Before you spend a dollar on AI, take this quiz.
Most practices that implement AI unsuccessfully weren’t let down by the technology. They were let down by implementing before they were ready. This chapter gives you a framework to assess your practice’s readiness honestly, before you commit time, money and staff goodwill to an implementation that may fail.
Be honest with your answers. The point is not to feel good about where your practice is. The point is to know where it is so you can address the gaps before they become expensive problems.
The Practice AI Readiness Assessment
Score each question from 0 to 3:
0 — No. This doesn’t exist in our practice.
1 — Partially. We have something but it’s incomplete or inconsistent.
2 — Mostly. We have this but it could be improved.
3 — Yes. This is documented, current and followed consistently.
Question 1 — Do you have documented processes and procedures?
Not in someone’s head. Not in an email chain from 2019. Written down, current, accessible to all staff and actually followed.
This is the single most important readiness indicator. Every AI tool you implement will need to reference your practice’s specific processes. If they’re not documented the AI has nothing to work from.
Ask yourself:
If your practice manager left tomorrow, could a new person find written instructions for every core task in the practice? If the answer is no, or not completely, your documented processes need work before any AI implementation begins.
Score 0-3: ___
Question 2 — Do you have a clear goal for AI implementation?
Not “I want to use AI” or “I want to be more efficient.” A specific, measurable goal.
Examples of clear goals:
Reduce no-show rates from 15% to under 8% within 90 days.
Eliminate after-hours phone calls by implementing AI after-hours booking.
Reduce time spent on billing exceptions from 3 hours per week to under 1 hour.
Examples of unclear goals:
Be more efficient.
Keep up with technology.
Do what everyone else is doing.
Ask yourself:
Can you write down in one sentence the specific problem you want AI to solve and how you will measure whether it has been solved?
Score 0-3: ___
Question 3 — Do you understand your current data environment?
Where does your patient data live? What practice management software are you using? What version? What customisations have been made to your standard configuration? Who manages your IT infrastructure?
AI tools need to integrate with your existing data environment. If you don’t know what that environment looks like you cannot assess whether an integration will work, or what will break when it doesn’t.
Ask yourself:
Can you describe your current software stack, practice management system, billing system, appointment system, clinical records system, and who is responsible for maintaining each?
Score 0-3: ___
Question 4 — Do you know where your patient data is stored and who has access to it?
On premise servers? Cloud based? Australian hosted? Overseas? Who has administrative access? When was access last reviewed?
This matters for two reasons. Firstly, Privacy Act obligations require you to know and control where patient data goes. Secondly, AI tools that process patient data need to integrate with your existing storage environment. If you don’t know where your data is, you cannot assess the sovereignty and privacy implications of adding an AI layer on top of it.
Ask yourself:
If a patient asked you where their health information is stored and who has access to it, could you answer accurately and completely?
Score 0-3: ___
Question 5 — Is your current software up to date and well maintained?
AI tools are built to integrate with current software versions. An outdated practice management system, an unpatched operating system or legacy hardware creates integration barriers that no AI vendor’s marketing material will mention upfront.
The practices that struggle most with AI implementation are often those running software versions that haven’t been updated in years. The integration that works perfectly on the current version of Best Practice may fail entirely on a version from four years ago.
Ask yourself:
When were your practice management system, operating system and hardware last updated? Are you running current versions or have updates been deferred?
Score 0-3: ___
Question 6 — Do you have adequate cybersecurity measures in place?
The Medibank breach happened through a stolen password and missing multi-factor authentication. AI tools expand your practice’s digital footprint, more integrations, more data flows, more potential entry points for a security breach.
Before you add AI you need to ensure the foundation is secure. Multi-factor authentication on all systems. Strong password policies. Regular staff training on phishing and social engineering. A documented response plan for a data breach.
Ask yourself:
Does every staff member use multi-factor authentication to access practice systems? When was your last cybersecurity review? Do you have a data breach response plan?
