The real cost of a discount: how promotional rate strategies erode RevPAR long-term

Every hotelier has faced the moment. Occupancy is soft, the revenue meeting turns tense, and someone suggests dropping the rate. It feels logical — a modest discount to stimulate demand, fill the house, protect cash flow. But what if that instinctive response is the single most expensive decision your hotel makes all year?

The evidence is unambiguous: reactive discounting does not generate the incremental demand hotels assume it does, and the true cost of a promotional rate is two to three times the face-value discount once you account for commission erosion, ADR dilution, demand displacement, and the long-term behavioural conditioning it inflicts on your guest base.

This is not opinion — it is the consistent conclusion of the most rigorous research our industry has produced over two decades. The landmark Cornell University studies by Professors Enz, Canina, and Lomanno, analysing more than 6,000 hotels using STR data, found that hotels discounting 15 to 25 per cent below their competitive set achieved higher occupancies but did not achieve higher RevPARs. In Asia-Pacific markets, the same discount tier produced a staggering -17.3 per cent RevPAR difference.

Yet here we are, an industry that spends billions annually training its guests to expect less and pay less — then wonders why ADR growth has stalled below inflation.

This article examines why the promotional architecture most hotels operate is structurally self-defeating, what the real cost chain of a discounted booking looks like, and how to build a rate strategy that treats your pricing as the commercial asset it is.

Before we dissect what goes wrong, it helps to understand the sheer breadth of what we are dealing with. The modern hotel operates within a promotional ecosystem of extraordinary complexity. A typical branded property may maintain fifteen to thirty active rate codes at any given time — advance purchase, member rates, corporate negotiated, group, packages, OTA promotional rates, opaque channels, mobile-only rates, and seasonal campaigns — each interacting with different distribution channels, commission structures, and guest segments.

The taxonomy is vast. Advance purchase discounts, the most universally deployed promotion, typically offer 10 to 15 per cent off Best Available Rate, rising to 30 or even 40 per cent for luxury properties requiring bookings several months ahead.

Length-of-stay promotions graduate from 15 per cent for three nights to 25 per cent for a week. Corporate negotiated rates carve 10 to 30 per cent off standard pricing. Group rates discount 15 to 30 per cent.

Opaque channels like Hotwire and Priceline sell distressed inventory at 30 to 60 per cent below published rates. And layered atop all of this sit the OTA participation programmes — Booking.com’s Genius offering 10 to 20 per cent off at over 390,000 properties, Expedia’s Member Pricing, and the mobile-only rates that further erode what guests actually pay.

Then there are the major brand campaigns. Marriott runs near-continuous promotional activity across its 9,500 properties. Hilton maintains a predictable quarterly cadence of bonus-point promotions and flash sales, with early-booking offers reaching 20 to 40 per cent off BAR.

IHG deploys flash sales of up to 25 per cent off. Accor, whilst scaling back the aggressive 40 per cent private sales of the pre-pandemic era, still runs summer and year-end campaigns at 20 to 33 per cent off. Each campaign trains hundreds of millions of loyalty members — Marriott Bonvoy alone has 271 million, Hilton Honors 235 million — to associate specific booking windows with discounted rates.

The question nobody in the boardroom is asking with sufficient urgency is this: when every channel, every brand, and every OTA is running promotions simultaneously, where does the “special offer” end and the actual market price begin? The answer, increasingly, is that the discounted rate is the market price — and that is precisely the problem.

The arithmetic of hotel discounting is more destructive than most operators realise, because the costs compound across multiple layers that are rarely calculated together. Let us walk through what happens to a room with a published rate of £200 when it enters the promotional pipeline.

Layer one is the face-value discount. A 20 per cent reduction takes that £200 room to £160. Straightforward enough — a £40 loss per room night. Most revenue discussions stop here. They should not.

Layer two is commission on the reduced base. If that £160 room is sold through Booking.com at an 18 per cent commission, the platform takes £28.80, leaving £131.20. Had the same room sold direct at full rate through the hotel’s own booking engine at a 2 to 3 per cent processing cost (illustrative example – usually can go up to 10%), it would have netted approximately £194. The gap between the two scenarios is not £40 — it is £62.80, a 31 per cent effective loss on what was marketed as a 20 per cent discount.

