In this article9 sections
A 62-room boutique property posts 81 percent occupancy for the month and the general manager calls it a win. Ownership, looking at the same period, sees gross operating profit down 6 percent year over year and asks what happened. Both are reading real numbers. Neither is reading the wrong one. The problem is that occupancy and profit answer different questions, and a hotel that tracks only the first one is flying with half the instruments dark.
2026 is the year that gap gets expensive. Labor cost per occupied room has climbed for two straight years, European flow-through sits near a third, and demand growth has slowed enough that a hotel can no longer out-occupy a cost problem. This is a working guide to the hotel performance KPIs that matter, in the order a revenue-literate owner or general manager should actually look at them, with the arithmetic behind each one and the specific numbers 2026 has produced so far.
Hotel KPIs: The Five Core Numbers
Hotel KPIs are standardized measures of how a property converts available rooms into revenue and, further down the chain, into profit. Five of them are not independent measurements. They are layers of one calculation, each adding what the previous one left out, and reading them in sequence is what turns a scoreboard into a diagnosis.
Start with occupancy: rooms sold divided by rooms available. It is the most intuitive number on the dashboard and the most misread, because it says nothing about price. Any property can push occupancy toward 95 percent by discounting hard enough, which is exactly why occupancy never travels alone. Pair it with ADR, average daily rate, which is room revenue divided by rooms sold. ADR measures what you charged on the rooms you actually filled and ignores the ones that sat empty.
Multiply the two together and you get RevPAR, revenue per available room. RevPAR is the number that settles the argument occupancy and ADR cannot settle alone, because it counts every room in inventory whether it sold or not. A hotel that fills 90 rooms at $120 and one that fills 70 at $150 can land on a similar RevPAR from two completely different strategies, and RevPAR is the only one of the three that shows you that.
Add every other revenue stream, food and beverage, spa, parking, meeting space, and RevPAR becomes TRevPAR, total revenue per available room. For a limited-service property this tracks RevPAR closely and adds little. For a full-service hotel or resort, TRevPAR can run 20 to 40 percent above RevPAR, and reporting RevPAR alone on that kind of property hides a real chunk of the earning base.
| Metric | Formula | What It Tells You |
|---|---|---|
| Occupancy | Rooms sold / rooms available | Demand capture, blind to price |
| ADR | Room revenue / rooms sold | Pricing on rooms actually sold |
| RevPAR | ADR × occupancy | Combined pricing and demand result |
| TRevPAR | Total revenue / rooms available | Full earning capacity, not rooms alone |
| GOPPAR | Gross operating profit / rooms available | What actually reaches the owner |
Bottom line: occupancy and ADR are inputs, RevPAR is the combined output, and TRevPAR widens the lens to every revenue stream a property runs. None of the four says whether the hotel made money. That is the next section.
GOPPAR vs RevPAR: Which Matters More?
GOPPAR, gross operating profit per available room, is RevPAR after operating costs are subtracted. It answers the question RevPAR cannot: of everything the hotel took in, how much actually reached the bottom line. In 2026 this has become the more important of the two, because U.S. hotel GOPPAR remains roughly 10 percent below 2019 levels even though revenue metrics passed 2019 benchmarks some time ago, and only about 11 percent of European independents track GOPPAR at all, according to Cloudbeds' 2026 State of Independent Hotels report.
Here is why the gap opened. Labour typically makes up 47 to 60 percent of a hotel's operating expenses, and U.S. hotel wages sit roughly 15.3 percent above 2019 levels against operating revenue up only about 12.8 percent over the same period. Revenue recovered. The cost base reset permanently and never gave that ground back. A hotel can post a respectable RevPAR and still have a worse year than the number suggests.
Work the math on one month for an 80-room independent hotel. Occupancy 72 percent, ADR $178, 30-day month.
