A Vital Sign Taken Once
A temperature taken once, with no history and no context, tells you almost nothing. Most Nigerian health businesses treat customer acquisition cost the same way — a single blended number, read in isolation, acted on in panic.
A temperature of 38.4 means very little on its own. It means something different in a two-year-old than in a seventy-year-old, something different at hour one of an illness than at day five, something different if the last three readings were climbing than if they were falling. A clinician who treats a single number as the whole story is not practicing medicine so much as performing the appearance of it. And yet customer acquisition cost — arguably the single most consequential number in any distribution-heavy health business — routinely gets treated exactly that way: one blended figure, reported at a board meeting, acted on as if it arrived with no history and no context attached.
The first fix is unglamorous and gets skipped constantly: pick a denominator you can actually defend. "Cost per lead generated" is not a real denominator in a business that hasn't converted the lead into anything — it measures activity, not outcome, and a business optimizing for cheap leads with no downstream discipline will happily generate thousands of them that never become revenue. The honest denominator is a paid consultation, a completed diagnostic episode, an activated employer account — something that only exists once actual value has changed hands. Everything upstream of that point is a funnel metric worth tracking, but calling it CAC dresses up an activity number in the clothing of an outcome number, and the confusion tends to survive exactly long enough to do real damage to a budget decision.
Why the number is naturally lumpy
Even with a defensible denominator, Nigerian health CAC does not behave the way a SaaS acquisition funnel behaves, and pretending otherwise is where a lot of imported playbooks go wrong.
Health decisions carry a trust step that most consumer purchases don't. Someone asks the price on a Monday and books on a Thursday, after talking to a spouse, checking with an employer, or simply sitting with the decision for a few days because health spending competes against a dozen other claims on a household budget that don't get deferred as easily as they should. That gap between enquiry and commitment is not a leak to be engineered away. It's a structural feature of how people decide to spend money on their own bodies, and a CAC model built on the assumption that a five-minute decision cycle is achievable will consistently read a healthy business as an underperforming one.
Geography changes the read as well, and not in a way that shows up cleanly in a spreadsheet unless someone deliberately builds it in. A clinic on Victoria Island, a branch in Ikeja, and a kiosk-style presence in Yaba carry different staffing costs, different footfall patterns, different opening hours shaped by different commuting rhythms, and different customer expectations about what a consultation should cost. Blending all three into one company-wide CAC produces a number that describes none of the three locations accurately, and a founder deciding where to expand next, working only off the blended figure, is making that decision with the geographic signal averaged out of the data entirely.
And financing adds a layer most consumer businesses simply don't have to reckon with. The person receiving care, the organization authorizing it, and the payer settling the bill are frequently three separate parties: an employer funding a screening benefit for staff who never see an invoice, a cooperative underwriting members through a scheme none of them fully understand the mechanics of, an individual paying cash because no scheme covers them at all. Nigeria's National Health Insurance Authority Act formalizes several of these tracks side by side — a Formal Sector programme, an Organised Private Sector programme, GIFSHIP for the informal sector, and coverage for vulnerable groups routed through the Basic Healthcare Provision Fund — and each one carries a different sales motion, a different decision-maker, and a different timeline to close. Each of those paths has a completely different cost to acquire and a completely different timeline to close, and collapsing them into a single acquisition number erases the very distinction that would tell a founder which path is actually worth investing more field effort into.
Consider what actually sits underneath a single field officer's acquisition number in a given week. Picture a rep whose Tuesday route runs six pharmacy visits across Ikeja, each one meant to end with a signed staff briefing or a batch of referral cards handed to the counter. One pharmacy's manager is away at a supplier meeting and the visit has to be rescheduled on the spot. Another visit runs forty minutes long because the owner wants to talk through pricing objections that a five-minute pitch was never built to answer. By mid-afternoon the rep has completed three of six planned stops, and whatever CAC number gets attributed to that day reads as expensive, in isolation, compared to a day where all six visits went smoothly and quickly. Neither day is more representative than the other. A CAC read that doesn't account for the ordinary variance of a field route — traffic, a manager's unavailability, a longer conversation that happens to close a bigger account two weeks later — will punish exactly the visits that took longer because they were doing real persuasion work, and reward the quick ones that may have produced nothing but a business card left on a counter.