Layer three is programme stacking. Hotels participating in Booking.com’s Preferred Partner programme pay an additional 3 per cent commission for ranking visibility, pushing total commission to around 21 per cent. Genius Level 1 adds a hotel-funded 10 per cent discount. Mobile-only rates can add another 5 to 10 per cent. Country-rate targeting adds more. Booking.com’s API examples show products discounted by up to 28%, and partner materials confirm that certain promotions and closed‑user‑group rates can stack under specific configurations. The net revenue on that original £200 room can fall below £115. The face-value discount was 10 per cent. The actual revenue destruction is over 40 per cent.

Layer four is ADR dilution across the property. If 30 per cent of a hotel’s inventory sells at a 20 per cent discount, the blended ADR drops approximately 6 per cent. For a 200-room hotel at £200 ADR, that represents roughly £2,400 per day in lost revenue — not from those discounted rooms alone, but from the mathematical depression of the entire property’s rate performance.

Layer five — and this is the one revenue managers most frequently underestimate — is displacement. During compression periods, every room sold at a promotional rate potentially displaces a full-rate booking that would have materialised closer to arrival.

HotStats data reveals that since 2019, global RevPAR grew 19 per cent, but distribution costs per available room surged 25 per cent over the same period. We are paying more to earn less in real terms. Juyo Analytics estimates that the combined impact of rate leakage, BAR compression, and direct conversion loss from wholesale and OTA rate erosion can silently cost a hotel 4 to 9 per cent of annual room revenue.

The industry shorthand, supported by layering these verified cost components together, is that the true cost of a hotel discount runs at approximately two to three times the face-value reduction. A 20 per cent discount does not cost you 20 per cent. It costs you 40 to 60 per cent of the revenue you would otherwise have captured.

Here is where the conversation shifts from accounting to psychology, and where the long-term damage becomes most insidious. The prevailing assumption behind hotel discounting is that lower prices attract incremental demand — guests who would not otherwise have booked. Professor Cathy Enz of Cornell stated it plainly: “The prevailing wisdom is that reducing room rates entices new consumers to enter the market and buy more rooms. This has never worked for the hotel industry.”

The demand for hotel rooms is relatively price-inelastic. When rates drop, existing demand does not meaningfully expand. Instead, the guests who would have booked anyway simply pay less. The 2023 study by Singh and Corsun, published in the Cornell Hospitality Quarterly and analysing 2,500 hotels through the pandemic, reinforced this finding: hotels that raised rates during COVID-19 showed a significantly greater positive impact on RevPAR penetration than those that discounted.

But the real damage is not just immediate revenue loss — it is what discounting does to how guests think about your pricing for every future booking.

The anchoring effect, one of the most robust findings in behavioural economics, explains why. First formalised by Tversky and Kahneman in 1974, anchoring describes how the first price a person encounters becomes the reference point against which all subsequent prices are judged. A 2024 economics working paper testing hotel room price anchoring specifically found an anchoring index of 0.53 — meaning over half of the distance between a high and low reference price influenced the participant’s final valuation. When your guest sees “was £250, now £180” on Booking.com, that £180 does not feel like a temporary promotion. It becomes their internal reference price for your hotel. The next time they search and find £250, they do not think “that’s the standard rate.” They think “that’s £70 more than last time” — and the loss aversion documented by Kahneman and Tversky tells us that perceived loss is felt approximately twice as intensely as the equivalent gain from the original discount.

This is the ratchet effect in action: prices are easy to lower and extraordinarily difficult to raise. INSEAD research confirms that “the value of the steady state price decreases, the more consumers anchor on the lowest price.” Every seasonal sale, every flash promotion, every Genius discount is quietly ratcheting down the price your guests believe your hotel is worth.