- Available room-nights: 80 × 30 = 2,400
- Occupied rooms: 2,400 × 0.72 = 1,728
- Room revenue: 1,728 × $178 = $307,584
- RevPAR: $178 × 0.72 = $128.16
- Ancillary revenue (F&B, parking, meeting space): $45,000
- Total revenue: $307,584 + $45,000 = $352,584
- TRevPAR: $352,584 ÷ 2,400 = $146.91
- Total operating costs for the month: $255,000
- Gross operating profit: $352,584 − $255,000 = $97,584
- GOPPAR: $97,584 ÷ 2,400 = $40.66
Now check flow-through, the share of new revenue that reached profit. The prior month ran $330,000 in total revenue and $88,000 in GOP. This month added $22,584 in revenue and $9,584 in GOP.
- Change in revenue: $352,584 − $330,000 = $22,584
- Change in GOP: $97,584 − $88,000 = $9,584
- Flow-through: $9,584 ÷ $22,584 × 100 = 42.4 percent
That 42.4 percent beats the HotStats European benchmark of roughly 35 percent, so this hotel is converting new revenue to profit better than the regional average, which is worth knowing before anyone celebrates the RevPAR line alone. See our full framework for total revenue management for how to build this cadence property-wide.
Bottom line: track revenue and profit side by side every month. A rising RevPAR with a flat or falling GOPPAR is not a good month, it is a discounting problem wearing a good month's clothes.
Reading RGI, MPI and ARI Correctly
RGI, MPI and ARI measure performance against a competitive set rather than in isolation, and they answer a question raw RevPAR cannot: was the result good, or did the whole market move? MPI (Market Penetration Index) is your occupancy divided by comp set occupancy, times 100. ARI (Average Rate Index) is your ADR divided by comp set ADR, times 100. RGI (Revenue Generation Index) is your RevPAR divided by comp set RevPAR, times 100, and it is mathematically the same answer as MPI times ARI divided by 100.
An index of 100 means fair share. Above 100 means you are outperforming the set on that dimension; below means you are trailing it. The diagnostic value comes from reading all three together rather than any one alone.
Take a hotel running 78 percent occupancy at an ADR of $178, against a comp set running 74 percent occupancy at $185.
- MPI: (78 ÷ 74) × 100 = 105.4
- ARI: (178 ÷ 185) × 100 = 96.2
- Hotel RevPAR: $178 × 0.78 = $138.84
- Comp set RevPAR: $185 × 0.74 = $136.90
- RGI: ($138.84 ÷ $136.90) × 100 = 101.4
This property is capturing more than its fair share of occupancy but at a rate roughly 4 points below its comp set, and the two nearly cancel out into an RGI just above fair share. That is a specific finding: the issue lives in rate, not demand capture, so the fix is a pricing conversation, not a marketing one. A high MPI paired with a low ARI is the most common and most expensive pattern in comp set benchmarking, because it means occupancy was bought with a discount instead of earned with positioning. Our RGI, MPI and ARI guide walks through the full diagnostic matrix.
Bottom line: never read RGI alone. It can hide a rate problem and an occupancy problem that happen to offset each other, and only MPI and ARI separately show you which lever actually moved.
What Does CPOR Actually Cost You?
Cost per occupied room (CPOR) is total room-related operating cost divided by rooms sold, and it sets the floor beneath which selling a room destroys value instead of creating it. Most operators can quote their ADR from memory and have no idea what their CPOR is, which is exactly what makes distressed-date pricing a guess instead of a calculation.
Labor is the dominant line inside CPOR, typically 60 to 70 percent of room-related expense, and 2026 data shows the pressure has not eased. Across roughly 5,000 U.S. hotels, all-hotel labor CPOR rose from $45.96 in Q1 2025 to $46.79 in Q1 2026, a 1.8 percent increase, while hours per occupied room actually fell 2.3 percent over the same period. That combination, cost up, hours down, means productivity is absorbing some of the wage pressure, but it is not eliminating it. For full-year 2025 the picture was sharper: average labor CPOR rose 12.8 percent, from $42.82 to $48.32, with the fourth quarter alone posting a 21.1 percent year-over-year jump as demand softened.
