An AI process audit tells you which three manual processes are worth piloting first, and why. It lists the work your teams do by hand, walks the busiest processes with the people who do them, and scores each on volume, judgement, data access, the cost of a wrong answer and ownership. The top three go forward with a measured baseline.
Everything else is parked with a written reason, so the same ideas do not come back every quarter.
This guide is for the operations manager, CFO or IT manager in a mid-sized Australian or New Zealand organisation who has been told the business needs to be "doing AI" and has to say what that means. The usual trigger is a board mandate with no named process behind it, or a new AI role with no first project to hand over.
Key takeaways
- Pick the first AI process from the work itself, not from a tool, a demo or the loudest request.
- Score every candidate on the same five criteria: volume, judgement, data access, cost of a wrong answer and ownership.
- Pure-rules processes belong in workflow automation, and processes with unreachable data need integration work first.
- Record a baseline for each of the three shortlisted processes before any AI runs.
- An audit for one business unit usually fits inside four weeks.
Why is the first AI process usually picked badly?
Australian organisations are already using AI. The National AI Centre's AI Adoption Tracker reported that 43% of businesses it surveyed had some level of AI adoption between December 2025 and February 2026. Most of that is people using general tools. Far less of it is a business process that now runs differently.
The gap opens at the first decision. Three ways of choosing tend to go wrong:
- Starting from the tool. Copilot or ChatGPT licences are bought, and the question becomes "what can we use this for?" The answers are the tasks the tool demos well, which are rarely the ones holding the business back.
- Starting from the loudest request. The executive who asked first gets their process done first, whether or not it has the data or volume behind it.
- Starting from the best demo. A vendor shows an agent handling a task end to end, and the business hunts for a process that looks like the demo.
Each one skips the same step: looking at the work. Gartner predicted in June 2025 that over 40% of agentic AI projects will be cancelled by the end of 2027, citing escalating costs, unclear business value and inadequate risk controls. Unclear value is exactly what a badly chosen first process produces.
How does an AI process audit run?
The audit is short and deliberately unglamorous. Most of the time goes on watching people work and reading the systems they work in.
- Long list collectedEvery manual process managers and teams can name
- Busiest walkedSit with the people who do the work, item by item
- Data checkedCan the inputs be reached by API, database or export?
- ScoredFive criteria, same scale for every process
- Owners confirmedA named manager agrees to own each shortlisted result
- Shortlist with baselinesThree processes, each with its starting numbers
Collect the long list
Ask every manager one question: which work in your area is done by hand, queues behind one or two people, or gets retyped from one system into another? Accept everything. A long list of several dozen candidates is common in a multi-site organisation, and nothing is scored yet.
Walk the busiest processes
Pick the ten or so with the most volume or the loudest complaints and sit with the people who do them. Watch real items, not the procedure document. The procedure says "check the invoice against the PO". The person doing it also checks the supplier's ABN, looks up a spreadsheet only they maintain and rings the site manager twice a week. Those extra steps are where the judgement lives, and they decide whether AI can help.
Check the data
For each walked process, find out where the inputs sit and whether they can be reached. An inbox, a SharePoint library, a line-of-business database with an API and a scanner in the corner are very different starting points. A process whose inputs only exist on paper or in one person's head may need a different first step entirely.
What are the five scores in an AI process audit?
Score every walked process on the same five criteria, one to five each. Write the definitions down so two people scoring the same process land close together.
| Criterion | Scores 1 | Scores 5 |
|---|---|---|
| Volume and backlog | A few items a month, no queue | Hundreds a week, a visible backlog |
| Judgement per item | Pure rules, or a decision only a senior person can make | Judgement a trained person makes from patterns in past items |
| Data reachable | Paper only, or locked in a system with no export | Digital inputs reachable by API, database or export |
| Cost of a wrong answer | Wrong output reaches a customer or regulator before anyone sees it | A person checks every output before it goes anywhere |
| Owner | Nobody will own the result | A named manager wants it and will measure it |
Two of these read the opposite way to what people expect.
The judgement scale is not a straight line from easy to hard. A process that is pure rules belongs in Power Automate or a workflow tool, where it runs the same way every time without AI. A decision that needs a senior person's full context is a poor first pilot. The sweet spot is the judgement a trained team member makes from patterns: which cost code, which category, which priority. Our guide to Power Automate vs AI agents covers where that line falls.
Cost of a wrong answer is scored on the controls around the process. High-stakes work with a person approving every output can score well. The same work with no review step scores badly until one is designed in.
What does a worked example look like?
