Enquiry to appointment: the number that matters

If I were given one metric to run a hospital group’s growth function on, it would not be cost per lead, reach, or share of voice. It would be the proportion of identified enquiries that become booked appointments, measured by unit and by service line, every week.

It is the least glamorous number in the deck. It involves no campaigns, no creative and no platform. It is also the only place where an improvement compounds across everything else you are doing, because every channel you run feeds into it and every point of improvement raises the yield of all of them simultaneously. Doubling your traffic doubles your media cost. Improving enquiry-to-appointment improves the return on media you have already bought.

And in most groups it is badly understood, unowned, and worse than anyone believes — not because people are careless but because the losses happen in places nobody is looking at, in fifteen-second increments, distributed across shifts.

Where enquiries actually die

I have sat through enough call audits to have a reliable list. In rough order of volume lost:

• Never contacted at all. The enquiry arrived in a system nobody was watching — a form submission to an inbox, a chat transcript, a missed call out of hours with no callback queue, a lead in a platform dashboard the agency checks twice a week.
• Contacted too late. The patient has already spoken to two other hospitals. You are now a price comparison rather than a choice.
• Contacted once. One ring-out counted as an attempt and the record closed. Patients in a consultation are not available at the moment you call; anyone who has tried to reach a working adult in an Indian metro between ten and seven knows this.
• Language mismatch. The agent and the caller do not share a comfortable language, so the conversation stays transactional, trust is not established, and the booking does not happen. This loss is recorded as a bad lead.
• No slot to offer. The agent can only offer a date eleven days out. The patient says they will call back and does not.
• Wrong specialty routing. The caller describes symptoms, the agent guesses, the booking lands with the wrong consultant, and the patient who attends a futile consultation does not return.
• Money not discussed. Nobody told them the consultation fee, whether cashless applies, or whether their TPA is empanelled. They go where someone answered that on the first call.
• Booked and not honoured. Technically a separate stage, practically the same failure — an appointment with no confirmation and no reminder is a coin toss.

Notice that exactly one of these is a marketing problem and none of them is a creative problem.

Response latency is the whole ballgame

Every study of enquiry handling in any industry reaches the same conclusion and healthcare is not exempt: the probability of reaching and converting an enquirer collapses with elapsed time, and the collapse is steepest in the first hour. In healthcare there is an additional mechanism. The anxious family member is calling several hospitals in one sitting. The first competent conversation frequently ends the search.

So measure latency, but measure it properly. The aggregate average will tell you nothing useful because it is dominated by the bulk of enquiries that arrive during staffed hours and are handled fine. Split it: by hour of arrival, by day of week, by channel, by unit, by service line. What you will find — I have found this in every group I have looked at — is that the problem is concentrated. Two shifts. Weekends. The hours after the last shift ends and before the first begins. A single channel whose enquiries route to a queue nobody owns.

Fixing two shifts is a tractable management problem. Fixing “our response time is poor” is not.

Contact centre design decisions that matter more than scripts

Most contact centre improvement programmes start with the script. The script is maybe the fifth most important thing.

Out-of-hours coverage. Healthcare enquiries do not respect office timings, and a substantial share of digital enquiry volume arrives in the evening and at night, when people are not at work and are worrying. If your centre closes at eight, you are systematically losing your most motivated enquirers. Extended hours, a night shift for high-value service lines, or at minimum a first-thing-in-the-morning callback queue that is worked before fresh enquiries.

Language routing as a design requirement, not a staffing accident. In a multi-state group this means knowing the language mix of each catchment and rostering against it, plus a transfer path that works rather than a promise to call back.

Specialisation by service line. A generalist agent handling oncology enquiries cannot do it well. Oncology, fertility, transplant and paediatric calls need agents who have heard the questions before and who are allowed to take eleven minutes rather than four. Average handling time as a universal target destroys exactly these calls, and they are your most valuable.

Attempt discipline. A defined contact cadence — multiple attempts across different times of day over forty-eight hours, with a WhatsApp or SMS touch between calls — and a rule that an enquiry cannot be closed as unreachable before the cadence is complete. Then audit the closures, because the quickest way for an agent to hit a productivity target is to close records.

Booking authority. The agent must be able to book a real slot in the real system on the call. If they can only “note your request and have someone call you back”, nothing else on this list matters.

