Everything we know, written down.
We build patient messaging for clinics across MENA, which means we spend our time on platform rules, documentation and the line between what software may answer and what it must not. All of it is here, including the parts that argue against buying anything.
Start here
Three references that cover most of it
Guide — Pillar guide
AI receptionist for clinics: a practical guide
Most material on this subject is written to sell software. This is written to help you decide whether you need it — including the cases where the answer is no.
10 min read
Guide — Pillar guide
WhatsApp for clinics: the rules nobody explains before you sign up
WhatsApp is where MENA patients actually message clinics. It is also the channel with the most rules, and most of them only become visible after a template gets rejected.
8 min read
Guide — Pillar guide
Deploying AI in patient messaging without creating a liability
The question is not whether AI in patient messaging is permitted. It is whether you can, six months later, reconstruct exactly what was said to a patient and why.
7 min read
Calculator
What is your inbox losing?
Four numbers you already have — enquiry volume, the share arriving out of hours, your two booking rates and your average first-visit value. The working is shown, so you can argue with the assumptions rather than the conclusion.
Also worth reading first
Guides
Long-form references on the things clinics have to get right before any software helps — channels, knowledge, and the clinical boundary.
Pillar guide
Deploying AI in patient messaging without creating a liability
The question is not whether AI in patient messaging is permitted. It is whether you can, six months later, reconstruct exactly what was said to a patient and why.
7 min read
Pillar guide
Instagram DMs are where aesthetic patients start. Most clinics lose them there.
A clinic can run a beautiful Instagram account and convert almost none of it, because the funnel does not end at the post. It ends in a DM at 11pm that nobody reads until Sunday.
7 min read
Pillar guide
WhatsApp for clinics: the rules nobody explains before you sign up
WhatsApp is where MENA patients actually message clinics. It is also the channel with the most rules, and most of them only become visible after a template gets rejected.
8 min read
Pillar guide
AI receptionist for clinics: a practical guide
Most material on this subject is written to sell software. This is written to help you decide whether you need it — including the cases where the answer is no.
10 min read
By specialty
How the same platform behaves differently in a dermatology inbox, a dental practice, a fertility clinic and a multi-branch group.
Solution
AI patient messaging for multi-branch clinic groups
Groups fail at messaging in a specific way: the brand has one Instagram account and the branches have different prices, different doctors and different rules.
3 min read
Solution
AI patient messaging for fertility and IVF clinics
This is the specialty where getting the boundary wrong does the most damage — and where the practical questions are so repetitive that answering them by hand is unsustainable.
3 min read
Solution
AI patient messaging for dental clinics
Dentistry has the clearest split in the inbox: a large volume of quotable, schedulable work, and a stream of pain messages that need a human immediately.
3 min read
Solution
AI patient messaging for dermatology and aesthetic clinics
No specialty is more exposed to the after-hours enquiry, and none has a sharper line between the question you should answer and the photo you must not.
3 min read
Comparisons
Where the alternatives genuinely win, written to be useful rather than to win an argument.
Comparison
Building your own clinic AI vs buying one
We are not neutral here, so this is written as a checklist rather than an argument. If you can tick the list, build it.
4 min read
Comparison
AI receptionist vs outsourced answering service
Both solve the same complaint — nobody is answering — and they solve it with opposite trade-offs. The choice mostly comes down to whether your enquiries arrive by phone or by message.
4 min read
Comparison
Chatbot vs AI agent for a clinic: what actually differs
The distinction is not marketing. The two architectures break differently, and the one that suits your clinic depends on how predictable your inbox is.
3 min read
Articles
Shorter pieces on operating a clinic inbox: response times, escalation design, no-shows, bilingual messaging and what to measure.
Operations
No-shows are a messaging problem before they are a policy problem
Most clinics respond to no-shows with a deposit policy. It works, and it also filters out patients you wanted. The messaging fixes come first.
4 min read
Measurement
The five numbers worth tracking on a clinic front desk
Most clinic messaging dashboards report volume, which nobody can act on. Five numbers actually change decisions, and one popular metric quietly causes harm.
4 min read
Language
Arabic, English, and everything in between
Bilingual patient messaging is not a translation problem. It is a detection, terminology and review problem, and getting it wrong is visible to the patient immediately.
5 min read
Design
When an AI should stop and get a human
Handoff is not the failure case of an AI receptionist. It is the feature that makes the rest of it deployable in a clinic at all.
5 min read
Knowledge
Building a knowledge base an AI can actually answer from
Clinics upload a folder of PDFs and wonder why the answers are vague. The problem is almost never the model. It is that the material was written for humans who could ask a follow-up question.
5 min read
Operations
The real cost of the message nobody answered
Every clinic knows it misses after-hours messages. Very few have measured it, and the arithmetic is usually worse than the guess.
4 min read
Then bring us your actual inbox
Forty real messages, photos and voice notes included. We will show you every answer, the source behind it, and the ones we would refuse.