Search-to-Booking Leakage: Six Places Healthcare Providers Lose Qualified Patients

01

Executive thesis

That space is not a single conversion event. It is a chain of six separate failures, each owned by a different part of the organization, each measured by a different team and usually reported to nobody in aggregate. Marketing reports traffic. The website team reports sessions. The call centre reports answered calls. Operations reports utilization. No one reports the number that matters, which is the proportion of qualified searchers who end up sitting in a consultation room.

The Search-to-Booking Leakage Map exists to make that chain visible and to force one shared calculation across marketing, website, booking and clinic operations.

02

Why more traffic can hide a broken healthcare conversion funnel

Small losses at several stages compound into a large final gap. This is arithmetic, not opinion, and it is the reason healthcare conversion funnel diagnosis fails when it is treated as a landing-page exercise.

Six stages at 80 percent each do not deliver 80 percent. They deliver 26 percent. Six stages at 60 percent deliver 5 percent. A clinic can therefore improve every individual stage to a level that sounds respectable in a monthly report and still convert fewer than one in twenty qualified searchers into an attended appointment.

Traffic growth hides this because it moves the only number most healthcare leaders see. Sessions rise, inquiries rise slightly, attendance stays flat and the conclusion drawn is that the market is competitive. The more accurate conclusion is usually that additional demand is being pushed through the same constricted chain, and that the cost of each attended appointment is quietly rising.

The Saudi context makes this more consequential rather than less. Digital health access is now normal behaviour, not an emerging channel. The Ministry of Health reported more than 16 million virtual appointments and medical consultations delivered during 2025. That is a national public-sector figure and it does not describe private-clinic booking behaviour, but it establishes the baseline expectation patients now bring to every provider they encounter. Patients who complete a government health transaction on a phone in two minutes do not become tolerant of a private clinic that answers booking requests during office hours only.

03

The six leakage points from discovery to attendance

The map separates six gates. Each has a distinct failure mode, a distinct owner and a distinct fix. Grouping any two of them together is what produces the vague diagnosis that "conversion is low".

1. Discovery loss. The patient searches and the provider does not appear. This includes absence from the local map results, absence from non-branded specialty and symptom queries, and absence from the directories and aggregators that occupy the first screen. Discovery loss sets the ceiling for everything below it. Clinical reputation does not close this gap on its own, which is the subject of the invisible specialist problem, and measuring it requires local visibility share rather than a single ranking.

2. Relevance loss. The provider appears but the result does not match the question being asked. A patient searching for a specific procedure lands on a generic department page. A patient searching in Arabic reaches an English page. A patient searching for a named condition finds a page that lists services but never names the condition. The click happens or it does not, and if it happens the page fails to confirm that the patient is in the right place.

3. Trust loss. The page is relevant but the evidence is thin. No named clinician, no credentials in a form a patient can interpret, no explanation of what the consultation involves, no public evidence of other patients' experience. This gate is where high-consideration specialties lose most heavily, because the patient's real question is not what the clinic does but whether this specific doctor is the right person to see.

4. Eligibility and access loss. The patient trusts the provider and still cannot establish whether the service applies to them. Which insurers are accepted. Whether a referral is required. Whether the clinic is open after work. Whether there is a female practitioner available. Whether anyone will speak their language. These are not marketing questions, they are operational facts, and they are absent from most Saudi clinic websites.

5. Booking loss. The patient decides to act and the booking path defeats them. Account registration before a first appointment, a form that fails silently, a phone line that rings out, a WhatsApp number with no reply for six hours, a channel switch from web to phone at the final step. This is the gate measured by the Booking Friction Index.

6. Attendance loss. The appointment is booked and the patient does not arrive. In Saudi Arabia this is a large and well-documented loss. A 2025 cross-sectional study of adult rehabilitation outpatients in Riyadh notes that Saudi studies have reported no-show prevalence between 11 and 30 percent depending on setting and service, against a global range of 12 to 80 percent.

