How do I know if AI is worth it for my business, or just hype?

What is genuinely overhyped?

Three things, consistently.

Anything sold as transformation without naming a process. If a proposal cannot tell you what changes on a Tuesday morning, it is a strategy document, and strategy documents do not save anybody any time.

Customer-facing chatbots as a first move. They are the most visible option and one of the least valuable, because they sit at the end of the process rather than fixing whatever generated the contact. Look at why people are getting in touch before automating the getting in touch.

Tool-first buying. Choosing the software before mapping the problem is how businesses end up with three subscriptions and no change in how the work flows.

What is genuinely working?

The unglamorous middle of the operation, where volume is high and judgement is low.

Reading and routing incoming work. Moving information between systems that do not talk. Drafting the first version of something a person then owns and improves. Summarising after the fact so nobody has to write it up. Checking documents against rules.

None of that will impress anyone at a conference. All of it removes real hours, and the hours are the point.

How do I find out cheaply, without hiring anybody?

Pick your most irritating process and count it.

How many people touch it, how many times a week, how long each time. Multiply it out into an annual cost. That number is the only thing that tells you whether any of this is worth pursuing in your business, and it takes an afternoon to produce.

Then ask whether the decisions inside that process could be written down. If they could, you have a candidate. If the answer is "it depends, you have to know the client", you have judgement work, and that is not what this is for.

If you would rather have a structured version, the Quick Check at the Quick Check page takes two minutes, gives you an indicative score, and does not require a conversation with anybody.

Is our business too small?

Almost certainly not, and small businesses often move faster because there are fewer systems in the way.

The test is repetition, not headcount. Three people doing the same rules-based task every week is a stronger case than thirty people doing genuinely varied work. Some of the clearest results I have seen have been in businesses under twenty people.

So when is the honest answer no?

When your bottleneck is not capacity, when your processes are unstable, or when you would be doing it because everyone else is.

If work is not flowing because of a people problem, a pricing problem or a demand problem, AI will not touch it. If the process is about to change, automating the current one wastes the money. And if the driver is a board asking what the AI strategy is, the right answer might be a considered "not yet", which is a legitimate position and one I have written for clients before.

A fair proportion of the businesses I speak to are not ready. Knowing that early is worth more than a project that stalls at month four.

The most useful next step

A free discovery call at our booking page is forty-five minutes on one process, not on my services. You leave with that process mapped, a straight answer on whether AI would help it, and a next step you could act on yourself without hiring anybody. Quite a lot of them end with me saying fix the process first.

our What to Expect page sets out how engagements actually run, including where they get uncomfortable. our guide on whether you need AI covers this ground in more depth.

Frequently Asked Questions

Your next step

Take the free two-minute Quick Check for an AI readiness score, see what the £995 AI Visibility Audit covers, or book a free 45-minute discovery call. Definitions for the terms used here are in the AI glossary and the Claude glossary.