Benchmarks vary by segment: budget properties typically run $25 to $45 CPOR, midscale $40 to $65, upscale $65 to $100 and luxury above $100. None of these numbers is a target for your property; they are context for your own trend line. If CPOR is flat and ADR is rising, the hotel is gaining ground. If CPOR climbed 6 percent while ADR moved 2 percent, there is a problem regardless of what any published benchmark says.
Bottom line: a rising CPOR is not automatically bad and a falling one is not automatically good. Read it against ADR and against your own history, never against a benchmark alone.
Pickup and Pace: Your Early Warning
Booking pace compares on-the-books demand for a future date against the same point in a prior period, and pickup measures how many rooms were added since the last check, typically since yesterday or the same day last week. Together they are the only forward-looking numbers in this entire list. Everything else describes what already happened; pace and pickup describe what is still changeable.
The independent hotels with the most accurate forecasts share four habits: they pull on-the-books and rate data twice weekly rather than daily or monthly, they segment the forecast into four to six booking categories, they reconcile forecast against actual every week for the period just closed, and they hold at least 18 months of pickup history for the property. A date can be behind on pace but picking up fast, which calls for patience, or behind on pace and flat, which calls for a rate or restriction change now, while the date is still 30 to 60 days out. Confusing the two is the single most common forecasting mistake at independent properties. Our demand forecasting accuracy playbook covers the full pickup-curve methodology.
Bottom line: pickup and pace are the two numbers worth a daily look. Everything downstream of them, ADR, RevPAR, GOPPAR, is a monthly or quarterly read on decisions that were already made.
Why Guest Reviews Move RevPAR
Guest satisfaction is a revenue KPI, not only a service one, and the research behind that claim is specific enough to act on. Cornell Hospitality Research, led by Associate Professor Chris Anderson, found that a one-point increase in a hotel's 100-point online reputation score is associated with up to a 0.89 percent increase in ADR, a 0.54 percent increase in occupancy, and a 1.42 percent increase in RevPAR. The effect held across both online and offline booking channels, and later research found hotels with a higher share of direct bookings also carry higher review scores, a plausible result of the extra service flexibility direct guests receive.
The practical KPI set here is short: guest satisfaction score from post-stay surveys, online review score across major platforms, and repeat guest ratio. None of these needs a dedicated tool beyond what most PMS and reputation platforms already export, and all three move the same lever, pricing power, from a different angle than occupancy or rate ever will.
Bottom line: a hotel chasing RevPAR through rate cuts alone is fighting the wrong problem if guest satisfaction is the actual ceiling on what it can charge.
Channel Mix: Direct vs OTA Math
Channel mix is the share of room nights sold through each distribution source, direct, OTA, GDS and wholesale, and it is one of the few KPIs where the underlying cost of each channel is public enough to calculate exactly. Booking.com commission typically runs around 15 percent before add-ons, and independent analysis of hotel contracts and invoices puts the true all-in cost, once optional visibility and promotion programs are folded in, at 18 to 30 percent at Booking.com and 17 to 23 percent at Expedia. A direct booking, once payment processing, booking engine cost and a fair share of marketing spend are counted, lands closer to 4 to 7 percent all-in.
On a $200 room night, shifting a single booking from an 18 percent OTA channel to a 5 percent direct channel is worth roughly $26 of pure margin, and the cancellation math compounds the case: OTA bookings ran a 21.8 percent cancellation rate in 2025 against 10.6 percent for direct, per Cloudbeds' 2026 State of Independent Hotels report built on 90 million bookings, so the OTA revenue a hotel does keep is also twice as likely to evaporate before arrival. None of this argues for abandoning OTAs, which still drive roughly 63 percent of independent hotel bookings and reach a single property cannot replace. It argues for treating commission as a cost of acquisition to be managed per date, wide open on dates that need the demand, tightened on dates that will sell out anyway. See our channel mix break-even framework for the date-by-date version of this decision.
Bottom line: channel mix is not a loyalty question, it is an arithmetic one. Track net revenue per channel, not gross bookings, or the cheapest channel in the report is the most expensive one in reality.
Building a KPI Cadence That Works
A metric reviewed at the wrong frequency behaves like noise or like a missed opportunity, and both look identical to a busy general manager. The fix is matching cadence to the metric rather than to the meeting schedule already on the calendar.