The parallel to retail is instructive. J.C. Penney ran 590 promotional events in a single year, with over 72 per cent of revenue coming from items sold at 50 per cent or more off list price. When new CEO Ron Johnson eliminated all promotions in favour of transparent everyday pricing in 2012, sales collapsed by 20 per cent, traffic fell 10 per cent, and the company lost $985 million in one year. The lesson was not that everyday low pricing fails. The lesson was that decades of promotional conditioning had made the discount the product — customers could no longer perceive value without the theatrical framing of a deal. Transaction utility, as Richard Thaler termed it, had become more important than the actual product.

Now consider: Marriott Bonvoy members account for 75 per cent of room nights in the United States and Canada. Between Marriott and Hilton alone, over 460 million loyalty members are exposed to regular promotional cadences — seasonal sales, bonus points campaigns, flash discounts. Are we genuinely stimulating incremental travel, or are we training the most commercially important guest segment in our industry to wait for the next sale? Research from Quantum Metric found that 50 per cent of consumers now prefer regular deals from brands throughout the year over concentrated sale events. The conditioning is already deeply embedded. Phocuswright’s 2025 data describes a travel market characterised by “shorter trips, downgraded plans, and last-minute bookings” — precisely the behaviour pattern you would expect from a consumer base trained to wait, compare, and optimise for the lowest price.

If the evidence against reactive discounting is this robust, why does it remain the default response across the industry? The answer lies in a structural discipline gap — the distance between what revenue management science prescribes and what actually happens in Monday morning revenue meetings.

The first and most consequential gap is forecasting. Demand forecasting is the single discipline that separates proactive from reactive pricing. As Dr. Cindy Heo of EHL’s faculty puts it, forecasting serves as “the basis for essential revenue decisions.” Yet Skift Research found that only 28 per cent of hotels use revenue management system technology, and when free tools like Expedia’s Rev+ are excluded, the figure drops to approximately 8 to 10 per cent.

This means the vast majority of hotels globally — hundreds of thousands of properties — are making pricing decisions manually, using spreadsheets, competitive rate shopping, and instinct. Manual pricing is inherently reactive. Without unconstrained demand modelling — the process IDeaS’ chief scientist Ravi Mehrotra describes as “creating an unconstrained demand forecast, which in turn produces a more accurate holistic forecast overall” — revenue managers cannot distinguish between a genuine demand shortfall and demand that simply has not yet materialised within the booking window.

This leads directly to the panic pricing phenomenon. A practitioner case study published on Hospitality Net describes a hotel with 40 to 45 per cent of rooms unsold one week before a major event — “a scenario that would cause most teams to panic.” Competitive intelligence showed competitors were filling rapidly. By holding firm and trusting the demand signals rather than the on-the-books number, the hotel finished at 97 per cent occupancy with higher ADR than the previous year’s event period. The revenue team that discounted at the same hotel in previous years had consistently left money on the table.

The diagnostic framework for identifying reactive discounting is, in practice, quite straightforward. If your hotel is discounting within 7 to 14 days of arrival without forecast validation, you are reacting. If rate changes are triggered by competitor price movements rather than demand signals, you are reacting. If discounts are applied uniformly across all room types and segments rather than targeted through rate fences, you are reacting. If promotions run without defined end dates or success metrics, you are reacting. And if pricing decisions are driven by general managers under occupancy pressure rather than revenue managers with forecast data, the organisational dynamics are structurally producing reactive outcomes.

The alternative — strategic promotional architecture — is built on three pillars.

First, every promotion is deployed against a forecast-identified need period, not a perception of weakness. A well-designed promotional calendar is planned 90 to 365 days in advance, with defined objectives, start and end dates, and appropriate rate fences for each campaign.

Second, the depth of any discount is calibrated to the minimum effective level, with a strong preference for value-added strategies — adding breakfast, parking, spa credits, or early check-in rather than cutting the room rate itself. The seminal rate fence research by Kimes and Wirtz at Cornell established that fences framed as “advantages” (you get a lower price because you met a qualifying condition) are perceived as significantly fairer than blanket discounts, and that customers become more accepting of differential pricing as they grow familiar with it.