Building a panel, not a headline number
The clinical instinct that translates well here is the one where a single vital sign never gets read alone. A raised heart rate paired with a normal blood pressure tells a different story than the same heart rate paired with a falling one. CAC deserves the same company. Track response time alongside it — how quickly a fresh enquiry gets a real answer, not an automated one. Track booking and attendance rate, because a business can hold a stable CAC while quietly bleeding people between "booked" and "showed up," which just relocates the same underlying problem to a stage nobody's watching. Track completed or claimable episodes rather than raw visits, since a visit that doesn't resolve into paid or reimbursable care isn't yet a finished transaction, whatever the appointment calendar says. Track contribution margin per episode, because a channel producing a low CAC on service lines that lose money on delivery is not actually cheap. It is just deferring the cost to somewhere else in the ledger. And track repeat and referral behavior, because a customer acquired cheaply who never returns and never refers anyone is a fundamentally different asset than one acquired at twice the cost who does both.
Read together, this panel tells you things the headline number alone cannot. A rising CAC alongside a stable repeat rate usually means the front of the funnel got more expensive to reach — competition increased, a channel saturated, an algorithm changed — while the underlying service is still doing its job once someone arrives. A stable CAC alongside a falling completion rate is a much worse signal dressed in a much better-looking number: the business is paying the same amount to acquire people who are increasingly failing to become finished transactions, and a founder looking only at the acquisition line will miss it entirely, sometimes for months.
What to do when the number actually moves
When CAC does move in a way that looks meaningful rather than merely noisy, the instinct in most rooms is to react immediately — cut the channel, cut the team, cut the budget. A better sequence borrows directly from clinical triage. First, verify the reading before trusting it: did the measurement window change, did an unpaid pilot or a promotional discount get folded into the denominator, did a reporting change quietly redefine what counts as a conversion. A meaningful share of alarming CAC movements turn out, on inspection, to be measurement artifacts rather than real signals, and acting on an artifact wastes real resources solving a problem that was never there.
Second, once the reading is confirmed real, locate where in the funnel the change actually originated, using the panel rather than guessing — did response time slip, did the booking-to-attendance rate fall, did a particular location's staffing change. Third, apply one bounded intervention at a time, with a specific review date attached, rather than restructuring five things simultaneously and losing the ability to tell which change actually mattered.
This matters because the vital-sign framing cuts in both directions, and the second direction is the one founders reaching for a spreadsheet tend to forget. A founder who sees CAC climb and responds by gutting field allowances or eliminating the human step in the sales process may be amputating exactly the trust-building work that makes healthcare distribution function in a market where roughly nine in ten working Nigerians earn their living outside formal, salaried employment, according to the National Bureau of Statistics's labour force survey — a market where a stranger vouching for a service in person still carries weight that a banner ad does not, and cannot, replicate. Cut the field officer's allowance to fix a number, and the number may fall for a month before the underlying trust-generation machinery quietly stops working and the real cost shows up somewhere else in the panel, later, and harder to trace back to its source.
The opposite failure is just as common and gets celebrated far more often than it should. A team that reports a falling CAC without checking completion rates may simply be buying unserved demand more cheaply — attracting more people who ask the price and never book, or book and never show, dressed up in a metric that looks, on its own, like unambiguous progress.
A temperature of 38.4, read alone, tells you almost nothing. Read against yesterday's reading, this morning's, and the trend across the week, it tells you whether to worry and roughly how much. CAC deserves exactly that discipline, and no more mysticism than a number on a chart actually earns.
Notes on sources
- National Health Insurance Authority Act, 2022 and its parallel coverage tracks (Formal Sector, Organised Private Sector, GIFSHIP, Vulnerable Group via the Basic Healthcare Provision Fund): nhia.gov.ng/nhia-act.
- 87.3% of Nigerian workers self-employed and 92.3% of employment informal, Q3 2023: National Bureau of Statistics, Nigeria Labour Force Survey.