Free tool

Hotel Overbooking Calculator: How Many Rooms to Overbook

The hotel overbooking calculator estimates how many extra bookings a hotel can accept for a night, from its own no-show and late-cancellation history, the revenue from an extra room and the cost of walking a guest. See the chance of a walk and the expected gain at each level.

Last reviewed · Published by Global Lux Agency

The night
Rooms you can sell for the night, after out-of-order rooms. Up to 2,000.
Your history for nights like this
% of bookings on the books that don’t arrive. Example 5%: use your own figure.
Cancelled too late to resell. Example 3%: use your own figure.
Revenue and cost
Usually ADR for the night, net of VAT. Less any no-show fee you would keep.
Relocation room, transport, waived charges, goodwill and compensation. Example only: enter your own.

Recommended overbooking

7 extra bookings

Accept up to 107 bookings for 100 rooms. Show rate 92%; critical ratio 33.3% (revenue ÷ revenue + walk cost).

Chance of walking anyone
23.9%
Expected walks
0.47
Expected rooms occupied
97.97
Expected gain vs none
£2,263
By overbooking level expected shows, chance of walking at least one guest, expected walks and expected gain vs none
ExtraShowsWalk riskWalksGain
192.920%0£414
293.840.2%0£825
394.760.9%0.01£1,227
495.682.9%0.04£1,602
596.67.1%0.11£1,924
697.5214%0.24£2,156
7 recommended98.4423.9%0.47£2,263
899.3635.9%0.81£2,218
9100.2848.9%1.27£2,011
10101.261.5%1.85£1,648
11102.1272.6%2.52£1,148
12103.0481.5%3.28£540

An estimate from a simple model: it assumes guests arrive independently and that demand fills the extra bookings. Groups and events arrive or cancel together, so be more cautious on those nights, and never walk a VIP or a returning guest when you can avoid it.

How the calculation works

  1. Show rate. 100% less your no-show and late-cancellation rates. Each booking is treated as turning up independently with that probability, so the number of arrivals follows a binomial distribution.
  2. Each overbooking level. For rooms + 0, 1, 2 and so on bookings, the tool works out the full probability of every possible number of arrivals. From that it gets the expected rooms occupied, the chance that more guests arrive than you have rooms, and the expected number of walks.
  3. Expected gain. Revenue per room × expected rooms occupied, less walk cost × expected walks, compared with not overbooking at all.
  4. Recommendation. The level with the highest expected gain. This is the newsvendor critical-ratio rule: one more booking is worth taking while the chance that all your rooms are already needed is below revenue ÷ (revenue + walk cost). When two levels tie, the tool picks the smaller.

Walking guests at a luxury hotel

The arithmetic treats every walk as a cost you can count. At luxury level the real cost is often larger: a walked VIP, honeymooner or loyal returning guest can be lost for good, and the story travels. Put that into the walk cost, plan who would be walked first (usually a single-night stay with a booking that can be honoured at a comparable sister property), and keep the overbooking level lower on nights with VIP arrivals, events or groups.

What the tool doesn’t do

  • It doesn’t supply no-show or cancellation rates: no reliable published average exists for luxury hotels. Use your own history for comparable nights.
  • It doesn’t model groups or correlated cancellations, multi-night stays, or room types. Run it separately for a room type if you overbook by type.
  • It doesn’t forecast demand. It assumes the extra bookings will come if you accept them.

Questions & answers

How many rooms should a hotel overbook?
Keep accepting extra bookings while the chance that every room is needed stays below revenue ÷ (revenue + walk cost). With a £450 room and a £900 walk cost that ratio is a third. For a 100-room hotel where 8% of bookings fail to arrive or cancel too late, that works out at about 7 extra bookings, with roughly a 24% chance of walking at least one guest and an expected gain of about £2,260 for the night. Raise the walk cost to £2,700 and it drops to 5, with about a 7% chance of a walk.
What should the cost of walking a guest include?
Everything it costs you: the room at the hotel you send the guest to (often at short notice and a higher rate), transport there and back, any first night or upgrade you waive, a gesture such as dinner or a credit, and compensation you have promised. For luxury guests also weigh the lifetime value at risk if a loyal guest or a VIP is walked. Most luxury hotels set this figure high, which pushes the recommendation down.
Where do the no-show and late-cancellation rates come from?
From your own history for comparable nights: the same day of the week, season, segment and booking channel. Measure both as a share of the bookings still on the books on the morning of arrival. The defaults in the tool are examples only; no reliable published average exists for luxury hotels, and rates vary widely between groups, leisure and corporate business.
Does a no-show fee change the answer?
The tool treats a no-show as revenue lost from that room. If you charge no-shows the first night, a no-show still leaves the room empty, but you keep part of the revenue, so the case for overbooking is a little weaker. Enter the revenue you would gain from re-selling the room, not the full rate, to reflect that.
What does the model assume?
That every booking turns up or not independently, with the same probability, and that demand is strong enough to fill the extra bookings you accept. In practice groups and wedding blocks arrive or cancel together, which makes walks more likely than the model shows, so be more cautious on those nights.

Sources & data

Every figure this tool uses by default is listed here with where it came from and when we last checked it. Spotted something out of date? Tell us.

  • Overbooking as a newsvendor problem (critical ratio rule)Revenue-management teaching treats the overbooking level as a single decision taken before the number of no-shows is known: the newsvendor model, where the optimum sets the probability of having too few no-shows equal to the critical ratio of the underage cost (an empty room that could have been sold) to the sum of the underage and overage (walked guest) costs. Lecture examples often model no-shows as Poisson; this tool models each booking as showing independently (binomial), which gives the same marginal rule. Practitioner sources word the formula differently (some add a fare ratio), so treat it as a planning model, not a forecast.Last verified 13 November 2026 · Consistent industry sources · Sources: Seoul National University OpenCourseWare

Add this tool to your website

Free for agencies, brokers and suppliers. Paste this snippet into your page — it resizes automatically and always uses our latest data. Please keep the credit link. In Notion, Ghost and other editors that support oEmbed, pasting the page link is enough.

Cite or link to this tool

Writing about this topic? You are welcome to link to or cite this tool.

For brands & suppliers

Reach the people planning luxury travel

Our tools are used by advisors, hotels, DMCs and travellers at the moment they are planning and pricing. Sponsorships, featured listings and custom tools are available.

Work with us