How to Tell Whether Your Practice Is Ready for AI
Whether a practice is ready for AI is rarely a question about technology. It is a question about documents, volume and who currently does the work. Get those three straight and the decision makes itself.
Here are the signals we look for, in both directions.
Signal one: your work arrives as documents
Referrals, invoices, contracts, statements, notes, plans. If the core of the job is reading something that arrives in a predictable-ish format and turning it into structured data, an agent can help.
If the work is mostly conversation, judgement or physical presence, an agent helps with the admin around it and little else.
Signal two: the volume is steady
Steady means every week, not every quarter. An agent improves with repetition — the corrections you make in week one teach it your conventions. Work that arrives in a seasonal spike never gets that far.
Signal three: someone owns the process today
If nobody can describe how an invoice gets coded or how a referral becomes a file, there is nothing to automate. Automation captures a process; it cannot invent one.
Signal four: the systems have a real API
Your practice software needs to accept data, not just display it. Where there is no usable API, the agent can prepare the data and a person still re-keys it — worth something, rarely worth a retainer.
Signal five: you are willing to review for a month
The first month is a corrections month. Practices that expect perfection on day one abandon the tool before the gain arrives.
The five signals you are not ready
- The ledger or file system is already a mess that nobody trusts
- Client data lives in scanned PDFs that are photographed badly
- Volume is under about 100 transactions or a dozen matters a month
- Nobody has authority to change the process if the agent finds a better one
- The driving motive is a competitor announced something
None of those are permanent. They are sequencing problems, and sequencing is fixable. The detail behind each is on the automation overview, and the twenty agents themselves are listed under specialised AI agents.
What readiness actually looks like
A practice that is ready can answer three questions without looking anything up: what arrives each week, who touches it, and how long it takes. If those answers are vague, the first project is not an agent — it is a process map.
That is a cheaper piece of work and it makes everything after it faster.
What we build
We build automation and digital employees for accounting, legal, medical and NDIS practices. If you want an honest read on whether your practice is one of the twenty that suits an agent, tell us what arrives each week and we will say so.
Start with one process
Tell us where the hours actually go and we will map it against what an agent can and cannot take. One process, measured before it is trusted. The first conversation is free and usually short.