When growth stalls, the reflex is to spend more on marketing. More channels, more budget, a new agency, a bigger creative test. It feels like progress because it is the one lever leadership can pull without touching anyone else's department. It rarely works, because the constraint is almost never at the top of the funnel. It sits in the handoffs: the moments where a patient moves from a click to a call, from a call to an intake, from an intake to an admission. Each handoff is owned by a different team, measured by a different number, and invisible to everyone except the people standing on either side of it.
I have run this diagnosis inside organizations that were convinced their marketing was broken. In most of them, marketing was the healthiest part of the machine. The leads were arriving. They were dying in transit.
The number that moves last
Marketing data reports the symptom last. By the time cost per admission drifts up on a dashboard, the friction that caused it has been compounding for weeks in the call-center queue, the referral inbox, or the scheduling backlog. The dashboard is a rear-view mirror pointed at the one part of the funnel that leaves a clean digital trail. Everything downstream (the voicemail nobody returned, the insurance question that stalled an intake, the discharge planner who stopped calling) happens in systems that were never wired to report upward.
So leadership optimizes what it can see. It tightens the ad account, swaps the landing page, renegotiates the media rate. Those are real levers, but they are levers on the cylinder that is already firing. Meanwhile the dead cylinder, the third handoff with no owner, keeps misfiring. No amount of top-of-funnel spend compensates for demand that arrives and then leaks out the side.
When the top of the funnel is healthy and the bottom is soft, stop buying more demand and start protecting the demand you have.
A pattern I see every quarter
Consider a behavioral health group with three facilities and a declining census. Leadership was ready to fire the agency. Spend was up, admissions were flat, and marketing was the obvious suspect. When we traced a week of inbound inquiries end to end, the picture inverted. The ads were producing more qualified calls than the year before. Roughly a third of those calls hit an after-hours line that routed to a voicemail box the weekend staff did not own and the weekday staff assumed the weekend staff checked. Families in an acute moment do not leave a second voicemail. They call the next provider on the search results page.
The marketing team was being blamed for a leak two departments downstream. Fixing the ad account would have done nothing. Assigning a named owner to the after-hours queue, with a service standard and a callback clock, recovered the demand that was already being paid for. The lesson generalizes: when the top of the funnel is healthy and the bottom is soft, stop buying more demand and start protecting the demand you have.
Make the journey observable
The fix starts with one view that follows demand from first contact through first clinical visit, with a named owner at every step. That is an operations project before it is a technology project, and it is the heart of how I run a diagnosis. The checklist I work through is short:
- Map every handoff between first contact and first visit, and write the name of a single accountable person next to each one.
- Attach one number to each handoff (conversion rate, response time, drop-off) so the step can be measured, not argued about.
- Instrument the seams so the number updates without anyone assembling a spreadsheet.
- Review the worst-performing handoff weekly until it stops being the worst, then move to the next one.
None of this requires a new marketing budget. It requires making the operating model observable and owned so the failure points stop hiding between departments. When leaders can finally see where patients are lost, the conversation changes from "spend more" to "fix the third handoff": a smaller, cheaper, more durable intervention. A hospital system I worked with found millions in leakage this way, none of it visible until the systems were connected at the seams.
Growth is an output, not an input
Growth is what a healthy operating model produces, not something you buy your way to when the model is broken. Build the system so the handoffs are visible and owned, and growth stops being a mystery you throw budget at. Leave the handoffs invisible, and you will keep paying full price for demand that leaks out before it ever reaches a clinician.
