Launching a service line with ROI-accountable marketing investment

# Launching a service line with ROIROIReturn on Investment: the ratio of net profit to the cost of an investment. A 300% ROI means each dollar invested returns $3.View full definition →-accountable marketing investment

HCA Healthcare puts something on the order of $5 billion of capital to work in a typical year, and every hospital in the group competes for it in the same forum. A hybrid OR, a fourth cath lab, a linear accelerator: each arrives as a pro forma with a volume ramp, a contribution margin per case and a payback date. Marketing money attached to those launches has a choice. It can sit in a departmental budget, where it is cut in the first soft quarter, or it can sit inside the capital case with its own promised case count, where it survives and gets audited.

This lesson follows one launch through the second route: what was committed, against what, and who held the pen.

The ask, written as a capital line

HCA's discipline is visible from outside the company: same-facility admissions, equivalent admissions and surgical case growth are reported every quarter, so a service line that was supposed to add volume either shows up in a number analysts already watch or it does not. Write the marketing ask so it lands in that same place.

The launch we will follow (illustrative numbers, one 300-bed hospital, a new interventional cardiology program):

• $600,000 of media, agency and call centre cost, committed across the first three quarters
• against 360 incremental billable cardiac episodes in the first twelve months
• of which at least 35% commercially insured
• reaching a run rate of 45 cases a month by month 12

How you get from spend to those 360 is the chain the funnelfunnelThe customer journey from awareness to purchase, typically Awareness, Interest, Consideration, Decision, Action, with prospects narrowing at each stage.View full definition → lesson maps and the cost-per-case method the acquisition costacquisition costCustomer Acquisition Cost (CAC) is the total sales and marketing spend divided by the number of new customers gained in a period. It measures how efficiently you grow.View full definition → lesson gives you. The committee does not want that arithmetic reproduced. It wants the two numbers at the end: 360 cases, and $1,667 of marketing cost riding on each one.

Then contribution, because a case is worth what it is reimbursed at, and reimbursement depends on who pays.

| Payer | Share of cases | Avg contribution per case |

|-------|---------------|---------------------------|

| Commercial | 35% | $12,000 |

| Medicare | 50% | $6,000 |

| Medicaid / self-pay | 15% | $2,500 |

Weighted contribution per case: (0.35 × 12,000) + (0.50 × 6,000) + (0.15 × 2,500) = $7,575. Across 360 cases, roughly $2.73 million of first-episode contribution against $600,000 committed. The follow-on imaging, rehab and repeat interventions belong in the multi-episode model the lifetime valuelifetime valueLifetime Value: the total revenue (or profit) a customer generates throughout their entire relationship with your business.View full definition → lesson builds; carry it as a separate labelled projection and keep it out of the number you are graded on.

What the money actually buys: months

The $600,000 does not buy 360 cases in the abstract. It buys them earlier. A staffed cath lab with two interventional cardiologists on income guarantees carries the same fixed cost whether it runs two cases a day or six. Put that fixed cost at $400,000 a month for the illustration, and the program covers it at about 53 cases a month. Year one averages 30. The lab loses money through the whole ramp, exactly as the pro forma said it would.

So the marketing promise is a date, not a volume. Left to referral relationships alone, a new program facing two established competitors can take well into year three to reachreachThe number of unique people exposed to your message in a given period. Unlike impressions, reach counts each person once, no matter how often they see it.View full definition → its breakeven run rate. If $600,000 pulls that crossing from month 30 to month 18, the system avoids twelve months of an under-loaded lab, and the avoided fixed cost dwarfs the media budget. Say that out loud in the meeting: we are buying months off the ramp, and the cases are how we count them.

Holding it: tranches and gates

Money committed in one lump is money nobody revisits. Split it.

• Tranche 1, $200,000, weeks 1 to 12. Gate: kept new-patient visits. Not enquiries, not leads. A lead is a cost.
• Tranche 2, $250,000, released only if cost per kept visit is within 20% of plan and commercial share of booked visits is at least 30%.
• Tranche 3, $150,000, released against billed episodes, with the six to ten week lag from first consult to billed intervention written into the gate date so nobody grades quarter one on quarter one's billing.

The service line administrator and an analyst from finance co-sign each release. Marketing supplies media data and takes scheduling and billing data as given. Where the two disagree, billing wins.

