Here’s a question nobody wants to hear: Maybe your business isn’t ready for AI yet. And that’s not a criticism — it’s often the smartest position to be in.
The numbers are striking: RAND Corporation found that over 80% of AI projects fail to reach production — twice the failure rate of non-AI technology projects. In the UK specifically, nearly half of all AI pilots never scale, and businesses that abandon them have typically spent over £300,000 finding that out. At Maly, we see the same pattern in the conversations we have with SMEs every week: the businesses that rush in without solid foundations spend months trying to fix hasty implementations, while the ones that got their basics right first are the ones actually seeing results.
So let’s talk about what “ready” actually looks like.
The Unglamorous Truth About AI Readiness
AI doesn’t fix broken processes. It automates them. If your current process is inefficient, unclear, or poorly documented, AI will just help you do the wrong thing faster.
Think of it like this: if your filing system is a mess of random folders with inconsistent naming, giving you an AI-powered search tool doesn’t solve your problem. You’ll just be searching through mess more quickly.
This matters even more if you’re in a regulated sector. In pharma, financial services, or any environment where compliance and auditability are non-negotiable, the cost of implementing AI on shaky foundations isn’t just operational — it can be reputational or regulatory.
The Real Readiness Checklist
Before you think about AI, ask yourself these questions. Be honest — there’s no judgement here, and no prizes for stretching the truth.
1. Do You Actually Know Where Your Time Goes?
Not roughly. Not “we spend too much time on admin.” Specifically.
If I asked you right now how many hours your team spent last week on customer queries, invoicing, data entry, and meetings — could you tell me? With numbers?
Why this matters: AI solves specific time problems. If you don’t know where your time goes, you can’t know what to automate. You’ll end up automating the wrong things, or automating things that didn’t need fixing in the first place.
You’re ready when: You can list your top three time-sinks with actual hour estimates, and you know these are consistent problems, not one-off issues.
You’re not ready when: Your answer is “everything takes too long” or “we’re just not efficient enough.”
2. Are Your Processes Actually Defined?
Can a new employee follow your processes by reading a document, or do they need to shadow someone for weeks to figure out “how we do things here”?
Why this matters: AI needs clear rules and patterns. If your process is “Dave just knows how to handle this,” you can’t automate it. The knowledge is locked in Dave’s head, and AI can’t access that.
You’re ready when: Your key processes are documented, even if it’s just bullet points in a shared document. People roughly follow the same steps each time.
You’re not ready when: Every employee does things their own way, or you regularly hear “well, it depends” when explaining how something works.
3. Is Your Data Actually Usable?
Where is your customer data? Is it in your CRM? Your email? A spreadsheet? All three, with none of them matching?
Do you have product codes, or do you just know which product is which? Are your customer names spelled consistently, or is “J Smith”, “John Smith”, and “Smith, John” all the same person?
Why this matters: AI is only as good as the data you feed it. Messy data means messy results. You’ll spend more time cleaning up AI errors than you would have spent doing the task manually. This is one of the most common problems we help businesses fix — and it’s almost always more solvable than people think, once there’s a proper system in place.
You’re ready when: You have one source of truth for key data (customers, products, orders), and it’s mostly consistent and up to date.
You’re not ready when: You’re regularly merging information from multiple sources, or you spend significant time cleaning data before you can use it.
4. Do You Have Someone to Manage This?
AI tools don’t maintain themselves. Someone needs to monitor them, refine them, fix them when they break, and train the team on them.
Why this matters: I’ve seen so many businesses buy AI tools, get excited, use them for a month, then quietly stop because nobody had time to keep them running. The tools slowly get more inaccurate, the team stops trusting them, and eventually they’re abandoned. That’s money wasted.
You’re ready when: You have someone (doesn’t need to be technical) who has capacity to spend 2–4 hours a week managing and improving the tool.
You’re not ready when: Everyone is already stretched thin, and the plan is “we’ll figure it out somehow.”
5. Can You Afford to Get It Wrong?
Not financially — though that matters too. I mean operationally. If your AI tool messes up for a day, a week, a month, what happens?
Why this matters: All AI tools make mistakes. They improve over time, but the early weeks are bumpy. If you can’t afford mistakes, you need extensive testing and human oversight, which multiplies your implementation cost and time.
You’re ready when: You can run the AI tool in parallel with your current process for a few weeks, compare results, and accept that there’ll be errors to fix.
You’re not ready when: Mistakes would damage critical customer relationships, cost you money you can’t spare, or create regulatory issues.
What Should You Do If You’re Not Ready?
If you’ve answered honestly and realised you’re not ready, what now? Getting ready for AI means improving your business in ways that have value whether you ever implement AI or not.
If your processes aren’t defined: Document them. Not to prepare for AI, but because it makes training easier, mistakes less common, and your business more resilient when people are sick or leave.
If your data is messy: Clean it up. Not for AI, but because messy data leads to poor decisions, mistakes, and wasted time right now. Moving from scattered spreadsheets to a single, well-structured database is often the single biggest operational improvement an SME can make.
If you don’t know where time goes: Track it for a month. Not to prepare for AI, but because you can’t improve what you don’t measure.
See the pattern? Getting “ready for AI” is really just getting your business basics right. These improvements pay dividends immediately — and if you do decide to adopt AI down the line, you’ll be starting from a position of strength rather than playing catch-up.
When You Are Ready: Start Small
If you’ve gone through this checklist and you’re genuinely ready, here’s my advice: start smaller than you think you should.
Don’t try to automate your entire order management process on day one. Start with automatically flagging overdue orders for follow-up.
Don’t try to replace your entire reporting workflow. Use AI to pull together the first draft of your weekly summary from a clean, reliable data source.
Pick one small, specific task. Get it working properly. Then, and only then, think about expanding.
The Real Timeline
Getting ready for AI might take three months. It might take a year. That feels like forever when everyone around you is talking about the AI revolution.
But here’s the thing: the businesses that rushed in without being ready are spending that same year — or more — trying to fix their hasty implementations, dealing with poor results, and sometimes writing off the whole thing as “AI doesn’t work for us.”
The businesses that took the time to get their foundations right first are spending a few weeks implementing something that actually works.
Which would you rather be?
Not sure where your business sits on this journey?
This is exactly the kind of conversation we have with SMEs every week — not a sales pitch, just a practical look at where you are, what’s holding you back, and what a sensible next step looks like.
If sorting out the foundations — your data, your processes, your systems — is where you’re stuck, that’s the work we do every day. We’re a Suffolk-based team with deep experience in regulated sectors, and we’ve helped businesses across the UK move from spreadsheet chaos to systems they can actually rely on.
Happy to chat if it’s useful. No obligation.
Where is your business on this readiness journey? Are you racing ahead, taking it slow, or still working on the foundations? I’d genuinely love to hear — there’s no right answer, and we’re all at different stages.