Hotel Overbooking: When It Pays and When It Backfires

Overbooking is neither good nor bad; what matters is whether anyone did the maths. Most people hear "overbooking" and think of airline passengers bumped at the gate. Hotels face something similar but not identical: instead of a seat there is a room, and instead of a passenger at the gate there is a guest arriving at the front desk near midnight to hear "I'm sorry, we're full." This article is written for hotel operators: what overbooking means in a hotel, when it is a sound revenue decision, and when it becomes an expensive mistake that ends up in your reviews.

One distinction first. Hotels experience two kinds of overbooking, and using one word for both muddles the discussion. The first is strategic overbooking: deliberately selling beyond capacity, based on history, to cover guests who will not turn up. The second is accidental overbooking: nobody decides anything, yet the same room is sold twice because of a stock delay between channels or a booking that was never entered. The first is a revenue management tool. The second is a gap in your systems.

What Hotel Overbooking Is, and How It Differs from Airlines

In a hotel, overbooking means accepting more reservations for a given night than there are rooms available to sell. A 60-room hotel taking 63 bookings for Friday night has oversold by three rooms. The logic is the same as an airline's: not everyone with a reservation shows up. Some become no-shows, some cancel on the day, some cut their stay short.

The difference lies in the consequences. An airline can offer the next flight within a set compensation framework. A hotel has no "next departure"; a guest it cannot accommodate has to be placed in another property, taken there, and in most cases the price difference is on the hotel. The industry calls this "walking" a guest. The whole overbooking decision rests on the balance between the cost of walking a guest and the revenue lost on an empty room.

Walking a guest is also more than a cost line. Failing to provide a stay the guest had confirmed means the hotel has not delivered on its commitment, and that responsibility stays with the hotel regardless of how carefully the oversell was calculated.

Strategic Overbooking: How to Size the Buffer

The raw material for strategic overbooking is no-show data. As we covered in How to Reduce No-Shows at Your Hotel, you lower the no-show rate first, and only the part you cannot remove becomes input for the overbooking decision. Never reverse that order.

An average is not enough; you need the distribution. Take the last 20 Friday nights at a 60-room city hotel. Each night sold out, but no-shows, same-day cancellations and early departures left some rooms empty (illustrative figures):

On an average Friday this hotel ends up with 3.5 empty rooms, so the instinct is to oversell by three or four. The right answer depends on the cost of walking a guest. Say the average room rate is 3,500 TL. Walking one guest might look like this: a comparable room elsewhere at 4,500 TL that night, 800 TL for transport and 1,200 TL of compensation on a future stay, 6,500 TL in total, and no revenue from the walked guest that night. Add even a small allowance for hidden costs such as a negative review and a lost repeat booking, and you reach 7,000 TL.

The decision rule follows from those two numbers. Each extra room sold earns 3,500 TL if it fills a room that would have stood empty, and costs 7,000 TL if it does not. Because the downside is twice the upside, the next oversold room is only worth selling if there is roughly a two-in-three (67%) chance or better that at least that many rooms will be empty. In the table, the third room clears the bar at 70%; the fourth does not at 50%. The right buffer is three rooms, below the 3.5 average.

Two practical consequences follow. First, the more expensive a walk is, the smaller the buffer: at a five-star property, or on nights when alternative rooms are scarce, it approaches zero. Second, the buffer has to be set night by night. Friday and Tuesday, summer and winter, trade-fair week and an ordinary week do not share the same distribution. A table broken down by day of week and season belongs naturally in your forecast report.

The Real Cost of Walking a Guest

The 7,000 TL above is only the part you can invoice. Many hotels set their overbooking buffer without even that figure in mind: they see the lost revenue of an empty room and treat walking as something to deal with "if it happens". The full picture looks like this:

• The alternative hotel: On a busy night a comparable room is hard to find and expensive; sending the guest to a lower category creates a complaint of its own.
• Transport and time: A transfer, luggage, an hour lost late at night. For the guest, the first night of a holiday or business trip is spoiled.
• Compensation: A discount, a free night or an upgrade on the next stay. Without it, the review will be harsher.
• Reviews and reputation: "I had a confirmed booking and was turned away at the door" is one of the most damaging things a guest can write about a hotel, and it stays visible for months.
• Lost guest value: A walked guest is unlikely to book direct again. If that guest was a loyal, direct-channel customer, the loss runs far beyond a single night.

As these items go into the calculation, the downside-to-upside ratio rises from two to three or four, and the safe buffer shrinks. The rule is simple: the more honestly you price a walk, the more realistic your overbooking buffer will be.

