave their teams time to ask questions before the tools arrived. That time is not wasted. It is the investment that makes everything else work.
The Readiness Work Is Not Soft. It Is Strategic.
There is a version of this that gets dismissed as touchy feely people management, separate from the real business work of AI adoption. That framing is wrong, and it is expensive.
When a business owner skips the readiness conversation and goes straight to implementation, they are not saving time. They are borrowing it. The problems that were not addressed at the start show up later, usually during rollout, when they cost more to fix and create more disruption than they would have at the beginning.
Readiness work is about understanding your actual starting point. Which processes are clean enough to automate and which ones will break if you try? Which roles have the capacity to absorb new tools and which ones are already stretched? Where is the appetite for change and where is the resistance concentrated? You cannot answer those questions by looking at a product spec sheet. You answer them by talking to your people and being honest about what you find.
The conversation we had this morning is not a soft warmup before the real work. It is the foundation that makes the real work possible. Without it, you are building on ground you have never actually tested.
What to Do Before You Buy Anything
If you are a business owner thinking about AI for your team right now, here is the most practical thing you can do before you speak to a vendor or sit through a demo.
Spend an hour mapping which roles in your business involve the most repetitive task work, the most data handling, and the most coordination between people and systems. Those are the areas AI is most likely to touch first. Then talk to the people in those roles. Not to announce that AI is coming, but to understand what they already know about it, what they are curious about, and what they are concerned about.
You will learn things from that conversation that no vendor will ever tell you. And you will have already done more than most businesses do before they sign a contract.
The technology is the easy part. The people are the hard part. Starting with the hard part is not a detour. It is the shortest path to an AI adoption that actually delivers what it promised.
Get in touch at www.xsiv.au/#form