Score 0-3: ___
Question 7 — Is your team ready for change?
Technology implementation fails as often from people problems as from technology problems. A team that feels threatened by AI, hasn’t been consulted about implementation or doesn’t understand what AI will and won’t change is a team that will consciously or unconsciously resist the implementation.
Change readiness doesn’t mean everyone is enthusiastic. It means the practice has a culture where change is communicated clearly, staff concerns are heard and new ways of working are supported with training and patience.
Ask yourself:
Have you had an honest conversation with your team about AI implementation? What it will change, what it won’t change and what it means for their roles? Have you addressed the fear of job replacement directly?
Score 0-3: ___
Question 8 — Do you have budget for implementation including the hidden costs?
The software subscription is rarely the largest cost of AI implementation. The hidden costs include staff time for setup and training, integration work if your software requires customisation, ongoing maintenance and updates, productivity loss during the transition period and potentially consultant support if the implementation becomes complex.
A $200 per month AI scheduling tool may cost $3,000-$5,000 to implement properly when staff time, integration work and the transition period are factored in.
Ask yourself:
Have you budgeted for the full cost of implementation, not just the subscription fee, including staff time, integration, training and a realistic transition period?
Score 0-3: ___
Question 9 — Do you have a plan for measuring success?
You cannot evaluate what you don’t measure. Before you implement any AI tool you need a baseline, your current no-show rate, your current billing exception volume, your current time spent on documentation, your current after-hours call volume, whatever metric is relevant to the problem you’re solving.
Without a baseline you have no way of knowing whether the AI improved things, made things worse or made no difference at all. And without that knowledge you cannot make informed decisions about whether to continue, adjust or abandon the implementation.
Ask yourself:
Have you documented baseline metrics for the specific problem you’re trying to solve? Do you have a measurement plan for 30, 60 and 90 days after implementation?
Score 0-3: ___
Question 10 — Do you understand your regulatory obligations?
As Chapter 4 outlined, there is no single regulatory framework governing AI in Australian medical practices. But there are obligations that apply to every practice implementing AI tools that handle patient data.
Privacy Act obligations regarding data collection, use and overseas disclosure. TGA registration requirements if the tool influences clinical decisions. Professional indemnity insurance. Does your policy cover AI-assisted decisions? Medicare compliance, particularly for billing tools that affect MBS item selection.
Ask yourself:
Have you reviewed your regulatory obligations before implementing each AI tool? Have you checked with your professional indemnity insurer? Have you confirmed where patient data will be processed and whether that meets your Privacy Act obligations?
Score 0-3: ___
Scoring your practice:
25-30 — AI Ready
Your practice has strong foundations. You can proceed with AI implementation with confidence. Start with your highest priority problem and work through the implementation chapters that follow.
18-24 — Mostly Ready
Your practice has good foundations but gaps that need addressing before implementation. Identify which questions scored lowest and address those gaps first. You can likely begin planning your implementation while addressing the gaps in parallel.
10-17 — Partially Ready
Your practice has some foundations but significant gaps. Rushing to implement AI now risks the pitfalls outlined in Chapter 5. Focus on the lowest-scoring areas, particularly Questions 1, 2 and 3, before committing to any AI tool purchase.
0-9 — Not Yet Ready
Your practice needs foundational work before AI implementation. This is not a criticism, most practices score in this range. The chapters that follow will help you build the foundations. Start with Chapter 8 on documented processes and work forward from there.
The honest truth about this assessment:
Most practices score between 6 and 15 on their first attempt. That’s not failure. That’s an accurate picture of where most Australian medical practices currently sit in relation to AI readiness and it’s exactly why so many AI implementations in healthcare underdeliver on their promise.
The gap between where your practice is and where it needs to be is not technology. It’s foundation work. The good news is that foundation work is entirely within your control. It doesn’t require a large budget, specialist IT knowledge or external consultants. It requires time, discipline and the willingness to document what your practice actually does.