Third, restrictions — minimum length of stay, advance purchase requirements, non-refundable conditions, closed-to-arrival controls — serve as the structural guardrails that protect full-rate demand from cannibalisation by promotional demand.

The properties that execute this discipline consistently see the results. One Bay Area hotel case study, documented by a revenue management consultancy, shifted from a habit of dropping rates 7 to 14 days before arrival to a disciplined approach using business intelligence to validate whether existing advance-purchase and length-of-stay discounts were converting. The outcome: occupancy remained virtually flat year-over-year, but the property achieved double-digit ADR growth every single month. The rooms were nearly identical. The revenue was not.

Perhaps the most structurally challenging dimension of the discounting problem is how OTA ranking algorithms have turned promotional participation into an addiction that penalises withdrawal.

Booking.com’s default sort algorithm — “Our Top Picks” — explicitly factors in commission level, Preferred & Genius programme participation, conversion rate, and net bookings (gross bookings minus cancellations). The platform confirms this openly: an accommodation’s ranking “can also be influenced by how much commission they pay us on bookings.”

Expedia’s algorithm operates similarly, with academic research by Eisfeld finding that a $1 increase in hotel price corresponds to a drop of roughly three positions in Expedia’s search results. The study concluded that “Expedia’s ranking algorithm tends to intensify price competition between hotels” and decreases market prices by approximately €1.89 on average compared to random ranking.

The feedback loop is elegant in its destructiveness. A hotel offers a Genius discount. Conversion rate rises. The algorithm rewards the property with a higher ranking. Visibility increases. Bookings grow. The hotel becomes dependent on the discount to maintain its algorithmic position. Withdrawal of the discount causes conversion to fall. The algorithm punishes the property with a lower ranking. Bookings collapse. The hotel reinstates the discount.

Booking.com claims Genius properties receive 70 per cent more search result views and up to 45 per cent more bookings. Industry analyses of Expedia search results suggest that properties in the top 10 positions may capture well over half of all clicks, and that most bookings occur within the first 15 results.

The commercial logic for joining these programmes is seductive. The commercial logic for leaving them is terrifying. This is the structural discount dependency trap — and it is amplified by the stacking phenomenon, where Genius discounts, mobile rates, country targeting, and seasonal campaigns compound to erode BAR by 30 to 40 per cent before the hotelier has made a single deliberate pricing decision.

The regulatory landscape is shifting, but slowly and unevenly. France, Germany, Austria, Italy, and Belgium have banned or restricted rate parity clauses. The European Commission designated Booking Holdings as a “gatekeeper” under the Digital Markets Act in May 2024, and by July, Booking.com had eliminated all rate parity obligations for EEA properties.

Spain’s competition authority fined Booking.com €413.24 million in July 2024 for abuse of dominant position — the largest such fine in Spanish history. Over 15,000 European hoteliers, represented through HOTREC and 30 national hotel associations, have filed a class action seeking to recover commissions paid during the parity era.

Yet legal freedom has not automatically translated to commercial freedom. As one industry analysis noted: “The law may no longer require parity, but their business model still does.” Booking.com uses automated web crawlers to detect even minimal price differences between the hotel’s direct website and the OTA listing. The 123Compare.me World Parity Monitor found that during peak season, European hotels achieved a direct-rate advantage on over 48 per cent of checked rates — progress, but still meaning more than half of rate comparisons either showed parity or OTA advantage.

Rate integrity recovery is possible, but it demands patience, discipline, and a deliberate channel strategy. The practical framework follows a phased approach.

Stabilisation requires a full audit of every active discount, promotional commitment, and programme participation across all channels — identifying where stacking is occurring and where unauthorised rate leakage exists.

Restriction involves closing discounted rate plans during high-demand periods, enforcing length-of-stay and cancellation controls, and reducing inventory allocation to high-commission OTA programmes.

Rebuilding means shifting investment from blanket rate cuts to value-add strategies and direct booking infrastructure — optimised websites, metasearch presence, email capture, and genuine closed-user-group pricing behind real authentication walls. Triptease data shows that hotels maintaining parity on metasearch increase conversion by up to 34 per cent, and that direct bookers are 18 per cent more profitable than OTA guests.