Who they report to. A contact centre reporting into a shared-services function optimises for cost per call. One reporting into growth optimises for appointments. I have seen the same team’s conversion change materially on the strength of nothing but a change in what their manager was measured on.

Doctor-slot availability is a marketing constraint

This is the part that is hardest to say inside a hospital, so say it with data rather than opinion.

If a consultant’s next available slot is consistently more than a few days out, marketing spend directed at their service line is largely wasted, and the enquiries it generates will show up as poor conversion that the contact centre gets blamed for. The binding constraint is clinical capacity. Until it moves, the funnel cannot.

There are real levers, and they belong to operations with marketing’s evidence behind them: a second consultant in the service line, dedicated new-patient slots protected from follow-up bookings, teleconsultation as a bridge for the second opinion, an extended OPD on the two days that are always full, and junior-consultant first consultations with a senior review. What you cannot do is fix it with better copy.

Put the slot-availability chart next to the conversion chart in the weekly review. Once both lines are visible in the same room, the conversation changes from whose fault it is to which constraint we are choosing to relieve.

The weekly meeting that actually moves it

None of this improves through a project. It improves through a recurring meeting with a fixed agenda and the right people in the room, and the meeting is the real intervention.

Forty-five minutes, weekly, same slot. Present: the contact centre lead, one unit representative on rotation, the digital lead, and someone from operations who can speak for slots. Not the agency, initially — add them once the data is trustworthy. The agenda is four items and does not change: last week’s enquiry-to-appointment by unit and service line; the latency breakdown with the worst two cells named; ten call recordings, listened to together, chosen by the contact centre lead rather than by marketing; and one committed action with an owner and a date.

The call recordings are the part people want to drop and the part that changes behaviour. Hearing an agent fail to answer a TPA question, or hearing a frightened caller be offered a slot nine days out, does more in four minutes than a quarter of dashboards. I would also insist the unit representative rotates, because the unit that attends the meeting improves for about six weeks afterwards regardless of what was decided.

Why this beats buying more traffic

The arithmetic is not subtle. Improving enquiry-to-appointment raises the output of every enquiry you already generate, at no additional media cost, permanently, across every channel and every unit. Buying more traffic raises output linearly with spend, usually at a rising marginal cost because you are moving into less qualified inventory.

There is a second and less obvious benefit. A group that handles enquiries well can afford to buy traffic others cannot, because its yield per enquiry is higher. The conversion work is what makes the media work affordable, which is the reverse of how most groups sequence the investment.

And a third: the conversion fixes are largely operational discipline rather than capital. Extended hours, a rostering change, a contact cadence, a reporting line. None of it requires a board approval of the kind media budgets do.

If you are starting this next quarter

1. Establish the number. One agreed definition of identified enquiry, deduplicated, and one of booked appointment, reconciled against the HIS. Expect the first honest number to be lower than the number currently circulating.
2. Break latency down by hour, day, channel and unit for four weeks before changing anything. Find the two shifts.
3. Close the unwatched channels. Every inbox, chat queue and platform dashboard that holds enquiries must flow into one queue with an owner. This alone usually recovers real volume.
4. Set and enforce a contact cadence and audit closures weekly for a month.
5. Give agents real booking authority in the live slot system.
6. Roster for language against each catchment’s actual mix.
7. Publish the slot-availability overlay in the same review as conversion, and cut spend where capacity binds.
8. Add confirmation and reminder touches for every booking, then measure honoured appointments separately from booked ones.
9. Only then ask for more media budget — and ask for it armed with a higher yield per enquiry, which is a much easier case to make.

What to watch out for

Two things go wrong when a group starts taking this seriously. The first is that conversion becomes a target and targets get gamed — agents book appointments the patient did not really agree to, and your honoured rate falls while your booked rate climbs. Measure both or you will optimise a fiction. The second is that the contact centre becomes the designated problem, which is unfair and counterproductive, since a good share of the loss is slot availability and channel plumbing. Name the constraint honestly each week and the team will work with you. Blame them and they will manage the metric instead.

More traffic is something you can buy. A conversation that reassures a frightened family at nine in the evening is something you have to build. Only one of those is a moat.

Questions people ask