Seen Is Not a Response

Ask most clinic owners how many of their enquiries turn into paying patients and you get a number delivered with total confidence and no supporting evidence. "About half." "Maybe sixty percent." The number is invented on the spot, generously, because nobody enjoys reporting a low number about their own business to a stranger asking questions. What they are actually describing is a feeling — a general sense that things are going fine, assembled from the patients they remember seeing this week and none of the enquiries that quietly evaporated before ever becoming a name on a schedule.

The fix for this is not a dashboard. It's a forty-eight hour trace, done by hand, on every fresh enquiry that arrives in one window: WhatsApp, Instagram DM, a phone call, a walk-in, a message through the Google Business Profile, an employer referral, and the one channel nobody puts on the org chart: a staff member's personal number that patients somehow acquired and now use directly. Follow each one from first contact to whatever it becomes — booked, paid, attended, or dead — and you get something no amount of guessing produces: a map of exactly where the funnel is losing people, rather than a vibe about it.

Hour zero to four

The first window is about response, and the discipline here is refusing to count things that look like a response but aren't. A message marked "seen" on WhatsApp is not a response. An automated greeting that fires the moment someone messages — "Thank you for contacting us, we'll respond shortly" — is not a response either, however much it resembles one on a screenshot. Classify every enquiry in this window into one of four buckets: answered by a human who addressed the actual question, clarified with a follow-up question that moved things forward, routed to the right person or department, or simply lost, no human touch at all inside four hours.

The instinct here is to turn this into a performance review, and that instinct should be resisted, because it usually points at the wrong target. The person holding the business WhatsApp line is very often mid-consultation with someone standing in front of them, which is the job working as intended. The data bundle on the office phone finished at eleven and nobody noticed until three. The front-desk officer who handles enquiries also handles walk-ins, restocking, and the phone that never stops ringing, and stepped out for lunch at the exact hour three separate leads arrived. None of this is a discipline problem. It's a coverage problem, and a coverage problem is invisible until you trace forty-eight hours and watch three good enquiries die during the same twenty-minute lunch break, on three different days, because there was never a backup. The test isn't about the person. It's about whether a backup exists at all.

Hour four to twenty-four

The second window is the handoff into an actual booking, and this is where a good conversation quietly dies for want of a concrete next step. "We have slots this week" is a sentence, not an offer. It requires the customer to do the next piece of work — pick a day, pick a time, remember to follow up — and asking the customer to carry that weight is where a warm enquiry goes cold. A concrete offer names a specific slot, a specific location, and a specific action the enquirer can take in the next message, not a general availability statement that requires a second round trip to become real.

The other way this window loses people is more mundane and more common: the booking exists only in someone's memory. A staff member takes a name and a preferred day over the phone, intends to write it down, gets interrupted by the next call, and the intention never becomes an entry anywhere retrievable. Three days later nobody remembers the conversation happened, and the person who was supposedly booked never gets a reminder, because there was never a record for a reminder to be sent against. This is not a technology failure requiring a booking system purchase. It is often just a notebook nobody is using consistently, sitting three feet from where the phone rings.

Hour twenty-four to forty-eight

The final window is where "booked" gets tested against whether anyone actually shows up and whether the visit converts into money. Booked is not revenue. A confirmed appointment that nobody reminds the patient about the night before has a real chance of becoming a no-show, particularly across Lagos traffic patterns where an appointment made with genuine intention on Monday can lose to a go-slow on Wednesday morning without anyone changing their mind about wanting care.

And even attendance is not the end of the trace, because for anyone whose care runs through an employer scheme or an insurance product, "paid" and "claimable" are two different states that get collapsed into one in casual reporting. A patient can attend, receive care, and still represent unresolved revenue for the business if the claim hasn't been submitted, approved, or reconciled — categories that exist because the person receiving care, the organization authorizing it, and the entity actually paying for it aren't always the same party, particularly under employer-sponsored or group coverage arrangements. A completed visit that becomes a stuck claim looks identical to a completed visit that becomes cash, right up until someone checks the account three weeks later and finds one of them never actually arrived.

What the trace actually produces

Run this honestly for two days and you end up with something that looks like a funnel: thirty enquiries in, twenty-five getting a useful first response, fifteen reaching a concrete booking option, ten confirming, six arriving, five completing whatever the actual service was. The value of this map is not the final number. It's that the four gaps between those stages point at four different and largely unrelated problems — response coverage, offer clarity, the scheduling handoff, and whatever happens once the patient is physically in the building — and each of those problems has a different, specific fix. A business that reads this trace as a verdict on staff competence will fix the wrong thing, probably by hiring someone or firing someone, when the actual gap was a data bundle running out at the same time every afternoon.

One more thing the trace should never do, and this matters more than it sounds like it should: it should never require copying a patient's actual clinical notes into whatever sheet or CRM is being used to track the commercial funnel. The instinct to paste in "reason for visit: suspected malaria, third episode this year" because it's already sitting right there in the chat thread is an understandable one — it feels like useful context. Under Nigeria's data protection framework, health information is treated as sensitive personal data with real handling obligations attached, and the Nigeria Data Protection Commission has been direct about the risk of health-related details ending up somewhere they can be misused or fed into systems that discriminate against the person they describe. A commercial capture sheet only needs to know what stage an enquiry is at and how long it sat there. It does not need to know why the person is sick, and building the habit of leaving clinical detail out of the commercial trace from day one is far easier than trying to unwind it later.

The objection worth taking seriously

There's a reasonable objection to all of this, which is that forty-eight hours is a small sample and a business shouldn't restructure itself around two unusually good or unusually bad days. I take that objection seriously enough that I'd push back on anyone treating a single trace as a permanent verdict. A trace run during a public holiday week, or during the week a key staff member was out sick, will not describe the business's real operating rhythm — it will describe an outlier and mislead anyone who takes it as gospel.

But the objection proves less than it seems to. The value of the exercise isn't the specific numbers from one forty-eight-hour window. It's that running the trace at all forces a business to build the muscle of watching an enquiry all the way through instead of only watching the ones that happened to convert. A business that runs this trace quarterly, in an ordinary week each time, ends up with something more useful than a single perfect measurement: a rough sense of where the leak tends to sit, and whether it's moving. That's a lower bar than perfect measurement, and it's a bar almost nobody clears, because almost nobody runs the trace even once.

What thirty enquiries actually taught

Go back to that funnel — thirty in, five completing. The instinct on seeing a number like that for the first time is to treat it as bad news, proof the business is failing somewhere. It usually isn't proof of failure. It's proof that until someone traced it, this business had no idea what its own funnel looked like at all, which is a different and more fixable problem than the one the raw ratio seems to describe. A business that knows precisely where its twenty-five useful enquiries lost ten of themselves between hour four and hour twenty-four is a business that can fix exactly one thing next month. A business that only knows "about half" convert has nothing to fix, because it has nothing specific enough to aim at.

The forty-eight hours cost nothing to run beyond attention. What it replaces is a number invented on the spot — "about half," said with a straight face — with an actual map of where, specifically, an enquiry that wanted to become a patient stopped being one.

Notes on sources

• Health-related information classified as sensitive personal data under the Nigeria Data Protection Act 2023, with attendant risk of misuse: NDPC, "Disability Details are Sensitive Health Data".