Marketing cannot compensate for gates four, five and six. No amount of additional spend at gate one repairs an unanswered phone or an unclear insurance policy. It simply increases the volume of patients arriving at the same blockage.

04

How to calculate compounded journey loss

The Search-to-Booking Leakage Map is expressed as six pass rates multiplied together.

Attended appointment rate = D × R × T × E × B × A

Where:

Three rules govern its use.

First, measure each gate separately. Qualified inquiry rate, booking completion and show rate are three different numbers with three different owners. A single blended conversion figure cannot be acted on because it does not identify who is responsible.

Second, identify the binding constraint, which is the earliest gate materially below its benchmark. Fixing the earliest severe failure lifts the input volume of every gate below it, so its effect is multiplicative rather than additive.

Third, identify the most expensive gate, which is rarely the same one. Cost weighting matters because acquisition cost accumulates as the patient moves down the chain. A patient lost at gate six has consumed the full cost of discovery, content, trust building, booking handling and a reserved clinical slot.

Wasted value at gate i = (patients entering gate i) × (1 − pass rate at gate i) × (cumulative cost per patient at entry to gate i)

The percentage loss tells you where the chain is weakest. The cost weighting tells you where the money is burning. Prioritization requires both.

05

Booking friction versus demand quality

The most common misdiagnosis in healthcare growth is to attribute a late-stage failure to early-stage demand quality. Booking completion falls, and the conclusion drawn is that the traffic is unqualified.

Sometimes that is correct. Broad awareness campaigns and price-led promotions do generate inquiries from people who were never going to attend. But demand quality and booking friction produce different signatures, and they can be separated.

Friction is indicated when losses concentrate at a specific step of the booking path, when the loss is worse on mobile than desktop, when it is worse in Arabic than English, when it is worse outside working hours, and when patients who fail online succeed by phone. Demand quality is indicated when losses are distributed evenly across the path, when the same pattern holds across devices and hours, and when patients who reach a human still decline to proceed.

The same discipline applies to attendance. The reflex answer to no-shows is to send reminders. The Saudi evidence supports that reflex only partially. A randomized controlled trial across three outpatient clinics at a Saudi hospital, with 1,499 patients enrolled between April and June 2011, found SMS reminders reduced non-attendance in general medicine, from 39.8 percent to 26.3 percent, and in neurology, from 43.9 percent to 29.3 percent. In obstetrics and gynaecology the difference was not statistically significant, at 26.6 percent against 27.9 percent.

The operational reading is that reminders address forgetting, and forgetting is a large but not universal cause of non-attendance. Where the cause is scheduling lead time, transport, cost uncertainty or ambivalence about the referral, a reminder changes nothing. Attendance loss requires diagnosis before it receives a tool.

06

A worked leakage model for a specialist clinic

The following scenario is illustrative. The pass rates are plausible planning assumptions for a single-site private orthopedic clinic in Riyadh and they are not DEMA measurements. Any clinic applying this model should replace every figure with its own data.

Assume 12,000 relevant monthly searches in the catchment area and a monthly digital marketing spend of SAR 30,000.

End-to-end conversion is 0.87 percent. Cost per attended appointment is SAR 30,000 divided by 104, or approximately SAR 288.

Now compare three single-gate interventions.

Discovery produces the largest volume gain because it is the earliest constrained gate and every downstream gate inherits its output. That is the multiplicative effect of fixing the binding constraint.

But volume is not the only lens. Apply cost weighting to the same baseline. The 35 booking attempts lost at gate five each carry the accumulated cost of everything above them, roughly SAR 130 per patient reaching that point on these assumptions. The 35 confirmed appointments lost at gate six carry that cost plus a reserved clinical slot. For context on the operational value of that slot, the Ministry of National Guard Health Affairs, in a case study submitted to the OECD Observatory of Public Sector Innovation and published in June 2024, estimated the average cost of a missed appointment at 1,315 Saudi riyals. That figure comes from a large public tertiary system and should not be applied directly to private-clinic economics, but it makes the point that a late-stage loss is not equivalent to an early-stage loss of the same percentage.