For example, consider a 250-staff professional services firm whose long list has eight candidates after the walks. The scores below are illustrative, to show how the method sorts them.
| Process | Volume | Judgement | Data | Error cost | Owner | Total |
|---|---|---|---|---|---|---|
| Supplier invoice coding to matters | 4 | 5 | 4 | 4 | 5 | 22 |
| Timesheet narrative clean-up before billing | 5 | 4 | 5 | 4 | 3 | 21 |
| Client intake and conflict check triage | 4 | 4 | 4 | 4 | 4 | 20 |
| Staff leave approvals | 4 | 1 | 5 | 4 | 3 | 17 |
| Board paper summaries | 2 | 3 | 4 | 3 | 4 | 16 |
| Tender response drafting | 2 | 3 | 3 | 3 | 5 | 16 |
| Archive file classification | 3 | 4 | 2 | 5 | 1 | 15 |
| Partner strategy memo drafting | 1 | 1 | 3 | 2 | 4 | 11 |
The top three are invoice coding, timesheet clean-up and intake triage. Each has a named owner and digital inputs.
Leave approvals scored well on volume and data but low on judgement. That is a workflow job, and it goes to whoever runs Power Automate. Archive classification scored well on judgement but has no owner and paper-heavy inputs, so it is parked with a note. When someone raises it next year, the reason it waited is already written down.
What happens to the processes that miss?
Most processes fail at least one gate. The audit should say what happens to each one, so that "not now" has a reason attached.
Which gate did this process fail?
- None, and it is in the top threeShortlist for a pilot
- Judgement scored 1 because it is pure rulesWorkflow automation, no AI
- Data cannot be reachedIntegration work first, then re-score
- No owner will take itPark with a written reason
- Wrong answers would reach customers uncheckedDesign a review step, then re-score
The data gate is the one most often missed. A process that scores well everywhere except data access is usually an integration project with an AI label on it. Getting the inputs out of the source system comes first, and our system integration work often starts there.
What should the audit produce?
An audit that ends in a slide deck of "opportunities" has not finished. The output should be specific enough that a pilot can start the following week.
| Opportunity deck | Useful audit | |
|---|---|---|
| Candidates | 20 use cases, unranked | Three ranked processes, the rest parked with reasons |
| Evidence | Workshop sticky notes | Walked items, data sources named, scores recorded |
| Starting point | Estimated benefit | Measured baseline per process |
| Ownership | "The business" | A named manager per process |
| Next step | A strategy phase | A scoped pilot with an accuracy threshold |
For each of the three, record the baseline before any AI runs. These are the numbers the pilot will be judged against.
- Items per week, and how many are waiting nowYour count
- Minutes per item, timed on real workYour timing
- Items reworked or corrected laterYour rework rate
- People who can do the hard itemsYour key-person count
- Accuracy threshold the owner will acceptSet by the owner
The accuracy threshold is the owner's call. It is the share of AI outputs that must match what their own team would have done. Until the AI reaches it, the AI runs alongside the team and nothing goes out. After that, a person still approves every output before it reaches a customer or system of record.
How long does an AI process audit take, and who should be involved?
An audit for one business unit or function usually fits inside four weeks. An organisation with several sites or divisions may run one per division, so each shortlist reflects how that division actually works. "Every branch does it differently" is a finding in its own right.
- Week 1Long listShort interviews with managers, every manual process captured
- Week 2Walk the busiestSit with the teams, time real items, find the data
- Week 3Score and test the gatesFive scores per process, data access confirmed with IT
- Week 4Shortlist and baselinesOwners confirmed, baselines measured, pilot scope written
Three groups need time in it. The people who do the work, because they know the steps the procedure leaves out. IT, because data access decides more scores than anything else. And the managers who will own the results, because a process with no owner does not survive its pilot.
Keep the audit separate from tool selection. If the business has already bought Copilot licences, that is fine, and the audit may find Copilot covers one of the three. It should not start from the licences. Where staff already use AI tools informally, run a shadow AI inventory alongside it, because what people use unofficially is good evidence of where the manual load sits.
How does a process audit fit an AI strategy engagement?
A process audit is the practical core of most AI strategy work. A broader AI strategy consulting engagement adds governance, risk and the operating model around it. The audit keeps that strategy tied to named processes and measured baselines.
Once the three are chosen, each goes through a pilot with its own exit point: build, run alongside the team against the baseline, then decide on the numbers whether to proceed. Our guide to moving an AI pilot to production covers that decision. For a finance example carried through, see automated invoice processing in Business Central.
Across our consultants' careers, including operations and data work in BHP, Rio Tinto and Senex Energy environments, one habit has carried over: watch the process being done before deciding what to build. When a shortlisted process needs something built, that work runs through our custom AI development service, on your infrastructure in Australia or New Zealand.
How Solve8 can help
Our consultants run the process audit as the first phase of an AI strategy engagement, on a fixed fee per phase agreed before we start, and the shortlisted processes carry straight into the build. If your board wants an AI plan and nobody has named a process, talk to us and bring the three your managers complain about most.
