Why Cutting Your Rate to Fill Rooms Loses Money
Cornell's 67,008-observation dataset: hotels priced 20–30% below comp set filled 15% more rooms and earned 12% less RevPAR. The evidence, and exceptions.
Cutting rate to fill rooms usually works exactly as intended. You fill more rooms — and still end up with less revenue per available room than the hotel that held its price. Cornell measured this across 67,008 hotel observations and found the pattern held in a boom and a recession. The reason is that lodging demand barely responds to price, so a discount moves guests between hotels rather than creating new ones. There are real exceptions, and they are at the end of this post.
The instinct is reasonable, and it is still wrong
Friday shows fourteen rooms unsold and forty-eight hours to go. Every one of those rooms earns nothing if it goes empty. Dropping $30 to move eight of them feels obviously correct.
The logic is sound for that one night in isolation. It stops being sound the moment it becomes how you price. Your comp set watches, your guests learn, and the discount stops being a tactic and becomes your position.
Cornell's data is about that position, not about a single Friday.
The numbers, in full
Enz, Canina and Lomanno analyzed 67,008 hotel observations from 2001 to 2007, a period spanning both a downturn and a boom (Competitive Hotel Pricing in Uncertain Times, Cornell Hospitality Report Vol. 9 No. 10, 2009). They compared each hotel to its own comp set — the small group of nearby hotels it competes with directly.
| Rate vs comp set | Occupancy vs comp set | RevPAR vs comp set |
|---|---|---|
| 20–30% lower | +15.2% | −12.2% |
| 10–15% lower | +4.5% | −8.3% |
| 5–10% higher | −0.5% | +6.7% |
| 20–30% higher | −10.9% | +10.3% |
Read the top row carefully. The deep discounters got what they wanted — fifteen percent more occupancy than their competitors. They earned twelve percent less per available room while doing it.
Now read the bottom row. Hotels priced 20–30% above their comp set sold 10.9% fewer rooms and made 10.3% more revenue per available room.
Independents get the worse end of the trade
The report also breaks out hotel type, and the independent series is harsher. In the 20–30% discount band, independent hotels showed a RevPAR gap of −16.31% against their comp sets. In the 5–10% premium band, independents showed +4.12%.
That makes sense structurally. A branded hotel discounting still has a loyalty program, a central reservation system and brand recall pulling business in. An independent discounting has only the price.
If you run an independent, the discount costs you more and buys you less.
The mechanism: demand hardly notices your price
The reason sits in a separate Cornell study. Canina and Carvell examined 480 hotels over eleven years, 1989 to 2000, and estimated the price elasticity of lodging demand at about −0.14.
Elasticity is simply how much demand moves when price moves. At −0.14, a 1% rate cut produces roughly a 0.14% increase in demand. Cut 10% and demand responds about 1.4%.
What actually drove lodging demand in that data was income and economic activity — not room rates. People travel because they have a reason and the money. They do not decide to take a trip because your Tuesday got cheaper.
Taking share is not creating demand
This is the sentence worth keeping. A discount does not add travelers to your town. It reassigns the ones already coming.
That is why the occupancy gain is real and the revenue gain is not. You captured share from three competitors, at a price that made the captured business worth less than the business you already had.
It also explains why the tactic is self-defeating at market level: your competitors can see your rate, and matching costs them nothing. Cornell noted that roughly 54% of hotels were priced below their comp set in both 2001 and 2004 — which is arithmetically a market talking itself downward.
The caveat the authors state themselves
Cornell is explicit that this analysis shows correlation, not causation. The data cannot prove the discount caused the lower RevPAR.
There is a plausible reverse story: weaker hotels discount because they are weak, and their lower RevPAR reflects the weakness rather than the pricing. That story is not ruled out.
What can be said is this. Across 67,008 observations, in every rate band, in good years and bad, lower relative price never coincided with higher relative RevPAR. If discounting worked, that pattern would have broken somewhere.
When discounting is actually the right call
Treating the above as an absolute rule is its own mistake. Rooms are perishable — an unsold night is gone permanently, and its marginal cost of sale is small. There are cases where dropping price is straightforwardly correct.
Genuinely distressed last-minute inventory. Inside 24 to 48 hours, most of your booking window has closed. Cloudbeds' 2026 State of Independent Hotels — 90 million bookings across 180 countries, 2025 data — put the independent booking window at 40 days. At day one, the remaining demand really is thin, and a rate above your cost per occupied room beats an empty room.
Structural oversupply, not a soft week. If three new hotels opened in your market this year, your old rate position may no longer exist. That is repositioning, and it should be deliberate.
Buying length of stay. A discount that converts a one-night booking into three is not really a discount on rate. It is a trade for volume you would not otherwise have, at a lower servicing cost per night.
A property with no history. New build, new room type, no reviews. Buying early trial and review volume is an investment with a defined end date.
Base business with ancillary spend. A contracted crew rate below your usual floor can still work if it brings reliable food and beverage revenue. That is a TRevPAR argument, not a RevPAR one.
The common thread: each is bounded. A defined window, a defined volume, a defined end. Chronic discounting has none of those.
The mistakes that make this worse
Dropping rate before pace tells you to. Pace is how fast a date is filling versus a comparable date. A quiet-looking Friday three weeks out is usually normal, not a problem.
Discounting the whole date rather than a segment. Cut your public best available rate and you cut it for every guest who was already going to book at full price.
Matching the cheapest hotel in the comp set. That hotel may be distressed, mispriced, or a tier below you. Copying it imports its position.
Leaving the discount up after the reason has gone. Rate is easy to lower and slow to recover, because the market has already recalibrated around your new number.
Never checking what the discount bought. Compare occupancy and RevPAR to the same date last year. If occupancy rose and RevPAR did not, you paid for volume.
The bottom line
Discounting reliably buys occupancy and reliably costs revenue per available room, and the effect is worse for independents than for brands. Demand for hotel rooms is close to price-inelastic, so the rooms you gain come from your neighbors rather than from new travelers. Keep discounting as a bounded tool for perishable, genuinely distressed inventory — and stop using it as an answer to the general feeling that a date looks empty.
The hard part is usually knowing where you actually sit before you touch the rate. Our market position module shows the live 25th, 50th, 75th and 90th percentile rate bands for your market, plus the headroom to your target in dollars. The choice to hold or cut then gets made against the market, not against a hunch.