AI Audit
Part of AI Consulting
What is worth automating, and what is not?
What is an AI audit?
An AI audit is a diagnostic pass over one specific business: someone sits with the team, goes through the work task by task, and establishes where AI or automation would genuinely save time. The subject is how a week actually runs, not AI in the abstract. Jobs worth automating get separated from the ones where it would be a distraction or a risk. Each one comes back with a rough sense of effort and cost. It isn’t a pitch dressed as a diagnosis, and it isn’t a generic “AI readiness” score out of ten. It’s an ordered picture of where the hours are, written by the people who’d build whatever they recommend, which is what keeps the estimates honest.
What an audit looks at.
The work, not the technology. It starts from where the hours actually go. The tasks people repeat, the points where a job passes between two tools and stalls, the report rebuilt from nothing every Monday, the same question answered forty times a month. Ordinary things. Nobody logs them.
Each one gets the same plain test. Would automation genuinely make this faster, cheaper or more reliable, and what would it take? Some get a clear yes with obvious payback. Some get a no. The volume is too low, the data is a mess, or a human judgement sits in the middle that shouldn’t be handed over. Some get “not with AI, but a process change would fix it in an afternoon”, which is often the most valuable finding in the document.
The output isn’t a lecture on what AI can theoretically do. It’s a map of where the hours and the money would come back, in one specific business, ordered by what to do first.
What the session actually looks like.
Less formal than the word “audit” suggests. Here’s the honest version of how a couple of days goes.
In your office, or over a call
Melbourne teams get someone in the room, because watching a job being done tells you more than hearing it described. Everywhere else in Australia it runs over a call and works nearly as well. Either way it's a founder, not a consultant we sent in our place.
We talk to the people doing the work
Not just whoever booked the audit. The person rebuilding that report every Monday knows exactly where the time goes and will explain it in about ninety seconds, properly asked. Half the value of an audit is that somebody finally asked them.
Nobody has to tidy up first
No preparation, no documentation, no making the process look better than it is. The spreadsheet nobody maintains and everybody depends on is exactly what we want to see. Workarounds are findings, and the messy version is the true one.
We say the unhelpful thing out loud
A process that isn't worth automating gets said in the room rather than padded into the report. Same when the real problem is two teams disagreeing about who owns a step, which no software has ever fixed. Selling a build for that would be taking money under false pretences.
Something short arrives days later
Three to five pages, written like a person wrote it, ordered by what to do first, with a rough cost against each line. Short enough that the people who need to read it actually will.
From audit to plan
An audit says what’s there. It doesn’t, by itself, say what order to tackle it in. Some businesses find the obvious wins clear enough to start on immediately, either with us building or with the findings handed to someone in-house. Others want the sequence worked out properly first, which is an AI strategy: a roadmap that orders the opportunities so each one sets up the next. No obligation is attached either way. The audit’s job is a straight answer, and what happens next is genuinely the client’s call. Anyone who’d rather size the opportunity up alone first can work through our AI readiness checklist, which covers the same ground for nothing.
Only book one if you want the real answer.
Worth saying before you pay for an audit. It’s only worth the money if you’re willing to hear that most of the list isn’t worth doing. A good share of what turns up is better solved by a setting nobody ever turned on, or a tool already sitting in the stack unused. We’d rather write that down than pad the list to justify the fee.
AI audit FAQs
How the business actually spends its time. The repetitive tasks, the manual handoffs, the places work piles up or gets re-done. Every process and tool gets the same question: would AI or automation genuinely give hours back here, and what would it cost to find out? It’s a look at real work, not a questionnaire about “AI maturity”.
An audit is the diagnosis: where the opportunities and the dead ends are. A strategy is the plan that comes out of it. Plenty of businesses take the audit first, then decide whether they want the roadmap or would rather act on the obvious wins themselves. One says what’s there. The other says what to do about it.
No, it’s scoped and paid as its own piece, and the findings stand on their own. Sometimes the honest answer is an off-the-shelf tool, a process change, or nothing yet. We’d rather give you a straight read and earn the build later than dress up a sales pitch as a diagnosis. You can act on the audit with us, with someone else, or in-house.
No. Don’t build us a process map, don’t tidy anything up, and please don’t get the team to write documentation for it. We’d honestly rather see the messy version, because the workarounds people have invented are usually where the best findings are hiding. Turning up and talking us through a normal week is enough.
Usually a session or two to look at the work, plus a short write-up, days, not weeks. It’s meant to be a fast, cheap way to get a straight answer, not a drawn-out engagement. A long audit tends to cost more than the first few wins it finds.
Three to five pages. An ordered read on where AI would help, what each item would roughly cost and take, and what to ignore. Enough to make a decision on, without wading through sixty pages nobody finishes. It arrives as a short written summary you can share internally.
One of us, Aidan or Caleb depending on whether the work is operational or marketing-shaped, and often both if it spans the two. Whoever turns up is the person who’d build whatever comes out of it. That’s the point. The recommendations stay grounded in what can genuinely be built and what it costs, rather than in what looks good on a slide, and nobody gets handed to an analyst who has never shipped the thing they’re recommending.
Happy to give a straight read on where AI fits
A couple of lines is enough. It goes straight to Aidan, not a sales team, and if it isn’t worth building he’ll tell you.
- You deal with Aidan directly, not a sales team.
- A straight answer on whether it's worth building, and what it'd cost.
- You own everything we build. No lock-in and no per-seat fees.
Thanks, we've got it.
It's gone to Aidan, not a shared inbox or a sales team. If it's urgent, call 0415 525 006.