“We should probably do something with AI.”

“We should probably do something with AI.”

I hear a version of this almost every week, and it is usually said with the same tired tone people use for flossing more or finally organising the garage. Something you know you should do, but nobody is sure where to start, and the longer it sits there unstarted, the more it feels like a bigger project than it actually is.

Here is my honest answer, the one I give before any pitch: most businesses do not need an AI strategy. They need one specific, boring task automated, the one eating ten hours a week from someone on staff who could be doing something better with their time. That is a smaller ask than “transform the business with AI,” and it is also the version that actually gets built and used, instead of demoed once in a meeting and quietly abandoned three weeks later.

Why the abandoned pilot happens

I have seen the abandoned pilot more than once, at more than one company. It follows a pattern: someone champions a proof of concept, it works great in a demo because demos are staged to work, and it never makes it past that stage into daily use. Not because the technology failed. Almost always because nobody scoped it down to a single, real, already-painful problem first. Instead it tried to solve “customer service” or “operations” in one shot, which is not a task, it is a department, and no single pilot survives being asked to fix a department.

A scoped pilot looks different. It has a name: invoice data entry, lead qualification calls, resume tailoring, appointment scheduling. It has a number attached: ten hours a week, or four thousand dollars a month in a subscription nobody is using fully. And it has one person who feels that cost directly enough to say yes to fixing it without a six-month committee process.

A smaller question than “should we do AI”

So before “should my small business use AI,” I would ask a smaller question: what is the one task, this week, that is costing you the most hours or the most money? Not the most interesting task. Not the one that would look best in a pitch deck. The one your team already complains about, the one where someone says “I spend half my Friday doing this by hand” and everyone nods because they have heard it before.

Start there. Automate that one thing, watch whether it actually gets used a month later, and only then decide whether there is a second thing worth doing. Everything else, the strategy deck, the multi-phase roadmap, the platform decision, can wait until you have one working example to build on instead of a slide with arrows on it.

Sometimes the answer is no, and that is fine

To be fair, sometimes the honest answer really is “nothing yet,” and that is a perfectly good answer too. Not every business has a task that clears the bar, and forcing an AI project into a business that does not need one wastes money on a solution looking for a problem. I would rather tell you that in a thirty-minute conversation than sell you a project you do not need.

If you have got a specific task you have been meaning to automate, or you are not sure whether you have one yet, that is exactly the conversation worth having. Find me through the contact section and we will figure out, honestly, whether there is something here.