Case study · Functional medicine
The enquiry system. That knows when not to answer.
One AI-assisted enquiry system: the clinic manager approves every reply, and clinical questions are never drafted by AI.
- Client
- A functional medicine clinic
- Sector
- Functional medicine
- Location
- Dungarvan, Ireland
- Engagement
- AI system build (V2)
Clinical: held for a person
01
The situation.
The clinic is a functional medicine clinic in Dungarvan, offering a catalogue of specialist tests and consultations. Most new patients start with a question, and the clinic manager, Lisa, answers each one personally.
Behind that personal service was a lot of manual juggling:
- Enquiries came in by email, Instagram, Facebook and manual entry, and each one was checked and answered by hand.
- Three years of client history lived in three separate Excel sheets, plus an export from the lab portal.
- Every week brought more than 20 manual status updates, tracking each sample from kit sent to results received from Nordic Labs.
Patients are making health decisions. Replies had to sound like the clinic, and anything clinical had to stay with a person.
02
What we built.
AI triage and reply drafting, with a human gate on every email. The system watches the clinic’s inbox, sorts each enquiry and drafts a reply in the clinic’s own voice. Lisa approves every email before it goes out. That was her condition from day one, and it carries through to version two.
Enquiry → reply
6 steps · a person approves
An enquiry arrives by email, Instagram, Facebook, or manual entry for phone calls.
The AI filters out spam and vendor mail.
It classifies the enquiry: a critical case, test information, symptoms mentioned, or a direct test query.
Clinical or regulated enquiries are never drafted by AI. They route to a “speak to your GP first” response and stop there.
Everything else is drafted from a library of 24 templates, carrying live pricing.
Lisa reviews it, edits if needed and approves it. Test orders then move across a five-stage board, advancing automatically when Nordic Labs emails arrive. Every update lands in one client record.
Clinic manager
03
How it works day to day.
Lisa works from one dashboard:
- a five-stage test-order board
- 3,216 live client records
- reporting on enquiry-to-test conversion, drop-offs and revenue
She edits the pricing and wording in the template library herself, with no development ticket. The version-two training document and walkthrough are in her hands to bring the wider clinic team onto the system, and weekly Thursday check-ins keep fixes moving.
Human approval, every time: in the first 30 days, 81 of 214 enquiries were identified as clinically sensitive and held back from drafting. That is the guardrail doing exactly what it was designed to do. No email leaves the clinic without Lisa’s approval.
04
The results.
- 214inquiries handled by AI in 30 days
- 21 hrsgiven back to the team in 30 days
- 3,216clients in one live record
- 81clinical inquiries flagged and held for a human
SourceThe clinic’s own system records, 30 days to 29 July 2026. Time saved calculated at 6 minutes per inquiry.
05
How we got there.
Apr 2025
Vision and V1
Mapped the clinic’s enquiry process and goals, then built a first version of lead handling, with Lisa approving every outbound reply.
Jun 2026
V2 delivered
Expanded into a central system: three years of Excel records and the lab export merged into one database, a five-stage test-order board, and a template library the clinic can edit itself.
Jul 2026 onward
Rollout and weekly support
The training document and walkthrough went to Lisa. Weekly Thursday check-ins.
06
What happened next.
With one live record of 3,216 clients, new options opened up:
- Reactivating dormant clients: 147 clients have been inactive for 12 months or more. Cohort email campaigns are next.
- A wider team rollout: the version-two training materials are with Lisa.
- Self-serve templates: Lisa edits pricing and wording herself.
- Conversion reporting: enquiry-to-test conversion, drop-offs and revenue, all from one record.
81 of 214 inquiries, nearly two in five, were held back from the AI. In healthcare, that is the result that matters most.
07
Tools and systems.
- The clinic’s own email accounts
- Nordic Labs (lab emails and the lab portal export)
- Excel (three legacy sheets)
- the template library (24 templates)
Start with one enquiry inbox. We will show you where AI pays, and where it should stay out.
Start with a conversation.
A free 45-minute strategy call with a founder. No pitch deck. You leave knowing where AI would pay in your business, whether or not you work with us.
On the call
- 01Where AI would pay in your business.
- 02What your first system could be.
- 03What your AI Blueprint would cover.
Before you book
What does it cost?
Every engagement starts with the AI Blueprint, priced up front. The first system is a fixed fee, set from what the Blueprint finds. Nothing is quoted cold.
Where does our data go?
Nowhere new. It runs in secure environments, on your cloud or your own servers. Every action is logged, and nothing trains anyone else's model.
How much of our team's time does it take?
Most at the start, while we learn how the work really happens. After that, a short weekly session with the people using it.
Who owns what you build?
You do. The systems, the data, the prompts and the documentation.
Pick a time that suits you.
Book a strategy callPrefer email? Write to hello@braind.io.