Why one market’s conversion rate cannot be compared with another’s
Comparing conversion across markets assumes an enquiry means the same thing everywhere, that journeys are the same length, that cases are clinically similar and that money arrives the same way. None of that holds. Channel mix, travel, paperwork, specialty weighting, payment route and language cover set the rate. Compare a market against its own past, read stage rates, use cohorts, and keep case value in view.
Sooner or later someone asks for the slide. Markets down the side, conversion rate along the top, sorted best to worst, with a red cell for the market that is letting everyone down. It is the most requested view in international reporting and the least defensible. Comparing conversion across markets looks like management and usually functions as a distraction.
The objection is not that the numbers are wrong. Each market’s rate can be perfectly calculated. The objection is that the rates are answers to different questions. Two markets can have the same desk, the same doctors and the same pricing, and still convert very differently for reasons that have nothing to do with how well anyone is working.
This matters because the ranking has consequences. Budget moves to the market at the top, a desk executive is told to explain the market at the bottom, and a medical value travel business gets reshaped around an artefact of measurement. I have watched teams withdraw from markets that were performing exactly as their structure dictated.
So here is what actually differs, why it differs, and what I would compare instead.
What comparing conversion across markets assumes
A conversion rate is a fraction, and a fraction only travels between two places if the numerator and the denominator mean the same thing in both. The moment you put two markets side by side you are assuming four things: that an enquiry was defined identically, that the journey between enquiry and arrival has the same shape, that the cases are clinically similar, and that the money reaches you by a similar route. None of those survive contact with a real international desk.
The rate also assumes that the hospital controls the variable being judged. It controls some of it: speed of reply, quality of the opinion, clarity of the estimate, follow up. It does not control flight connections, document processing, currency controls, the school calendar or whether a referring doctor abroad is on leave. A ranking mixes the two and then hands the whole thing to someone as a performance review.
That is the difference between a diagnostic and a scoreboard. The funnel view I argue for in the international patient funnel is a digital product is useful precisely because it separates the parts you can redesign from the parts you can only plan around.
The denominators are not the same thing
The first problem is upstream of any market difference. The mix of channels varies by market, and each channel produces a different kind of enquiry.
A market where most demand arrives through open messaging channels will produce a denominator full of questions, price checks, relatives asking on behalf of someone who has not decided anything, and people who will never travel. A market where most cases arrive through a referring doctor or a partner will produce a denominator of pre qualified cases with reports already attached. The second market will convert far better, and it tells you nothing about the desk. It tells you that someone else did the filtering before the case reached you.
Duplication skews it further, and unevenly. In markets where a family will contact you on several channels at once, and where an uncle abroad and a cousin in India both make their own approach, one case can appear as several records. In markets where a single partner forwards a tidy list, it will not. Unless duplicates are merged into one case, you are comparing a market that was counted once with a market that was counted three times over.
Until you have a written definition of an enquiry applied identically everywhere, the comparison is arithmetic on sand. That definition work is the first half of measuring international demand honestly, and it is worth doing before anyone builds a comparison slide at all.
The journey is longer from some places
Once a family has decided, the distance between decision and arrival varies enormously, and almost all of it sits outside the hospital.
Travel itself differs: a land border crossing and a short flight are not the same commitment as a long journey with connections, and for some places flights are infrequent enough that a single delay moves the case by weeks. Paperwork differs in how long it takes to assemble and how far a family must travel to submit it. Moving money across a border is straightforward from some markets and slow and expensive from others, and a family that must arrange funds through several intermediaries will pause at exactly the point your dashboard calls a drop off. Seasons differ too: weather, school terms, religious calendars and agricultural cycles all move travel in ways a hospital marketing calendar never anticipates.
The rules governing entry and documents change from time to time and differ by country, so treat them as facts to be checked with the relevant authority and with your own international desk rather than as something a marketing plan can assume. What a growth leader needs to take from this is simpler: a longer journey produces a lower conversion rate in any fixed window, without anybody doing anything wrong.
Specialty mix changes the rate more than anything you do
Different conditions produce different journeys. An urgent cardiac case moves quickly because the family has no room to deliberate. A planned orthopaedic procedure can sit for a year while the patient weighs up time off work. Cancer care involves a sequence of opinions and often a decision to begin treatment at home first. Transplant cases carry approvals and a donor, which puts them on a timeline of their own.
