A family of four walks out of your office having paid ₹1,42,000 for nine nights in Kerala. The consultant who closed it is pleased. The customer is pleased. Four months later your accountant is still matching a houseboat invoice against a hotel bill against three UPI screenshots from a driver in Munnar, and nobody in the office can tell you within ten thousand rupees what that trip actually earned.
This is the ordinary condition of a travel agency. Price is set on a Tuesday; cost is discovered over the following quarter. In between, the business runs on the comforting arithmetic of turnover — the season was busy, so the season was good — which holds right up until the bank balance disagrees.
The fix is not a better accountant. It is treating every rupee you commit to a vendor as a record attached to a trip, written at the moment you commit it, rather than at the moment somebody sends you a bill.
Price is a decision. Margin is a discovery.
Package price is set by a handful of forces that have almost nothing to do with cost: last season's sheet with a percentage on top, a round number the customer can say out loud, and whatever a competitor quoted on WhatsApp an hour ago. It is a single decision, taken once, by one person, usually in a few minutes.
Cost is the opposite in every respect. It is fifteen or twenty separate commitments, made by three or four people, spread across six weeks — some before the customer has paid a rupee, some after they have come home and posted the photographs. No single person sees all of them. There is no moment at which the whole cost of a trip exists in one place, which is precisely why the margin never does either.
There is a quick test for this. Ask an agency owner which package was their most profitable last year. Most will name their best-selling one. Those are different questions, and in a lot of agencies the honest answer to the first is a package that sold nine times, not ninety.
The five lines that eat a package
Every itinerary, however exotic, decomposes into the same cost families. It helps to name them, because you can only chase what you have named.
- Accommodation — room category and per-night rate, seasonal and peak-date surcharges, extra beds, and whether you booked room-only or with meals.
- Transport — vehicle class, kilometres run against kilometres quoted, driver allowance, tolls, parking and interstate permits.
- Guides and escorts — per-day fees, the premium for a language you promised, and the tour manager's own seat, room and food.
- Activities and entries — tickets, ferries, safaris, adventure operators, and the small on-ground payments a guide makes and claims back in cash.
- Permits, insurance and paperwork — entry permits for restricted areas, monument fees, visa handling charges and travel insurance.
The first two are almost always quoted. The last three are almost always assumed — carried in somebody's head from the last time this itinerary ran. Assumed costs are exactly the ones that move. A safari operator revises rates in April, a monument changes its fee, a long weekend adds a surcharge nobody circulated. You find out when the invoice lands, on a trip you priced in January.
A vendor booking is a record, not a phone call
The hotel is booked by calling a sales manager you have known for nine years. That relationship is one of the most valuable assets in the business, and nothing here suggests replacing it. The problem is not the phone call. The problem is that the phone call is the only evidence the booking exists.
A vendor booking, written down properly, carries six things: which trip it belongs to, which vendor, what service, what dates, what it costs, and what state it is in — requested, confirmed or completed. Requested and confirmed are not the same word. An unconfirmed room forty-eight hours before departure is a different kind of emergency from an unpaid one, and only one of the two can be solved with money.
The roster underneath it matters just as much: hotels, transporters, guides, caterers and activity operators, each with contact details, the services they actually provide, and a GSTIN. That last field is not bureaucracy. A supplier you cannot identify on paper is an input credit you cannot claim, and on a trip you are billing with your own GST invoice, that is real money left on somebody else's table.
A trip cost you cannot see until the season ends is not a cost. It is a surprise with a date on it.
The invoice that arrives after the trip is over
Travel has a timing problem that most businesses do not. Your customer pays before they travel. Your vendors bill you afterwards — sometimes weeks afterwards, on a credit cycle that has nothing to do with your departure dates. Revenue and cost for the same trip land in different months, and often in different quarters.
In that gap, things happen. The driver runs three hundred kilometres more than the quote. The family adds two nights in Alleppey on the ground. The guide buys entry tickets in cash and claims them back a fortnight later. A hotel raises a damages claim. None of this is a scandal; all of it is invisible until somebody remembers.
The discipline that fixes it is small. Record the expected cost when you make the vendor booking, and update it when the actual invoice arrives. That single habit turns every trip from provisionally profitable into measurable, and turns the difference between expected and actual into a number you can look at — per trip, per vendor, per package — instead of a vague feeling that Kerala was harder work this year.
Group departures: fixed costs against variable pax
A group departure is a different animal from a family booking, and pricing it per head hides that. The coach costs the same whether fourteen or twenty-two people board it. The tour manager costs the same. The guide costs the same. Hotels and meals scale with the number of travellers; very little else does.
Which means margin per passenger is a curve, and the shape of that curve is the most important number in the whole departure. Somewhere on it sits the seat at which the trip breaks even. Agencies that have split fixed from variable know exactly where that seat is. Agencies that priced from a per-head sheet find out in the final week, when two families withdraw and the coach still has to be paid for.
This is where the state of each booking earns its keep. A departure showing fourteen confirmed and six quoted is a completely different commercial position from twenty confirmed, even though both fill the coach on paper. Moving every booking honestly through enquiry, quoted, confirmed, partially paid, paid, travelling and completed is not process for its own sake. It is the difference between knowing you have a departure and hoping you do.
The discount given at the closing table
Almost every trip is sold slightly below its sheet price. Eight thousand off to close today. A free airport transfer. A room upgrade promised verbally and paid for silently. These giveaways are made by the person with the least visibility into the trip's cost, under the most pressure, at the last possible moment.
None of that is wrong — consultants have to be able to close. But the giveaway has to land somewhere you can see it. A discount written as an explicit line on the invoice, alongside the package, add-on and event lines with the CGST and SGST split intact, is a discount you can total at the end of the season and set against the business it bought. A quietly lower price is just a number nobody can question.
The same applies to what gets given after the trip. A goodwill credit for a bad hotel room is a genuine cost of that departure and belongs against it, not in a general expenses bucket where it becomes anonymous. So does the commission you paid whoever referred the booking.
Per-trip margin, as a report you can run on a Tuesday
Strip it all down and a trip has three numbers. What it was sold for, net of discount. What has actually been collected against it. What has been committed to vendors for it. Everything else — the itinerary, the thank-you message, the review — is commentary on those three.
Once vendor bookings hang off the trip and invoices hang off the booking, both sides live in the same place: an outstanding-invoices view telling you what customers still owe, a vendor-payable view telling you what you owe, and revenue by booking underneath. The margin question stops being a fortnight of reconstruction and becomes something you run before lunch.
It also changes the conversation with your team. “You closed eleven trips last quarter and two of them lost money” is a coaching conversation. Without the numbers it is only an accusation, so it never gets had — and the same two trips get sold again next season at the same price.
Repricing next season with numbers instead of nerve
Rate sheets arrive from hotels and transporters before every season, and most agencies do the same thing with them: pass on the increases they were told about, absorb the ones they were not, and move the headline price by whatever feels defensible.
With a season of per-trip history the conversation gets specific. This package's cost drifted while its price barely moved. This vendor raised twice in one year and generated a complaint both times. This itinerary day is the second most expensive thing in the trip and the one customers mention least. Each of those has an obvious action — reprice it, change the hotel, drop the day, or stop selling the package.
The reverse is just as useful. Almost every catalogue contains one or two quiet, unglamorous packages that earn far more per departure than the ones the office is proud of. They are usually under-sold, for no better reason than that nobody knew.
None of this requires a finance department. It requires that money going out is written down against the trip it belongs to, at the time it is committed, by the person committing it. Do that for one season and the question at the top of this page stops being rhetorical.