AI Is Not Your Problem. Your Processes Are.

The Call I Keep Getting

"Craig, we need an AI strategy. Everyone's talking about it. Our competitors are doing it. We don't want to get left behind."

I understand the feeling. Every vendor is adding AI features. Every conference talk mentions AI. Every LinkedIn post is about the transformative power of machine learning. The pressure to do something — anything — is real.

But here's the awkward truth I have to deliver in almost every one of those conversations: AI is not your problem. Your processes are.

And if you invest in AI before you fix your processes, you won't transform your business. You'll just automate your dysfunction at scale.

What AI Actually Needs

AI is not magic. It is a tool that requires three things to work:

1. Clean, structured data. AI models are only as good as the data you feed them. If your data is inconsistent, duplicated, or incomplete, AI will magnify those problems.
2. Well-defined processes. AI works best when applied to a specific, well-understood operational problem. If you can't describe the process clearly, you can't apply AI to it effectively.
3. Integrated systems. AI needs to draw data from multiple sources. If your systems don't talk to each other, your AI is working with half the picture.

Now ask yourself honestly: does your business have these three things?

Most businesses in this sector don't. Their data lives in multiple places with inconsistent formats. Their processes are undocumented and vary by person. Their systems operate in isolation.

Throwing AI at that situation doesn't help. It makes it worse.

The Real Opportunity

Here's what I tell every business that calls about AI:

Fix your operations first. Then evaluate AI.

The businesses that will benefit from AI are not the ones that buy it first. They're the ones that have clean data, documented processes, and integrated systems — because they can deploy AI on top of a stable foundation and get predictable results.

Meanwhile, the businesses that chase AI without fixing their foundation will spend a lot of money on tools they can't use effectively, blame the technology, and conclude that AI doesn't work.

Where AI Actually Helps in This Sector

I'm not anti-AI. There are genuine applications in this sector. But they're specific, not transformative:

• Artwork proofing: AI can flag common artwork issues before they reach production
• Order routing: AI can learn which orders go to which production method and suggest routing
• Demand forecasting: AI can analyse historical order patterns to inform stock purchasing
• Customer service triage: AI can categorise and route customer enquiries based on content

Notice a pattern? Every single one of these requires clean data, a defined process, and integrated systems to work.

What You Should Do Instead

Before you buy an AI tool, here's a six-month plan that will deliver more value:

1. Audit your operations. Understand what's actually happening in your business. Where are the bottlenecks, the workarounds, the data quality issues?
2. Clean your data. Deduplicate, standardise, and document your data. This is unglamorous work that pays for itself many times over.
3. Document your processes. Write down how key workflows actually work. Not how you think they work. How they actually work.
4. Integrate your systems. Close the gaps between your ERP, eCommerce, accounts, and production systems.
5. Then — and only then — look at AI. With clean data, documented processes, and integrated systems, you can evaluate AI on its actual merits rather than on hype.

This is what a Clarity Audit covers. Six areas of your business assessed against what good looks like. Every finding quantified. Every recommendation costed. And an honest assessment of whether AI is something you should be thinking about now or later.