The correct sequencing conclusion is therefore not "fix discovery" or "fix booking". It is that discovery sets the ceiling and should be resourced for growth, while booking and attendance determine the efficiency of everything already being spent and should be resourced for margin. A clinic that increases discovery without repairing booking pays more per patient for a larger volume of the same leak. The relationship between these losses and blended acquisition cost is developed further in the healthcare acquisition equation.

07

The Search-to-Booking diagnostic

Six questions, six owners, six thresholds. Answer them with data rather than impression.

Two conditions should trigger escalation regardless of the individual scores. The first is any gate where the Arabic pass rate is materially below the English pass rate, because that is an access failure rather than a marketing failure. The second is any gate where no owner can be named, because an unowned gate does not improve.

08

A prioritized leakage-reduction plan

Fix now, within thirty days. Publish eligibility facts on every service and clinician page, covering accepted insurers, exact location with parking guidance, opening hours including evenings, languages spoken and whether a referral is needed. Remove account registration from first-time booking. Establish a maximum response time for every inbound channel and measure it. Confirm every booking immediately in the channel the patient used, in the language they used.

Build next, within one to two quarters. Instrument the six gates so each produces a monthly number with a named owner. Rebuild clinician profiles as decision pages rather than biographies. Close the Arabic and English parity gap at the gates where the diagnostic shows the widest divergence. Introduce a reminder sequence, then measure its effect by specialty rather than assuming it works, because the Saudi trial evidence shows the effect is real in some clinics and absent in others.

Measure continuously. Report the six pass rates and the end-to-end attended appointment rate as a single monthly view seen by marketing, operations and clinical leadership together. Track cost per attended appointment rather than cost per lead. Re-run the discovery and booking audits quarterly, since both degrade without maintenance.

The connective logic across all six gates, and the reason a patient who fails at one gate rarely returns to an earlier one, is set out in the DEMA 7-Touch Patient Decision Journey.

09

Methodology, limitations and compliance boundary

This article presents a framework and a worked planning scenario. It does not present original DEMA research findings.

The worked model uses illustrative assumptions for a single-site private orthopedic clinic in Riyadh. The pass rates are planning values selected to demonstrate compounding arithmetic. They are not measurements, not benchmarks and not client results. Any provider using the model should substitute its own data before drawing conclusions.

External evidence is labelled by geography and method. The virtual consultation figure is Saudi Ministry of Health national public-sector volume for 2025 and does not describe private-clinic booking behaviour. Saudi no-show prevalence of 11 to 30 percent is reported in a 2025 cross-sectional study published in Health Science Reports, drawing on prior Saudi literature across varied public and hospital settings. The SMS reminder results come from a randomized controlled trial conducted in 2011 at a single Saudi hospital across three specialties, published in the Journal of Taibah University Medical Sciences, and the age of the trial should be weighted accordingly. The SAR 1,315 cost per missed appointment is self-reported by the Ministry of National Guard Health Affairs in a case study published by the OECD Observatory of Public Sector Innovation in 2024, and reflects a public tertiary system rather than private-clinic economics. International comparison data from Tebra's sixth annual Patient Perspectives survey, an online survey of 3,964 United States adults fielded on 29 June 2025, reports that 65 percent of patients would switch providers for a better overall digital experience, up from 55 percent in 2024. That is a United States commercial survey and no inference about Saudi patient behaviour should be drawn from it.

Nothing in this article constitutes medical, legal or regulatory advice. Providers implementing eligibility disclosures, reminder programmes or booking data collection should confirm compliance with Saudi health advertising rules and applicable personal data protection requirements before deployment.

Last reviewed August 2026. Review cycle annual.

Request a Search-to-Booking Audit. DEMA measures the six gates for a single site or a multi-location group and returns the binding constraint, the cost-weighted priority and a sequenced remediation plan.

Continue exploring

Related reading from the same authority programme.