So a market that sends you mostly complex or planned cases will show a slower, lower conversion rate than a market that sends you mostly urgent ones, permanently. If you insist on a comparison, compare within a specialty, and even then look at the stage pattern rather than the headline rate. Comparing a market weighted to oncology with a market weighted to elective surgery is comparing two different businesses that happen to share a desk.
Who pays sets the pace
The payment route may be the single largest structural difference between markets, and it is the one most often left out of the analysis.
A family funding treatment themselves moves at the speed of their own savings and their extended family’s contributions, and the estimate is the hinge on which everything turns. A patient whose employer is paying moves at the speed of an approvals process in another country. A case funded through a sponsoring institution may be nearly certain to arrive and still take months, because the money and the authorisation travel through several desks. A market dominated by any one of these will look completely unlike a market dominated by another.
There is a practical consequence for the desk: the work that improves conversion is different in each case. For self funding families it is an estimate that survives contact with reality, which I have written about in the estimate that survives arrival. For sponsored cases it is paperwork discipline and patient follow up with the sponsoring office. Ranking the two against each other tells the desk nothing about which of these to do.
Language and time zones shape the conversation
Where your team can reply in the family’s own language, in writing, at a time when they are awake, cases progress. Where the conversation depends on a shared third language and an available interpreter, more cases stall in the middle, and they stall quietly, because a family that is struggling to explain themselves usually stops replying rather than complaining.
Time zones do quieter damage. A market several hours away from you gets its answer at the start of your next working day, and by then the family has already heard back from somewhere else. Cover at the hours when enquiries actually arrive is a rota decision rather than a marketing one, and it moves conversion in a way that no campaign will.
Treat this as service design, not as a quirk of a market. It is a capability you can build: cover in the languages your demand actually arrives in, written replies that can be read aloud to an elder at home, and a response window that matches the hours when the family is awake rather than the hours your office is open. Until that cover exists, a market with poor language cover will convert worse than a market with good cover, and the number is measuring your staffing, not the families.
What to compare instead
Drop the ranking and use three comparisons that hold up.
First, a market against its own past. Same definitions, same structure, so a movement means something. This is the comparison that catches a new competitor, a service problem or a route drying up.
Second, stages within a market. Where do cases stall here, and is that stage moving. Stage rates are diagnostic because each one maps to an owner: first reply to the desk, opinion to the clinical coordination, estimate to pricing and the desk together, travel to logistics support. A single blended rate has no owner, which is why it produces blame instead of action.
Third, cohorts rather than calendar months. Follow the enquiries created in a period forward and let them mature, and mark clearly which cohorts are still open. Slow markets are punished by calendar reporting for being slow, which is circular.
There is a fourth comparison worth building towards, which is a test inside a single market. Change the reply template, add cover at a different hour, put a written estimate where a verbal one used to go, and watch the same market with and without the change. That is the only setting where the structural factors hold still long enough for a difference to mean what you think it means.
Alongside all three, keep the value side visible. A market that converts modestly but brings complex, longer staying cases can be worth more than a market that converts briskly, which is why I keep cost per arrived international patient and case value on the same page. The domestic version of this argument, in enquiry to appointment, holds here too: a conversion rate is only meaningful next to the definition that produced it.
When the board asks for the league table anyway
You will still be asked, and refusing to answer is not a strategy. Give them something better than a ranking.
What I present is a page per priority market: the direction of travel against its own history, the stage where cases stall, the structural constraints that shape what is possible there, and the one thing being changed this quarter. Then a single line on what would have to be true for the market to deserve more investment. Boards respond well to this because it answers the question they were actually asking, which is where to put the next rupee of effort, not which colleague to be disappointed in.
Bring the constraint page into the meeting rather than promising it later. Constraints written down once tend to end the same argument every quarter, and they make it obvious when something has genuinely changed in a market rather than in the way the market was counted.
If a ranking is unavoidable for some reason, rank on movement rather than on level: which markets improved against themselves. That at least rewards work rather than geography.
The reporting change I would make first
Take the comparison slide out of the monthly pack and replace it with stage rates by market and a cohort view. Say why, in one sentence, in the pack itself, so the change is understood as a correction rather than as a team avoiding scrutiny.