When Overbooking Makes Sense

Strategic overbooking is a legitimate revenue management tool when all of the following are true:

• You have reliable no-show and cancellation data by channel. No-shows are recorded in the PMS with the right code rather than closed as cancellations, and you have several months of history you can split by day of week.
• Flexible rates make up a large share of your mix. When many bookings carry free cancellation and pay-at-hotel terms, expected empty rooms are higher. On nights dominated by non-refundable rates, the buffer should shrink.
• You have an agreed alternative hotel nearby. A comparable property, briefed in advance and at an agreed rate, cuts both the cost and the stress of a walk.
• The buffer is monitored during the day. Overbooking is not a number you set once and forget. Afternoon confirmation calls on the day of arrival and incoming cancellations shrink or close the buffer in real time.

When Overbooking Backfires

The same tool used on the wrong night produces losses rather than profit. Avoid overbooking, or set the buffer to zero, when:

• There is a major event in town: On trade-fair, concert, congress or holiday nights almost every guest shows up and alternative rooms are close to impossible to find. Historical averages mislead on these nights.
• Groups and long stays dominate: Groups rarely no-show, and walking a seven-night booking creates a seven-night problem.
• You have no data: "A few people usually don't show" is a hunch, not a distribution. Overbooking without data is not a forecast; it is a bet.
• It puts direct and loyal guests at risk: Turning away a commission-free, returning guest damages the most valuable channel the hotel has.

Accidental Overbooking: The Decision Nobody Made

Much of the overbooking hotels actually experience is not strategic at all; nobody said "let's oversell". The same room was simply sold on two channels at once. The four most common sources are:

1. Lag between the channel manager and an OTA: The last room sells on the website at 14:02, OTA inventory updates at 14:07, and in those five minutes the same room sells on the OTA too. A channel manager reduces this risk, but sync speed and connection errors still need watching; we cover the integration side in Choosing a Channel Manager and PMS.
2. Unreleased allotments: If rooms held for an agency or tour operator are not returned to general inventory after the release date, or the release is forgotten, the same room can be sold from both sides.
3. Phone bookings never entered: A reservation taken over the phone and noted in a notebook or a message sits outside inventory, and the room sells elsewhere in the meantime.
4. Out-of-order rooms: If a room that should be off sale stays open in the PMS, the hotel sells beyond its real capacity without realising it.

Accidental overbooking has no upside, only the cost of walking guests. The fix lies in process, not in buffer calculations: a single source of inventory, phone bookings entered during the call, allotment release dates tracked in a calendar, and a daily check of channel sync errors. We look at the full channel mix in How to Optimize Your Hotel Distribution Mix.

When It Happens: A Walk Procedure

Even the best calculation fails on some nights. Leaving that night to the front desk's improvisation at midnight multiplies the cost. A written procedure should cover these steps:

1. Decide early: If there is still an excess on the afternoon of arrival, after confirmation calls and cancellations, make the decision then, not when the guest is at the desk.
2. Choose the right guest: Protect loyal and direct-booking guests, long stays and special-occasion bookings. The criteria are written down in advance, not left to the shift.
3. Tell the guest before arrival: A guest who is called before setting off and offered a solution is far more understanding than one who finds out at the door.
4. Same or higher category, free transport: The hotel covers the first night at the alternative property and the transfer. For multi-night stays, the guest is brought back the next day.
5. Record and follow up: Note the walk on the guest's profile so they are recognised and compensated on their next booking. For OTA reservations, the platform's relocation rules may also apply, so check your contract.

Overbooking Is a Data Decision

Every step in this article depends on the same thing: knowing which booking came from which channel on which terms, which segment produces no-shows, and exactly when and where inventory was decremented. If that information is scattered across the PMS, the channel manager, the phone and email, strategic overbooking becomes guesswork and accidental overbooking becomes an inevitable surprise.

Convertels is not a PMS or a channel manager; those systems stay in place. Convertels is the layer built on top of the direct channel: the direct booking system shows rate and cancellation terms clearly, the hotel call center records phone bookings during the call, and the hotel CRM system brings each guest's channel, history and pre-arrival contacts into a single record. That way you know which guests to protect and which channel produces empty rooms before midnight, not at it.

In short: for a hotel with the data, strategic overbooking turns rooms that would sit empty into revenue, with a buffer set below the average number of empty rooms and decided night by night. Accidental overbooking is a process gap that no buffer calculation can offset; it has to be closed.

---

---