Most AI projects don’t fail because the technology was bad. They fail because a small team jumped to “which tool?” before answering “are we ready, and ready for what?” These questions are the check-up worth doing first. Answer them honestly with your team — the gaps you find are usually more useful than the answers you’re sure of.
Use this as a self-assessment. Read each question, and for each one give yourself a plain answer: yes, clearly / sort of / no, not yet. The “sort of” and “no” answers are where the real work is. You don’t need every box checked before you start — but you should know which boxes are empty and choose to move anyway, with eyes open.
People & buy-in
AI adoption is a people project wearing a technology costume. If your team isn’t with you, the best tool in the world will sit unused.
- Can we name one real problem our people feel every week — the admin that eats a clinician’s afternoon, the inbox that never empties — that AI could realistically ease?
- Who on the team is genuinely curious about this, and who is worried? Have we actually asked, or are we guessing?
- Have we been honest that the goal is to give people their time back, not to reduce headcount? Would our team believe us if we said it?
- Is there one person who will own this — not build it, but champion it and answer questions when the rest of the team gets stuck?
- Do we have even an hour of time to give people to learn, or are we expecting adoption to happen in the cracks of an already-full day?
Data & privacy
Your data is the raw material and the risk. For a dental practice, a nonprofit holding donor records, or a physiotherapy clinic with patient files, this section isn’t optional.
- What information would this tool touch — and does any of it include patient, client, or donor details we’re legally or ethically bound to protect?
- Do we know where that data would go? Would it leave our systems, and if so, to whom, under what agreement?
- If a patient, client, or donor asked us plainly “is my information safe with this?”, could we give an honest, specific answer?
- Are we clear on our obligations — HIPAA for a medical office, donor-privacy commitments for a nonprofit, basic customer trust for a retail shop — and does this tool respect them?
- Do we have a rule for what our people should never paste into an AI tool, and does everyone actually know it?
Workflows & friction
AI helps most when it slots into how work already happens. If using it means ten new steps, it won’t stick.
- Have we mapped the actual workflow we want to improve — the real one, not the tidy version in the manual?
- Where does the friction actually live: the blank page, the repetitive typing, the digging for information, the waiting on a hand-off?
- Would this tool remove steps for our people, or just add a new one alongside everything they already do?
- Can we start with one narrow, low-risk task — drafting appointment reminders, summarizing a meeting, sorting an inbox — instead of rewiring everything at once?
- What happens on a day the tool is down or wrong? Do our people still know how to do the work the old way?
Guardrails
Trust is easy to lose and slow to rebuild. Guardrails are how you keep AI useful without letting it cause quiet harm.
- For anything the tool produces that reaches a patient, customer, or donor, is there a person who reviews it before it goes out?
- Do we know where a human must stay in the loop, and where it’s genuinely safe to let the tool run on its own?
- When the AI is uncertain or out of its depth, does it hand off to a person — or does it guess and hope?
- Can we explain, in plain language, how a given output was reached — or is it a black box we’re asked to trust blindly?
- Who is accountable when the tool gets something wrong? If the honest answer is “no one, really,” that’s the gap to close first.
Success measures
If you can’t say what “working” looks like, you won’t know whether to keep going. Decide before you start.
- What specifically would be better in ninety days — hours saved, faster responses, fewer errors, a team that feels less buried?
- How will we actually know? Is there a simple, honest way to measure it, even roughly?
- Are we measuring the thing that matters to our people and the people we serve — or just what’s easy to count?
- What would tell us to stop or change course? A frank “this isn’t worth it” needs to be an allowed answer.
- If it works, do we know what we’d try next — and if it doesn’t, have we protected ourselves from a costly, hard-to-undo commitment?
What to do with your answers
Look at where the “no, not yet” answers cluster. If they’re mostly in People & buy-in, slow down and bring your team in before you touch a tool. If they’re in Data & privacy, get that settled first — no efficiency gain is worth a breach of trust. If your gaps are in Workflows or Success measures, you’re likely closer than you think; you just need to narrow your first project and define what winning looks like.
A family business, a small clinic, and a two-person nonprofit will all answer these differently — and that’s the point. Readiness isn’t a score to beat. It’s a clearer picture of where you stand, so you can move deliberately instead of hopefully. The teams that do best with AI aren’t the ones with the most technology. They’re the ones who asked the honest questions first.
If you’d like to talk through your answers with someone who does this every day — no jargon, no hard sell — book a free discovery call. We’ll look at where you’re ready, where you’re not, and what a sensible first step would actually be.