Performance marketing for hospitals: the spend traps

Performance marketing in a hospital group looks like performance marketing anywhere until you examine what the numbers are actually buying. The dashboard shows leads at a cost per lead that improved month-on-month. The unit shows flat footfall. Both are accurate. Between them sit about eight structural traps, most of which reward the agency and the platform while quietly transferring money out of the business.

None of these traps is a scandal. They are the default behaviour of a system where the incentives of the people optimising the account are not the incentives of the hospital, and where the metric everyone agreed to is easier to move by changing what you count than by changing what you achieve.

Here is the list, in roughly the order I would go looking for them in an account I had just inherited.

Trap one: bidding heavily on your own brand

Brand search is the cheapest conversion in the account and therefore the most seductive. Someone searching your hospital’s name plus “appointment” will convert at a rate nothing else touches, which makes your blended cost per acquisition look excellent and your incrementality close to zero. You are paying to appear above your own organic result.

There is a defensive case, and it is real: aggregators and occasionally competitors bid on your name, and in some categories the paid result above your listing genuinely intercepts a patient who would have reached you. So do not switch it off entirely. Cap it. Set a fixed monthly ceiling on brand terms, review whether a competitor is actually present rather than assuming, and — this is the part people skip — exclude brand spend and brand conversions from the blended efficiency number you report. Otherwise every agency review becomes an argument you cannot win, because the easiest way to improve the headline is to pour more money into the term that was already yours.

Trap two: fighting aggregators on their own ground

Doctor-discovery platforms and appointment aggregators out-rank you on generic condition and specialty searches, and they will out-bid you too, because their economics differ from yours fundamentally. They monetise a click across many hospitals; you monetise a patient at one. They can pay more for the same keyword and still profit.

The instinct is to escalate. I have watched a group raise bids on “best cardiologist in” queries for a quarter, win a modest share of a very expensive auction, and generate enquiries that were overwhelmingly comparison-shoppers already talking to three hospitals. The better answer is uncomfortable: concede the top of the generic funnel, compete where the aggregator is weak. That means consultant-name searches, procedure-specific and sub-specialty intent, locality plus specialty, second-opinion journeys, and your own doctor profiles built well enough to rank organically. It also means deciding deliberately whether to be present on the aggregators as a paid participant — a strategic question about channel dependence, not a media-buying one, and it belongs with the leadership rather than in the account.

Trap three: lead forms that manufacture volume

Native lead forms on social platforms will give you volume at a cost per lead that makes the monthly review pleasant. A large share of those leads will not answer the phone, and of those who do, a meaningful share did not knowingly enquire about a hospital — they tapped through a pre-filled form on a feed.

The tell is a widening gap between lead count and contactable-lead count, and the fix is friction. Add a qualifying question the patient has to type. Ask for the city or the locality. Require the condition or the service. Your cost per lead will rise, possibly sharply, and your cost per appointment will fall. If your agency is compensated on leads, expect resistance framed as concern about volume.

I would go further. Stop reporting cost per lead upward at all. Report cost per contactable enquiry and cost per booked appointment. The day we changed the definition, three of our best-performing campaigns became our worst, which was information we had been paying not to have.

Trap four: the emergency-query trap

This one is both a spend trap and an ethical problem, and it deserves more attention than it gets.

Queries that signal an acute emergency — chest pain right now, stroke symptoms, an accident — do not belong in a lead-generation funnel. Nobody having a cardiac event fills a form and waits for a callback. If your campaign serves them a lead form, you have taken money to delay care, and a delayed callback to a genuine emergency is an incident, not a missed opportunity.

What to do instead: serve emergency intent with a click-to-call unit, the ambulance number, and the address, and route those calls to a line that is answered in one ring by someone trained to triage rather than to convert. Measure success as call connection rate and time to answer, not as cost per lead. And check your keyword lists for acute-symptom terms sitting inside a broad match group, because that is where they end up without anyone deciding to put them there.

Trap five: agency incentive misalignment

Most healthcare agency contracts in India pay a percentage of media spend, a fee per lead, or some blend. Both structures pay the agency to do something other than what you want.

Percentage of spend pays them to spend more, which is why every quarterly review contains an opportunity requiring additional budget. Cost per lead pays them to find the cheapest definition of a lead, which is how you end up with lead forms full of untraceable numbers and brand search quietly consuming a disproportionate share of the budget. Neither structure pays anyone for the enquiry-to-appointment rate, which is where your actual losses are.

