Tulu Health has launched in the US, betting that AI built for patient relationships can plug the churn draining membership clinics.
Dr Adil Khan didn’t set out to build a healthcare company in the US first. Tulu Health, the AI-driven platform he runs as CEO, started in India, then moved to the UAE, then Southeast Asia. Only now, after three prior markets, has it arrived in the country most people would assume comes first.
There’s a reason for the detour. India, Khan says, is a market “where healthcare complexity is measured not in millions of patients but in billions.” Fragmented systems. Payer landscapes that shift by region. Clinical teams working from decades of institutional habit rather than a shared playbook. If a platform could survive that, the thinking went, everywhere else would be easier by comparison.
What it learned along the way wasn’t really about India, the UAE, or Southeast Asia specifically. It was a pattern that kept repeating: longevity clinics and medspas were losing members they didn’t even know they were losing. Quietly. At scale. Every single day.
A number most clinics don’t track closely enough
Membership churn at longevity clinics and medspas runs 15% to 20% a year, according to Tulu Health [1]. That number doesn’t announce itself the way a bad marketing quarter does. It shows up slowly, in a member who came in twice and then stopped answering the phone, in a follow-up that never got scheduled. By the time a clinic notices, a good chunk of that revenue is gone, and replacing a lapsed member costs three to five times more than keeping the one already there.
Khan’s read on why this keeps happening is blunt: the software clinics already use wasn’t built for it. Legacy EHRs, generic CRMs bolted on after the fact, standalone booking apps – they each solve a piece of the problem, but none of them follow a member’s relationship with a practice from the first phone call through years of membership.
Tulu Health’s answer is built around what it calls a Patient Twin, essentially a continuously updated profile of each patient that four separate AI tools read from and write back to. AI Recall goes after patients who’ve gone quiet, reaching them in whichever language and channel they’re actually likely to respond to. AI Front Desk picks up every inbound call or message around the clock, so nothing lands in voicemail. AI Care Navigator tracks who’s supposed to be doing what, next follow-up, next test, next check-in, and flags the gaps. A fourth tool, AI Benchmarks, is still coming and will let a clinic see how it stacks up against comparable practices.
None of the four require a full rebuild. Clinics can plug in whichever one addresses their most expensive problem first, then add from there.
Why Khan keeps saying this isn’t about the software
Ask Khan what the hardest part of entering a new country is, and he doesn’t point to the technology.
“Entering a new healthcare market is not a product problem. It’s a trust problem,” he said. “You cannot deploy AI inside a clinic without people who understand the clinical, operational, and regulatory context of that specific market.”
That’s the reasoning behind bringing in Dr Sanjiv Chopra, Faculty Dean for Continuing Medical Education at Harvard Medical School, whose involvement in the company’s clinical governance predates the US launch and now extends board-certified oversight into it.
Tulu Health says it already runs inside more than 50 longevity clinics and medspas spread across three continents, a base it’s using as its evidence for the US pitch rather than a fresh promise.
There’s an easy version of this story where AI is the interesting part. It probably isn’t. Longevity medicine works on the premise that people keep showing up, for monitoring, for adjustments, for the slow accumulation of small interventions that only pay off over years. A member who quietly stops answering the phone isn’t just a line item disappearing from a clinic’s books; it’s someone who was supposed to be on a longevity plan and, for reasons a spreadsheet won’t catch, isn’t anymore.
“Every new market is a new chance to prove that the right AI, in the right hands, changes what’s possible for patients,” Khan said. Whether that holds will come down to something less flashy than the algorithm: whether clinics actually use it to keep the people they already signed up.
