AI consulting in Newcastle, NSW: work out where AI actually helps your business.

Looking at how a business actually works, finding the jobs that are repetitive and written down, and deciding which of them are worth automating and in what order.

The work itself is plain. Somebody sits with the people doing the job, writes down what happens, and looks for the parts that happen the same way every time.

What it covers

• How the business actually works. Not the diagram. What people really do, including the spreadsheet nobody mentions.
• Which jobs repeat. The ones with the same steps every time are the candidates. The ones needing a judgement call are not.
• What it would take. For each candidate: what has to exist, what it touches, and what happens when it gets something wrong.
• The order. Which one first, and which ones are not worth doing at all. This is the part that earns the fee.

The output is a short document naming what to do, in order, with the ones to leave alone named as well. Building it is a separate job and it is the easier half.

The tools are cheap and plentiful. What is scarce is knowing which job in your particular business is worth handing to one, and that answer does not come out of a product demonstration.

Every small business owner in Newcastle is being sold something with AI in the name. Most of them are perfectly good products, and most of them are being bought before anybody has asked what job they are for.

What follows is a familiar pattern: a subscription, a fortnight of enthusiasm, and then nobody uses it because it did not fit how the work actually happens.

The deciding is the part that stops that, and it is cheaper than the tools. It is also the part nobody sells, because there is no subscription attached to it.

It will not make a judgement call you would want to stand behind, it will not fix a process nobody has written down, and it will not know anything about your business that you have not told it.

Being clear about the ceiling is the part of this conversation that gets skipped, and it is what makes the rest of it usable.

The limits worth knowing

• It does not have judgement. It will produce a confident answer to a question that needed somebody to think. That is the failure mode, and it is why the limits get written before anything goes live.
• It cannot fix an undefined process. If three people do the same job three ways, there is nothing consistent to hand over. The writing down is the work.
• It knows only what you gave it. Your prices, your area, your rules. Nothing else, and it should say so rather than guess.
• It does not remove the responsibility. If it sends something wrong to a customer, that is your business speaking.

The federal government's AI ethics principles put accountability and transparency at the centre for the same reason, and they are short enough to read in one sitting.

The job has to be written down, the facts it needs have to exist somewhere, and somebody has to own reading what it does. Missing any of those, building is early.

Almost every automation that quietly fails was built before one of these was true, and each of them is a job for the business rather than for us.

What has to exist

• The job, written down. The steps, in order, as they actually happen. If it cannot be written, it cannot be handed over.
• The facts it answers from. Prices, hours, area, rules. Written once, kept somewhere it can be changed the same day.
• The limits. What it must never say or decide. Written before it goes live, not after a complaint.
• Somebody who reads what it did. For the first while, every message. This is the difference between a thing that improves and a thing that drifts.

Sit with the work, list the repeated jobs, cost them honestly, pick one, write its facts and limits, then build the smallest version and read everything it does.

1. Sit with the work

With the people who do it, not with the owner's description of it. The gap between those two is where most of the findings are.

2. List the repeated jobs

Everything that happens the same way every time. Include the small ones, because frequency matters more than size.

3. Cost them honestly

How often, how long, and what happens when it is missed. This is what turns a list into an order.

4. Pick one

The one with the best ratio of frequency to difficulty. One done properly beats five started.

5. Write its facts and its limits

What it answers from, and what it must never say. An hour, and it decides whether customers like the result.

6. Build the smallest version and read it

Then read everything it does for a fortnight. Then, and only then, the next one.

The looking is short. Writing down how a job actually happens takes longer than anybody expects, and it is the part that determines whether the building works.

The order it happens in

• The work gets looked at. With the people doing it. A short piece of work and it produces the list.
• One job gets written down properly. Steps, facts and limits. This is the part that stretches, and it belongs to the business.
• The smallest version gets built and read. Then corrected. Then the next job, which is always faster than the first.

A business that wants something no tool currently does is asking for software to be built, which is a different job with a different timeline. It is worth naming as one rather than discovering it.

Ask what they will look at before recommending anything, whether they sell the tools they recommend, who owns what gets built, and what they would tell you not to do.

We do this work in Newcastle, so we are not neutral. These are the questions we would want asked of us.

• What will you look at first? How the business actually works. An answer that starts with a product has skipped the job.
• Do you sell the tools you recommend? Worth knowing. We build on tools we do not own, and we say when the honest answer is that you need nothing.
• Who owns what gets built? The business, in accounts in its own name, including anything holding customer details.
• What would you tell me not to do? A consultant with nothing on that list has not looked properly.
• Am I locked in? Ours runs month to month and always has. Any arrangement you cannot leave at the end of a month is worth a second look.

Plenty of businesses go through this and find one small thing worth doing. That is a good outcome and it is worth saying, because the alternative is a project nobody needed.