The timeline is not trivial. Market-level analysis suggests a 6 to 12-month cycle from meaningful discount reduction to visible rate recovery. Cornell research indicates hotels using dynamic pricing strategically achieve RevPAR uplifts of up to 7.3 per cent. The billboard effect — Cornell’s finding that OTA listings generate 3 to 9 additional direct bookings per OTA reservation — remains a viable mechanism for leveraging OTA visibility without full dependency, provided the hotel’s direct channel offers genuine value differentiation.

The uncomfortable truth embedded in all of this research is that pricing is not a tactical lever to be pulled when occupancy softens. Pricing is the single most powerful signal your hotel sends to the market about its value proposition. Every discount says to the guest: this room is not worth what we normally charge. Every seasonal sale says: if you wait, we will charge you less. Every OTA programme participation says: we value algorithmic visibility more than our own rate integrity.

The hotels that will outperform over the next decade are not the ones offering the cleverest promotions. They are the ones building the organisational discipline to say no — no to the panic discount, no to the blanket flash sale, no to the stacking of OTA programmes that systematically transfer margin from the hotel’s balance sheet to the platform’s. They are investing in forecasting capability, constructing intelligent rate fences, deploying value-add strategies over rate cuts, and treating their direct channel not as a cost centre but as their most important commercial asset.

Professor Enz’s conclusion from two decades ago remains as relevant — and as widely ignored — as ever: “Hotels in direct competition make more money when they maintain their price structure and do not discount to fill rooms.” The data has not changed. The question is whether we are finally willing to act on it.

The real cost of a discount is not the percentage you advertise. It is the compounding, invisible erosion of the rate your market believes you are worth. And once that belief is lost, recovering it is the hardest commercial challenge in hospitality.

Sources:

Cornell University / Academic Research

https://news.cornell.edu/stories/2003/04/hotel-room-discounting-doesnt-increase-revenues https://scholarship.sha.cornell.edu/chrpubs/13/ https://scholarship.sha.cornell.edu/chrpubs/184/ https://journals.sagepub.com/doi/10.1177/19389655231184475 https://journals.sagepub.com/doi/10.1177/19389655211016829 https://pmc.ncbi.nlm.nih.gov/articles/PMC10323521/ https://sites.insead.edu/facultyresearch/research/doc.cfm?did=41566

Distribution & OTA Dynamics

https://partner.booking.com/en-us/solutions/genius-0 https://partner.expediagroup.com/en-us/resources/blog/travel-marketplace-visibility-guide https://luiseeisfeld.github.io/assets/docs/HotelRankings_Eisfeld_TSE.pdf https://www.littlehotelier.com/blog/get-more-bookings/preferred-partner-program/ https://www.triptease.com/resources/metasearch-the-rate-parity-battleground https://www.hotel-online.com/news/how-ota-giants-use-your-margins-to-build-loyalty-and-what-smart-hoteliers-are-doing-instead

Rate Parity & Regulatory

https://bookbetterdirect.com/background-booking-coms-record-fine-in-spain/ https://www.hotelmanagement.net/tech/ai-and-end-rate-parity https://www.uknewsgroup.co.uk/world-parity-monitor-europe-gains-pricing-control-as-hotels-tackle-rate-parity-2/ https://scidaproject.com/2024/05/15/blog-6-a-new-summer-designation-booking-com-the-dma/

Revenue Management & Pricing

https://juyoanalytics.com/the-true-cost-of-hotel-distribution/ https://hospitalityinsights.ehl.edu/hotel-demand-management https://www.hospitalitynet.org/opinion/4130250.html https://www.researchgate.net/publication/45714666_Price_fencing_in_the_practice_of_revenue_management_An_overview_and_taxonomy

Behavioural Economics

https://www.hbs.edu/news/articles/Pages/rajiv-lal-on-jcpenney.aspx https://www.pon.harvard.edu/daily/negotiation-skills-daily/the-anchoring-bias-consumers-beware/ https://www.ucd.ie/economics/t4media/WP2024_07.pdf