When we inherited this account, there were 14 ad accounts, 14 billing setups, and 14 clinic managers who each thought they were doing it right. Total monthly spend: ₹3.2L. Booked appointments from ads: nobody actually knew. The lead data lived in WhatsApp threads, front-desk notepads, and a CRM that three out of fourteen clinics had bothered to set up correctly.
This is the multi-location healthcare problem. And it's more common than it should be.
The multi-location problem nobody talks about
Fragmented ad accounts mean fragmented data. When each clinic is running its own campaigns, you lose the ability to see which locations, creatives, and offer types are actually driving seated patients. You end up with 14 small accounts, each with too little data to optimise properly, each burning money on broad audiences because no single account has enough conversion history to train the algorithm.
The worst part: the metric everyone was tracking was cost per lead, not cost per booked appointment. They were optimising for form fills. Whether those people showed up at the clinic was considered a front-desk problem, not a marketing problem. This is how multi-location healthcare accounts quietly bleed money for years.
Step 1 — Consolidate everything into one Business Manager
The first conversation with ownership was about consolidation. Every clinic's campaigns moved into a single Meta Business Manager, under one billing account, with one unified reporting dashboard. The initial resistance was predictable: clinic managers felt like they were losing control of their own marketing spend.
What they gained was significantly more valuable. Shared custom audiences across all 14 locations. Cross-location retargeting pools large enough to actually function. A single creative testing environment where a winning hook or offer at one location could be deployed across all 14 within 48 hours. And for the first time, an apples-to-apples comparison of how each clinic was performing relative to the others.
Step 2 — Geo-fencing by catchment area
The default targeting for every clinic had been city-wide. Mumbai or Pune or Nashik, depending on location, with a blanket 15–25km radius. This sounds reasonable until you look at patient data. Across all 14 clinics, more than 78% of new patients came from within 3km of the clinic location.
City-wide targeting was burning roughly 60% of each clinic's budget reaching people who were never going to travel 14km for a dental appointment when there was a clinic closer to their home. We rebuilt every campaign with geo-fenced radii between 2.5km and 4km depending on each clinic's local competition density. Cost per click went up slightly. Cost per booked appointment started falling immediately.
Step 3 — Wire offline conversions
The attribution gap was the hardest fix, but it was the most important. The CRM we were working with had an API endpoint. We built a simple webhook that pushed "appointment booked" and "patient seated" events back into Meta's Offline Conversions API with a 24–48 hour delay.
This changed everything about how the algorithm was optimising. Instead of telling Meta to find people who fill in forms, we were telling it to find people who show up for dental appointments. The lookalike audiences rebuilt themselves around this signal within 2–3 weeks. Cost per seated patient started falling. The front-desk no-show rate also dropped, because the algorithm was now reaching people with genuine intent rather than impulse form-fillers.
The creative split that worked
Running ads for 14 dental clinics might seem like a creative challenge, but the category splits itself naturally. There are high-value treatments — implants, aligners, smile makeovers — that command premium CPLs and attract patients with high lifetime value. And there are routine treatments — cleanings, fillings, check-ups — that drive appointment volume but at lower per-visit revenue.
These cannot share the same campaign. High-value treatment ads need longer consideration content, treatment comparison creative, and before/after social proof from previous patients. Routine visit ads need urgency, proximity, and a low-friction booking CTA. We split budgets — 40% to high-value treatment campaigns, 60% to routine visit volume campaigns — and built separate landing pages for each category. Conversion rate on the high-value campaigns tripled when we stopped forcing them to compete with volume-oriented creative.
The 38% number — what it means
The headline reduction was cost per booked appointment, not cost per lead. This distinction matters enormously in healthcare marketing, where a "lead" is a form submission and an "appointment" is a patient in the chair with a treatment plan. Most healthcare marketers report on the former because it's easy to measure. The latter is what actually drives revenue.
Before consolidation and offline conversion tracking, the chain was spending ₹3.2L/month with no clear picture of how many of those rupees turned into seated patients. After 6 months with the new setup, spend was essentially flat at ₹3.4L/month, but the number of booked appointments from ads increased by 62%. Cost per booked appointment dropped 38%. And for the first time, the marketing team could walk into a review meeting and tell ownership exactly what each rupee of ad spend produced.
If you're running ads for a multi-location healthcare business and your data looks anything like this, we'd be happy to take a look. We review your setup before the call so you're not listening to generic advice from someone who hasn't looked at your numbers.