What has worked better: a flat retainer for the work plus a bonus tied to booked appointments in named service lines, with a floor on contactability. Agencies will tell you they cannot be held accountable for appointments because the contact centre is not theirs. That is a fair objection and the right response is to give them visibility into the contact centre data and a seat in that weekly review — at which point the ones who were serious get better and the ones who were not become obvious within two months.

Trap six: creative that breaches medical advertising norms

Indian medical advertising is constrained by professional conduct regulations that bind your doctors personally, by drug and magic-remedies legislation, and by advertising standards codes — and the practical risk is not usually a regulator. It is a complaint from a competitor, a screenshot in a WhatsApp group of local physicians, or your own medical council-registered consultant discovering that an agency put a success rate next to his face.

The recurring offenders are simple and preventable. Guarantees and cure claims. Success-rate percentages without a cited source and a denominator. Comparative superlatives — best, number one, most advanced — where nothing substantiates them. Patient testimonials used as clinical evidence rather than as experience. Before-and-after imagery in cosmetic and bariatric work. Price-led surgical offers presented like a retail discount. Naming a competitor. Any implication that a consultant personally guarantees an outcome.

Build a one-page creative standard, get it signed by the medical director, and put a clinician in the approval path for every service line where a claim appears. Then hold the agency to it contractually, because the reputational cost of a pulled campaign lands on the hospital and not on them. I have had to withdraw a live campaign on a Friday evening after a consultant objected to a line he had never been shown, and the hour I spent apologising was entirely deserved.

Trap seven: optimising a campaign against a full calendar

A campaign that generates demand for a consultant whose next slot is eleven days away is not performing, whatever the dashboard says. The enquiry converts elsewhere. This is the trap that looks least like a marketing problem and costs the most.

Before scaling spend on any service line, look at the slot availability for the consultants that line depends on. If utilisation is already high and the waiting list is long, additional spend buys you nothing but an increase in your own lost-enquiry rate. Redirect it to a service line with capacity, or to a unit with capacity, and say plainly in the review that capacity is the constraint. That sentence is unpopular and it is usually true.

Trap eight: one campaign run across every unit

Group-level campaigns are efficient to produce and wasteful to run. A single city campaign sends an enquirer from one end of the city to whichever unit the landing page defaults to, and patients do not travel across a metro for a consultation they could have twelve minutes away. The enquiry is generated, counted, called, and lost on geography.

Worse, the cost of that waste is invisible, because the unit that received the misrouted enquiry records it as a poor-quality lead and the unit that should have received it never knows it existed. Radius targeting per unit, a landing page that resolves to the nearest unit, and a routing rule in the contact centre are not sophisticated tactics. They are the difference between a campaign and a leak. The only place I would run genuinely group-level spend is brand awareness and quaternary care that patients do travel for — transplants, complex oncology, specialist paediatric surgery — where the catchment is the region rather than the locality.

What to optimise for instead

• Cost per contactable enquiry, deduplicated across channels, replacing cost per lead everywhere it currently appears.
• Cost per booked appointment by service line and unit, accepting that the number is an estimate and reporting it anyway.
• Non-brand share of spend, tracked as a discipline so brand search cannot quietly grow.
• Contactability rate by campaign, which exposes lead-quality problems within a week rather than a quarter.
• Call connection rate and time to answer for every click-to-call and emergency asset.
• Total service-line volume against total service-line spend on a rolling quarter, as the reality check on everything above.

If you are starting this next quarter

1. Pull a keyword-level report and calculate brand versus non-brand share of spend. Do this first; it usually reframes the whole conversation.
2. Audit for acute-symptom terms in broad match and move them to call-only assets this week.
3. Redefine the reported metric to cost per contactable enquiry and restate the last two quarters on the new basis so the comparison is honest.
4. Add a typed qualifying field to every native lead form and accept the volume drop.
5. Overlay slot availability on spend by service line and cut spend where capacity is the binding constraint.
6. Sign a one-page creative standard with the medical director and route claims through a clinician.
7. Renegotiate the agency contract at the next cycle to retainer plus appointment-linked bonus, and give them contact centre visibility as part of the deal.

Performance marketing for a hospital is mostly an exercise in refusing the metrics that are easiest to move. Get that refusal into the contract and the reporting template, and the rest is housekeeping.

Questions people ask