AI tools are now built into almost everything a business already uses, and new ones appear every week. Many promise to save hours, cut costs or replace whole tasks. Some deliver. Many end up as another subscription that a few people try once and then forget.
The difference rarely comes down to the tool itself. It comes down to whether the business asked the right questions before buying. Here are five worth asking. They are not the whole picture, but if a proposed tool cannot get through all five, the problem is usually not the AI.
1. What problem are we solving, in numbers?
“We should be using AI” is not a problem. “Answering routine customer emails takes two people most of every morning” is. If you cannot describe the problem and roughly what it costs today, you will not be able to tell whether the tool helped.
2. Is the process clear enough to hand over?
AI is good at doing a well-understood task faster and more consistently. It is much less good at fixing a task that nobody can describe in the same way twice. If your team does the same job three different ways, the first step is agreeing one way, not automating all three.
3. Where does our data go, and who can see it?
Every AI tool needs data to work with, and that often includes customer details, financial figures or internal documents. Before buying, you need clear answers about where that data is processed and stored, whether it is used to train the provider’s models, and how this fits your obligations under UK data protection law. A tool that cannot answer these questions plainly is a risk, however clever it is.
4. Who checks the output?
AI can be confidently wrong. That is manageable when someone reviews the output before it reaches a customer, a contract or a set of accounts. It is not manageable when nobody does. For every use, decide who is responsible for checking the result, and how much of their time that will take. Sometimes the checking takes longer than the task did.
5. What does success look like in 90 days?
Agree a simple measure before you start: time saved, response times, error rates, or something else that matters to you. Then check it. A short, honest trial on one process tells you far more than a long rollout across the whole business, and it makes the decision about what to do next much easier.
The question behind the questions
Underneath all five is a simpler one: is this the right problem to solve first? Sometimes the answer is an AI tool. Sometimes it is a small process change, a better connection between two existing systems, or a feature you already pay for and do not use.
We do not earn commission on any software we recommend, so our only interest is the answer that works for your business. If you are weighing up an AI purchase, a short conversation with our AI consultancy team before you sign is usually worth having.