- Daily: pickup, booking pace, on-the-books occupancy
- Weekly: RevPAR, ADR, occupancy trend, RGI/MPI/ARI, cancellation rate
- Monthly: GOPPAR, TRevPAR, CPOR, channel-level net revenue
- Quarterly: flow-through, average length of stay trend, booking window shift, guest satisfaction trend
Two rules make this stick. First, give every metric an owner: an unowned number on a dashboard is decoration nobody is accountable for. Second, pair every volume metric with a value metric so a target can never be hit by spending more than it saves. Occupancy travels with ADR. Direct booking share travels with net contribution, not gross share, since a jump in direct bookings bought through expensive paid search can cost nearly as much as the commission it replaced.
Bottom line: five KPIs reviewed on the right cadence with a named owner beat twenty metrics on a dashboard nobody has time to read.
Frequently Asked Questions
What is the most important hotel performance KPI?
There is no single answer that fits every property, but GOPPAR is the closest thing to it for an owner, because it is the number that survives comparison across properties of different sizes and it is what actually reaches the bank account. A revenue manager works from pickup and pace day to day; an owner works from GOPPAR and flow-through.
What is a good GOPPAR for an independent hotel?
There is no universal figure; GOPPAR depends heavily on segment, location and star tier. The more useful question is your own trend: U.S. hotel GOPPAR overall remains roughly 10 percent below 2019 levels, so a property tracking flat or improving GOPPAR against its own history is already outperforming the broader recovery.
How often should I check pickup and pace?
Daily, for any date inside the next 30 to 60 days. Pace and pickup are the only two numbers in a hotel's KPI set where checking daily changes the outcome, since they are the only ones you can still act on before the date arrives.
Is RevPAR or occupancy more important?
Neither alone. Occupancy ignores price and can be inflated by discounting; RevPAR combines both but still says nothing about cost. Read occupancy and ADR together, then read the resulting RevPAR against GOPPAR before drawing any conclusion about the month.
What's a healthy CPOR for my hotel?
It depends on segment: roughly $25 to $45 for budget, $40 to $65 midscale, $65 to $100 upscale, and $100 plus for luxury. Treat these as context, not a target, and judge your own CPOR against your own ADR trend rather than the published range.
Do I need STR or CoStar data to benchmark my hotel?
It helps but it is not the only route. A defined comp set of 5 to 10 genuinely comparable properties, reviewed at least twice a year, produces a workable RGI, MPI and ARI even without a paid benchmarking subscription, provided the comp set is chosen for genuine competition rather than comfort.
Can two hotels with the same RevPAR have very different profit?
Yes, routinely. One property can hit a RevPAR of $130 through disciplined pricing and controlled costs, and another can hit the identical $130 through heavy discounting that drove occupancy but wrecked flow-through. RevPAR cannot tell the two apart; GOPPAR can.
When should a hotel outsource revenue management?
Below roughly 40 rooms with a simple rate structure, a well-run owner can often manage the core KPIs directly. Above that, or with multiple segments, channels and a comp set that shifts seasonally, the daily pickup and pace discipline this guide describes becomes a full role, and most independent owners are better served handing it to a dedicated revenue manager than doing it as a fourth or fifth job.
Conclusion
A hotel does not need twenty KPIs reviewed with equal urgency. It needs five or six read at the right frequency, with GOPPAR and flow-through answering the question RevPAR cannot, pickup and pace catching what is still changeable, and guest satisfaction treated as the pricing lever the Cornell research shows it to be. Get that structure right and the monthly review stops being a scoreboard recap and starts being the meeting where the next quarter's pricing actually gets decided.
If your hotel's KPI reporting still stops at occupancy and RevPAR, get in touch and we will walk through what a GOPPAR-level dashboard would show for your property.
Written by
Revenuenaire ExpertThe Revenuenaire revenue management team: hotel and short-term rental pricing specialists writing practical, data-backed guidance on dynamic pricing, OTA optimization and revenue strategy.
