Knowledge & data
Lightweight practice knowledge base
How do referrals work again? How long does the new surface disinfectant need to stay on? Who covers for us over the holidays? In most practices, the same experienced assistant answers questions like these – in between two patients at reception.
You'll get …
- 1
Know-how that doesn't depend on one assistant
Prescription and referral routes, appointment types, hygiene standards and holiday arrangements are written down and easy to find.
- 2
Look it up, don't interrupt
New assistants get answers, with the source, instead of interrupting their experienced colleague at reception.
- 3
Deliberately light – patient data stays out
The core is process knowledge you can ask questions of. Changes to standards only go in as drafts with sign-off; the records stay in the PMS.
01Intro
The light practice knowledge base makes organisational processes and standards easy to find: prescriptions and referrals, appointment types and durations, hygiene checklists, equipment training, holiday arrangements, onboarding for new assistants. It's a structured system where the team looks things up instead of asking around – and where AI answers the standard questions, with the source. Deliberately "light": the value lies in process knowledge you can ask questions of, not in a fully automated build.
Worth knowing: patient data stays out – that's non-negotiable. The knowledge base holds processes and standards; the PMS holds the records. And "light" is deliberate: start small, one area first, and make it searchable. A fully agentic system with autonomous routines is rarely needed in a practice – and only conceivable with a strict data-protection concept.
02How it works
- 01
Take stock: where does your process knowledge live?
In heads, on notes in the treatment room, in an old "practice manual" folder nobody opens. Also: which questions interrupt whom most often, and what does only your longest-serving assistant know?
- 02
Build the structure – small and documented
A clear filing system: processes, standards, team organisation. With simple rules – and a firm line that patient data doesn't belong in it.
- 03
Fill one area first
Usually prescriptions, referrals and appointment types – the topics that cause the biggest queues at reception. Existing material (manual, notes) moves in curated, not copied.
- 04
Make it searchable
The team – and new assistants in particular – can ask the knowledge base questions and get answers with the source. That's the core of the light version: look things up instead of interrupting.
- 05
Optional: AI drafts, you sign off
Once that works reliably, the AI can suggest changes to standards (a new hygiene rule, new holiday dates) – but only as a draft, applied after sign-off. A practice rarely needs more.
03Stages
For practices, Tier 1 is usually the goal, not just the starting point – process knowledge you can ask questions of already solves most of the problem.
Starter and core: a knowledge base you can ask
Processes, standards and team organisation are organised so you can ask them questions. New assistants look things up instead of interrupting their colleague at the desk. For many practices, that's enough.
Next step: the AI drafts (with sign-off)
Changes to standards and processes are prepared as drafts and applied after sign-off – the knowledge base stays up to date without anyone maintaining it by hand.
Only with a data-protection concept
A fully agentic system with autonomous routines is rarely needed in a practice and only conceivable with a strict data-protection concept – deliberately not part of the standard offer.
04Demo
The light practice knowledge base of the fictional “Dr Winter Practice” – on the left the order in which processes and standards live (patient data deliberately stays out), on the right three everyday moments with it. Click through and watch where it lights up in the tree.
The architecture behind it has two layers: at the bottom the vault – structured folders and notes that belong to the business. Above it the AI layer with documented rules: it reads only at first, then writes with approval, and sensitive areas stay locked. Calendar and tools dock on, routines run within clear limits.
The proof of this model: this website came about this way – I work in such a system myself every day, with AI as an employee inside it.
05Technical implementation
- The knowledge layer belongs to the practice: a Markdown vault (e.g. Obsidian), Notion or a similar system – your content, no vendor lock-in.
- The AI layer is an assistant with access to the vault (e.g. Claude Code as the agentic layer), working to documented rules on which areas it may read and (with sign-off) write to.
- A hard system boundary: patient data stays in the practice management system; the knowledge base holds only processes and standards.
- Permissions in stages: read access first, write access only as drafts with sign-off – sensitive areas excluded.
- Upkeep routines are set up from the start, but kept lean – the light version depends on a well-maintained core, not on automation.
- Typical content: prescriptions and referrals, appointment types, hygiene and equipment standards, holiday and cover arrangements, onboarding for new assistants.
for context · This shows how something like this usually gets built – not how it has to be built for you. What actually fits is worked out in the project: around your existing setup, your team, and what stays easy to maintain.