Where the promise breaks

Mix drift beats volume. Suppose the campaign delivers 380 cases, 6% above plan, but commercial share slips to 25% while Medicare rises to 60%. Weighted contribution falls to $6,975 and first-episode contribution lands near $2.65 million: volume beaten, contribution missed by about 3%. Broad-reachreachThe number of unique people exposed to your message in a given period. Unlike impressions, reach counts each person once, no matter how often they see it.View full definition → media, discounted screening offers and community health fairs all pull mix that way. Put contribution on the scorecard, not case count, or the team will optimise honestly toward the wrong thing.

Capacity kills a pro forma quietly. Track third next available appointment for the new program every week. Once it passes roughly two weeks, extra acquisition media buys no-shows and competitor referrals: the patient books, waits, gets seen elsewhere, and your call centre absorbs the complaint. The standing rule is to pause consumer media at that threshold and move the money into referring physician work, which the referral lesson covers, until access recovers.

Quarterly reporting punishes launches at their worst moment. Quarter one shows the full spend against a fraction of the eventual cases, because the billing lag has not cleared. Report by cohort: this quarter's spend against the episodes that spend eventually produced. Systems that refuse to do this cancel good launches in month four and keep the capital asset they cannot fill.

AttributionAttributionA framework for assigning credit to the touchpoints that contributed to a conversion, so you can measure which channels and interactions actually drive results.View full definition → overclaim costs more than it wins. Some of the 360 arrived because a cardiologist sent them and would have come regardless. Hold out two comparable ZIP codes, or one secondary market, and compare. Where lift cannot be shown, write "attributed", not "caused". A marketing leader caught once claiming referral volume as campaign volume is not believed on the next capital request, and that is a permanent tax on every future launch.

The sunk launch is the failure nobody logs. The pro forma said three years, so year two spend renews without anyone making a decision. Write the kill criterion before the first dollar moves: if the month-12 run rate is under 30 cases with commercial share below 30%, the program drops to a maintenance budget and the remaining money returns to the pool.

When case volume is the wrong promise

Intermountain Health owns SelectHealth, so a share of the patients its hospitals treat are covered by its own insurance plan. An extra elective procedure on a member is revenue on one side of the house and claims cost on the other. A launch pro forma there has to split promised volume by who carries the risk, and the defensible marketing promise may be share gained among non-members plus leakage kept inside the system rather than gross cases. Same campaign, opposite sign, depending on the contract.

Apollo Hospitals sits at the other end. With no Medicare or Medicaid, and a large share of care paid out of pocket or through insurers and third-party administrators, mix is a question of acuity and price band rather than payer class. Apollo reports occupancy and average revenue per occupied bed each quarter, which sets a trap for a launch: filling beds with low-acuity work lifts occupancy and drags ARPOB down, so the campaign reads as growth in one line and dilution in the next. Apollo also reports its digital business as a separate segment with its own stated path to profitability, which is the right shape for a launch investment: visible, time-boxed, and judged against a date somebody named in advance.

🎬 [VIDEO: "Healthcare Marketing ROIROIReturn on Investment: the ratio of net profit to the cost of an investment. A 300% ROI means each dollar invested returns $3.View full definition → Explained" - youtube.com - a concise walkthrough of connecting campaign spend to patient acquisition and downstream value]

For background reading on service line strategy and hospital growth economics, the American Hospital Association publishes useful free material.

The one page that survives the meeting

Six lines, and everything else is annex: the committed spend, the promised cases with their payer split, the months taken off the fixed-cost ramp, the tranche gates with dates, the kill criterion, and the holdout design that will tell you whether any of it was incremental. If a line cannot be checked by someone outside marketing, it does not belong on the page.

Key Takeaways

• Ask for the money as capital, not budget. A launch defended as case volume against a payback date is reviewed on its own terms; a promotional line item is cut when the quarter tightens.
• The promise is a date as much as a number. Marketing's economic contribution to a new program is shortening the period the fixed asset runs under-loaded.
• Release money in tranches against gates someone else scores. Kept visits, then commercial share, then billed episodes, with the billing lag built into the gate dates.
• Beat the volume and miss the contribution and you still lost. Ten points of mix drift wipes out a 6% volume beat, so put contribution on the scorecard.
• Write the kill criterion before you spend. Launches rarely fail loudly; they renew quietly into a second year nobody re-approved.