Ever refreshed a hotel booking page twice in one hour and watched the price change? You weren’t imagining it. Hotel pricing today runs on algorithms that recalculate value dozens or even thousands of times a day, weighing everything from local events to how many rooms are already sold down the hall. What looks like a random number on a screen is actually the output of a fairly sophisticated pricing engine, one that blends decades-old revenue management theory with modern AI.
Understanding how that number gets built can help travelers time their bookings better and give hoteliers a clearer picture of why “the same room” can cost wildly different amounts depending on the day. Below are seven of the biggest forces shaping the rate you see when you search for an empty room tonight, next weekend, or six months from now.
1. Supply and Demand in the Local Market
At its core, hotel pricing is still a supply-and-demand exercise, just executed at high speed. When a market has more rooms than travelers want on a given night, rates soften to fill beds. When demand outstrips supply, whether because of a conference, a holiday weekend, or a sold-out stadium event, rates climb quickly because every unsold room represents lost revenue that can never be recovered once the night passes.
This dynamic played out vividly during the 2026 World Cup final weekend, when New York City hotels saw ADR surge more than forty percent year over year, with a single Saturday night showing more than a hundred percent lift in both ADR and RevPAR compared to the same night the year before. Among the Top 25 Markets, New York City saw the largest increases in ADR at 41.5% to $425.03 and RevPAR at 40.3% to $377.07, with a 105.1% lift in ADR on the Saturday before the World Cup final.[1] That kind of spike shows just how directly a temporary demand shock can override a hotel’s normal pricing pattern.
2. Seasonality and Local Events
Beyond the big headline events, hotels build pricing calendars around predictable seasonal rhythms, ski season in the mountains, spring break in Florida, summer in coastal resort towns. Revenue managers layer in known recurring events too: trade shows, festivals, graduation weekends, and sports schedules that repeat year after year and create reliable demand bumps.
Modern pricing tools now go further than a simple calendar of known dates. Dynamic pricing is the practice of adjusting hotel rates in real-time based on demand, seasonality, local events, and competitor behavior.[2] Systems increasingly pull in signals beyond the obvious, since predictions can be based on real-time data on occupancy, competitor pricing, local and global events, and even the weather to adapt prices to expected peak and low periods.
3. Booking Window and Reservation Pace
How far in advance you book, and how quickly rooms are filling relative to history, matters enormously. Revenue managers track “pace,” meaning the rate at which reservations for a future date are being picked up compared to the same point in previous booking cycles, to decide whether to raise or hold rates.
If guests are starting to book 60 days out instead of 30, rates should reflect that rising urgency.[2] Conversely, when a normally busy date is pacing behind schedule, hotels often trim rates to stimulate bookings before the date arrives, since rooms stop earning the moment rates fall behind demand and one slow day on pricing is enough to lose the reservation.[3] This is also why last-minute bookings can go either way: sometimes hotels drop rates to fill remaining inventory, and other times they raise them because only premium rooms are left.
4. Competitor Rates and Market Positioning
No hotel prices in a vacuum. Revenue teams constantly monitor what comparable properties nearby are charging, a practice often called rate shopping, and adjust to stay competitively positioned without leaving money on the table. If a comparable hotel across the street raises its rate for a weekend, that often signals rising demand worth capturing too.
The sophistication of this competitive tracking has grown considerably. Competitive intelligence has become more contextual, with the best AI hotel pricing software not just ingesting competitor rates but interpreting them.[4] Understanding which competitor moves actually reflect real demand shifts, and which are noise, helps hotels avoid chasing every price change a neighbor makes.
5. Distribution Channel and Where You Book
The channel through which a room is booked, whether that’s a hotel’s own website, a call center, an online travel agency, or a corporate booking tool, can influence the price a guest ultimately sees. Hotels pay commissions on OTA bookings, so many push rate parity or slight direct-booking discounts to encourage travelers to book straight through the property, improving the hotel’s net revenue even when the sticker price looks similar.
The value gap between channels can be substantial. SiteMinder reported that direct hotel and short-term rental websites reached an average booking value of $516 in 2025, while OTAs averaged $312, meaning channel mix now affects both occupancy and margin.[3] That’s a meaningful reminder that the channel you choose doesn’t just affect what you pay, it affects what the hotel actually keeps.
6. Room Type, Guest Segment, and Length of Stay
Not every guest pays the same price for the same room type, and that’s intentional. Hotels segment demand by traveler type, corporate, leisure, group, or long-stay, and by rate plan, refundable versus non-refundable, package versus room-only, because each segment has a different willingness to pay and a different cancellation risk profile.
This segmentation has real financial consequences that go beyond simple occupancy numbers. Managers typically split the forecast by direct, OTA, corporate, group, long-stay, and repeat guests, because a full house with high commission costs can still hurt net earnings.[3] That’s why a hotel might occasionally turn away a low-value group booking in favor of leaving rooms open for higher-margin transient guests expected to book later at a better rate.
7. AI-Driven Dynamic Pricing Systems
Perhaps the biggest shift in the past few years is who, or what, is actually setting the price. Revenue management used to rely heavily on human analysts working from spreadsheets and historical booking curves. That model has been steadily replaced by self-learning software that reacts to markets in near real time.
The scale of this shift is striking. Legacy platforms relied on fixed rules and human fine-tuning, remaining static unless someone manually rewrote the logic, but AI has changed that, with self-learning pricing engines now updating themselves thousands of times per day.[5] The financial impact is being documented too, since real-time dynamic pricing powered by AI can increase ADR by 10 to 15%[5] in some cases, while separate industry data suggests machine-learning pricing engines can deliver up to 15 to 20% higher RevPAR within six months for hotels new to a revenue management system[6].
8. Market Tier and Location-Driven Pricing Power
Where a hotel sits on the quality spectrum, and where it’s physically located, shapes not just its base rate but how much pricing flexibility it has. Luxury and upper-upscale properties have consistently outpaced budget hotels in rate growth recently, largely because affluent travelers have kept spending even as price-sensitive segments pulled back.
The data on this divide is fairly stark. Luxury and upper-upscale hotels have been operating at roughly 67.3% occupancy with a $281 ADR and $189 RevPAR, while midscale and economy tiers sit near 54.5% occupancy with an $86 ADR and just $47 RevPAR, with the high end forecast to lead RevPAR growth at roughly 5.3% even as the economy segment posts flat-to-negative demand.[7] Location amplifies this further. Individual markets can diverge sharply from national averages, as seen when Las Vegas Strip hotel RevPAR fell 10.9% to $149.13 in 2025 even as Strip ADR held at $199.79, still 24.5% above the U.S. national average[8], showing that a single market’s fortunes can move independently of the broader industry.
What This Means for Travelers and Hoteliers
For travelers, the practical takeaway is that hotel rates are rarely arbitrary, even when they feel that way. Booking earlier for high-demand dates, being flexible on room type, and comparing direct rates against OTA listings can all make a measurable difference given how segmented and channel-sensitive pricing has become.
For hotel operators, the trend line is clear: hotels that adopt modern AI revenue management will be able to respond to volatility and execute pricing decisions consistently even with lean teams, while those that don’t will still be working hard, just in the wrong way.[9] With national performance metrics for 2026 pointing to modest, uneven growth, precise pricing has stopped being a nice-to-have and become one of the clearest levers separating hotels that thrive from those that merely survive.
The empty room sitting on a booking site tonight isn’t priced on a whim. It’s the product of demand curves, competitor benchmarking, booking pace, channel economics, and increasingly, an algorithm quietly recalculating value in the background. Once you see the machinery behind the number, that price on the screen starts to make a lot more sense.
