What to Sort Out Before You Touch an AI Tool
What to Sort Out Before You Touch an AI Tool
If you’re weighing up whether it’s finally time to bring AI into your business, the honest answer is probably not yet, not in the way you might be picturing it. Not because AI isn’t useful, but because most of what makes it useful somewhere else depends on groundwork most small businesses haven’t had the chance to get to yet.
Why AI readiness matters more than the tool you pick
It’s tempting to treat this as a shopping decision: which tool, which subscription, which integration. But AI doesn’t fix a messy process, it just runs the mess faster. If the underlying information is scattered, undocumented, or living in someone’s head, adding AI on top doesn’t simplify anything. It adds a new layer of complexity to a foundation that wasn’t ready to support it.
AI readiness: the state of a business’s operations, documentation and decision-making being clear and consistent enough that AI tools can actually work with them reliably, rather than needing constant correction.
The three things worth checking before you pay for anything
1. Where does information actually live? If the honest answer is “in my head, a few group chats, and one spreadsheet nobody else opens,” that’s the first thing to fix. AI can only work with what it can see. If your pricing, your process steps, or your client history exist only in memory, there’s nothing for a tool to draw on yet.
2. Is there a process, or just a habit? A habit lives in your head and shifts depending on your mood, your memory, or how busy the day is. A process is written down the same way every time, so someone else, or a tool, can follow it without guessing. AI needs the second one. Habits are the most common thing we find when we start an Operations Health Check Audit with a new client, and they’re usually the first thing worth turning into an actual process.
3. What could actually be handed over? Not everything should be automated, and that’s fine. Start with the repetitive, low-risk work: the questions that have the same answer every time, the admin that doesn’tneed significant judgement. Leave the genuinely judgement-heavy decisions with a person for now.
A simple way to test whether you’re ready
Pick one task you’re considering handing to AI, and ask three questions:
Does it follow the same steps every time? If not, document the steps first, then reconsider
Is the information it needs written down somewhere, not just remembered? If not, get it written down before automating it
Would getting it wrong be annoying rather than risky? If risky, keep a person reviewing it for now
If you answered yes to all three, the task is a genuine candidate. If not, the process work comes first, not the tool.
Where to start this week
None of this requires new software. It’s mostly a tidy-up of what you’ve already got: writing down the two or three things that only live in your head, separating your real processes from your habits, and being honest about which tasks are actually low-risk enough to hand over.
Businesses that do this groundwork first tend to get far more out of AI than businesses that buy a tool and hope the mess sorts itself out afterwards. Structure creates freedom here too: the clearer your systems are before AI arrives, the less oversight it needs once it does.
If you know your systems need work before AI makes sense, that’s what our Ops Health Check Audit, and AI Readiness & Automation offers, are built for: a structured look at what’s actually ready, and what needs sorting first.