About this episode
The phone rings. No one answers. The patient moves on.
That gap — between a call and a missed call — is where clinics quietly lose revenue, trust, and new patients every single day. Tanmay Patel, Co-Founder of Lyngo, joined the team at HMDG on their No Appointment Necessary podcast for an 85-minute conversation about what AI phone answering actually does for clinics: what it handles well, where it falls short, how to evaluate vendors without a technical background, and what separates implementations that deliver from the ones that quietly make things worse.
No hype. No vendor pitch. Just a direct conversation about a question most clinic owners are already asking.
What you'll learn
- Why treating AI phone answering as infrastructure — not a gadget — changes how you evaluate and buy it
- The real cost of a missed call, and why most clinic owners consistently underestimate it
- How open APIs protect your clinic from being locked into a platform that stops serving you
- Why "it sounds human" is a terrible standard for assessing an AI phone system
- What actually happens to your front-desk team after AI is introduced — and why it's not what most people fear
- How to evaluate AI vendors even if you have no technical background
- Why a suspiciously cheap AI phone answering product should give you pause before you sign anything
- When AI phone answering clearly works — and when you should not use it at all
The conversation
Missed calls don't just inconvenience patients — they cost you
The conversation opens with something most clinic owners know but rarely measure: how often calls go unanswered, and what that actually means in real terms. The problem isn't random. It clusters. Early mornings when practitioners are arriving and getting set up. Lunchtimes when the desk is thin. Peak booking windows when the phone and the waiting room are competing for the same pair of hands.
"Most clinic owners think they're missing maybe 10 or 15 percent of calls. When they actually pull the data, it's closer to 40 percent during peak hours. That's not a small leak — that's a structural problem, and it's been there the whole time."
The episode makes the case that missed calls don't just lose a booking — they erode the kind of patient trust that's hard to rebuild. Someone calling in pain, or calling to follow up after a referral, doesn't usually call back. They find a clinic that picks up.
Where AI phone answering fits — and where it doesn't
A significant portion of the episode is spent on honest use-case mapping. AI phone answering is not a universal solution. It works well in specific situations, and it fails — sometimes visibly — when it's deployed in others.
"Where it works: after-hours calls that currently go to voicemail, overflow during booking spikes, first-contact triage for straightforward enquiries. Where it doesn't work: complex clinical questions, emotionally distressed callers who need a human immediately, situations where your processes aren't defined enough for anyone — human or AI — to handle consistently."
The conversation also addresses the clinics who "tried AI and it didn't help." The finding is consistent: it's almost never the technology. It's that AI was handed a job it wasn't set up to do — no clear escalation rules, no staff briefing on what the system handles versus what it hands off, no definition of what success actually looks like.
White-labelled wrappers, open APIs, and the vendor questions that matter
The episode goes deep on vendor evaluation — the part most clinic owners skip because it feels too technical. The conversation simplifies it considerably. The central question isn't about features. It's about architecture.
"A white-labelled wrapper on a third-party AI model has a fundamentally different data exposure profile to a purpose-built system with a direct API connection to your practice management software. You don't need to understand how it works under the hood — but you do need to know which one you're buying."
Covered in detail: why standardising on a single all-in-one platform is usually a long-term mistake, what an open API actually is and why it matters for clinic independence, and the specific red flags that indicate a vendor is re-packaging someone else's product rather than building something purpose-fit for healthcare.
What happens to your front desk after AI is introduced
One of the most practically useful parts of the episode deals with implementation and change management — specifically, how to bring your team along rather than around it.
"The single biggest implementation mistake clinic owners make is not the technology choice. It's not preparing the team: not setting clear escalation rules, not briefing staff on what the system handles and what it passes through, and expecting AI to fix processes that were already broken."
The episode is direct about what changes for reception teams: the volume of repetitive call-handling drops. The time freed up goes back to in-clinic patient experience, follow-up calls, and the work that actually requires a human. For most clinics that have implemented well, that's been the outcome staff appreciate most.
Who this episode is for
- Clinic owners actively considering AI phone answering
- Owners who tried AI and came away disappointed — this episode addresses why directly
- Practice managers responsible for front-desk operations and admin
- Healthcare founders evaluating their technology stack
- Anyone sceptical about the hype around AI but open to a grounded conversation
This episode is not for clinics looking for quick wins without changing how they operate, or anyone expecting AI to fix processes that don't work.