Then do one piece of homework per market: write down the enquiry channel mix, the dominant payment route, the specialty weighting and the language cover. One page, no figures needed, just the shape of the market. Half the arguments about conversion end the moment those four facts sit next to the rate, because everyone can see at once that two markets were never running the same race.
Finally, agree with the leadership team what a market review will judge. My preference is direction, stage movement and the quality of the plan. Not position in a table that was never measuring what people thought it was.
Questions people ask
It means putting the enquiry to arrival rate of one source market next to another and reading the difference as performance. The practice assumes that an enquiry means the same thing in both markets, that the journey between enquiry and arrival has the same shape, that the cases are clinically similar and that the money arrives by a similar route. Those assumptions rarely hold for international patient demand.
Not at all. They are useful within a market and over time, and they are very useful at stage level, where each stage maps to someone who can fix it. What breaks is the comparison between markets, because the rate then mixes things the hospital controls, such as reply speed and estimate quality, with things it does not, such as travel routes, paperwork timelines and payment mechanisms.
Because different channels produce different enquiries. Open messaging channels bring questions, price checks and relatives asking on behalf of someone undecided. Referring doctors and partners bring cases that have already been filtered, often with reports attached. A market fed mostly by referrals will convert better regardless of how the desk performs, since the filtering happened before the case reached the hospital at all.
Urgent conditions move fast because families have little room to deliberate. Planned procedures can sit for a long time while a patient weighs time away from work. Cancer care usually involves several opinions and sometimes treatment started at home. Transplant work carries approvals and a donor timeline. A market weighted towards complex or planned cases will show a slower rate permanently, whatever the desk does.
Because it sets the pace. Families funding treatment themselves move at the speed of their own savings and the estimate becomes the hinge. Employer funded cases move at the speed of an approvals process abroad. Institutionally sponsored cases may be almost certain to arrive and still take months as authorisation passes through several offices. Each pattern produces a different rate and needs different work from the desk.
Ask for a page per priority market showing direction against its own history, the stage where cases stall, the structural constraints in that market and the one change being made this quarter. That gives you something to decide about. A ranking gives you someone to be disappointed in, which is a poorer use of a review meeting and tends to produce stop start investment.
Tell them the rates answer different questions and offer movement instead of level: which markets improved against themselves, and why. Then show the constraint page for each priority market. Boards are usually asking where the next effort should go rather than which market is best, and a structured answer to the real question lands better than a table that quietly misleads everyone who reads it.
No. Challenge is sharper when it is aimed at a stage. If first replies are slow, that is the desk and it can be fixed this month. If estimates keep changing, that is pricing discipline. If cases stall after agreement, it is usually logistics support. Stage level accountability is harder to argue with than a blended rate, and it gives the team somewhere to start.
Where a team can reply in the family’s own language in writing, at a time when they are awake, cases progress. Where the conversation depends on a shared third language and an interpreter who may not be available, more cases stall in the middle and do so silently. That difference is a staffing decision inside your hospital, so the rate is partly measuring your own cover rather than the market.
Because international journeys are long and the length varies by market. Calendar reporting divides this month’s arrivals by this month’s enquiries, which compares two unrelated groups and penalises slow markets for being slow. Cohorts take the enquiries created in a period and follow them forward. The current cohort always looks poor, so mark clearly which are open and let the closed ones carry the argument.
Look at acquisition cost and case value together rather than at conversion alone, and insist on seeing definitions alongside rates. A market that converts modestly while sending complex, longer staying cases can be worth considerably more than one that converts briskly. Also ask how long a market has been worked, since referral relationships in medical travel compound slowly and early numbers understate them.
Stage timestamps on every case, a cohort view built on first contact date, and fields that hold channel, payment route and specialty so the rate can be read with its context attached. Nothing exotic is required. The difficulty is agreement on definitions rather than technology, and the build should wait until those definitions are signed by the desk, marketing and finance.
The mechanical change can happen in a month. The cultural change takes longer, because a ranking is easy to read and the replacement asks people to hold more context. Expect a couple of review cycles before the new view is trusted, and keep the old slide available for a short while so anyone can see for themselves how differently the two versions describe the same quarter.
Withdrawing from a market on the strength of a ranking. The market that looks worst is very often the one with the longest journey, the most complex cases or the slowest payment route, and those are also markets where relationships compound over years. Before exiting anything, write down the four structural facts about that market and see whether the rate was ever telling you what you assumed.
