Two Routes To The Same RevPAR

Fourteenth article in this silo, and the last one this session. Canadian hospitality is unusually well served by published performance data and unusually badly served by explanations of what to do with it.

The Verdict, Stated First

Five claims, in descending order of confidence.

One. On our own arithmetic rate and occupancy are not interchangeable, because occupancy carries a variable cost per room sold and rate does not.

Two. On our own arithmetic a discount requires far more occupancy than operators expect, with a 20 percent cut needing 22.5 additional points to stand still.

Three. Both results turn entirely on cost per occupied room, a figure most hotels do not compute.

Four. We found a widely republished Canadian benchmark presented as an annual forecast that is a single month's data, which we demonstrate below.

Five. And on the published national data, rate rather than occupancy has been doing the work, which the arithmetic above suggests is the better of the two.

The fourth is the one a hotel could act on wrongly today, ours. Benchmarking annual performance against a June figure will tell a good hotel it is failing.

Our Grades For These Claims

Applying the scheme this publication uses throughout.

Grade A for the performance data, which comes from CoStar press releases and industry reports and is internally consistent where it should be[1][5][6].

Grade A for the period mismatch we document, since we obtained both sources and the figures are identical to the cent.

Grade A for our own arithmetic, which is reproducible from the RevPAR identity.

Grade D for our cost per occupied room assumption, which is invented, and which every conclusion in the second half depends on.

Grade C for the forecast material, which is a commercial real estate firm's outlook reported by a retail trade site.

A Note On Method

Everything here is checked to 29 August 2026.

We obtained CoStar press releases for full-year 2025 and for June and July 2026[1][5][6], and an industry report summary and a forecast report[2][3].

We did not obtain any underlying dataset, all of which are sold by subscription, so every figure here is from a press release or a summary of a paid report.

We found no Canadian benchmark for cost per occupied room, which is the single most important input to the arithmetic in this article and which we have therefore invented and flagged.

All arithmetic is ours. The hotel, its room count, its rate, its occupancy and its variable cost are invented to demonstrate a structure.

This article discusses hotel operations and is not accounting or business advice.

Where Canadian Hotels Actually Sit

The published position, from the strongest sources we obtained.

CoStar reported that Canada's hotel industry recorded its highest annual top-line performance on record for full-year 2025, with British Columbia reporting the highest absolute levels in each key metric at 70.4 percent occupancy, CAD 257.03 ADR and CAD 180.92 RevPAR; that Quebec at negative 1.3 percent and Ontario at negative 0.6 percent were the only provinces to see an occupancy decline while all posted ADR and RevPAR increases; and that among major markets Vancouver registered the highest absolute levels at 78.4 percent, CAD 284.44 and CAD 223.05[1].

An industry report summary puts the national figures for 2025 at occupancy stabilised at 66 percent, ADR risen to $216 and RevPAR at a historic high of $143, and states that growth was primarily driven by strong pricing strategies[2].

Four observations, ours.

Sixty-six percent national occupancy is the number to hold onto, and it is well below what a hotel operator's intuition tends to suggest.

The provincial spread is wide. British Columbia at 70.4 percent against a 66 percent national figure means several provinces sit materially lower.

Vancouver at 78.4 percent against British Columbia's 70.4 shows the same effect within a province, so a city benchmark and a provincial one are different instruments.

And the attribution is the useful part, ours. Growth driven by pricing rather than by occupancy is precisely the distinction the rest of this article is about.

One caution on comparing yourself to any of these, ours. National and provincial figures blend limited-service and full-service properties, which have different rates, different occupancies and very different cost structures, so a single-segment hotel is not the average of them.

A Benchmark Quoted From The Wrong Period

A finding we did not expect and can demonstrate precisely. Ours.

A hospitality trade site summarising a quarterly lodging outlook states: "Canadian hotel markets show divergent 2026 forecasts with national occupancy slightly below 2025 at 73.0% versus 75.6% while ADR rises to CAD252.63 from CAD239.72 and RevPAR edges up to CAD184.33."[4]

A CoStar press release dated 23 July 2026 reports June 2026 data for Canada: occupancy 73.0 percent, down 3.5 percent; average daily rate CAD 252.63, up 5.4 percent; RevPAR CAD 184.33, up 1.6 percent[5].

Four observations.

The three figures are identical to the cent. 73.0 percent, CAD 252.63 and CAD 184.33 appear in both.

One source presents them as a 2026 forecast against 2025. The other identifies them as a single month, June 2026, against June 2025.

The percentage changes settle it, ours. CoStar's occupancy figure is down 3.5 percent year over year, which is consistent with a month and not with the summary's framing of a forecast slightly below 2025.

And the annual figures make it impossible. National occupancy for full-year 2025 was 66 percent[2], so a 75.6 percent "2025" comparator in the summary cannot be a Canadian annual number.

There is a further clue in the same source, ours. The quarterly product it summarises covers six major markets rather than all of Canada[4][7], so even correctly quoted it would not be a national figure.

How We Caught It

The method, because it generalises. Ours.

Four observations.

We were not looking for it. We were collecting benchmark figures and noticed two sources giving the same three numbers for different periods, which is not a coincidence available to chance.

The confirmation came from the year-over-year percentages, which a forecast summary tends to drop and a monthly release always carries.

The cross-check was the annual figure from a third source, and 66 percent annual against a claimed 75.6 percent for the same year is not a rounding difference.

And the generalisable test is one line, ours. When a benchmark is quoted without a period, find the same numbers somewhere that states one.

The habit it comes from is worth naming, ours. We record the period beside every figure we collect, which costs nothing and is the only reason two identical triples in different files were visible at all.

Why That Would Mislead A Hotel

The practical consequence, ours.

Four observations.

June is a strong month and the year is not. Benchmarking an annual result against a June figure compares twelve months including February to one month in early summer.

The gap is large. Seventy-three percent against sixty-six is seven points, and RevPAR of CAD 184.33 against $143 is a difference of about 29 percent.

So a hotel performing at the national annual average would appear to be failing by a wide margin against the misquoted figure.

And the direction of the error is the dangerous one, ours. It tells a competent operator to cut rate, which the rest of this article argues is usually the wrong move.

It would also misprice a transaction, ours. A hotel valued on a RevPAR multiple against an inflated comparator looks cheap when it is not, and the same error runs through any lender's covenant set built on the wrong benchmark.

The Two Routes

Our own arithmetic on an invented hotel, and the core of this article.

A 150-room hotel at an average daily rate of $180 and occupancy of 68 percent. That is 54,750 available room-nights, 37,230 sold, RevPAR of $122.40 and rooms revenue of $6,701,400. Assume a variable cost per occupied room of $35, covering housekeeping, laundry, amenities, utilities and in-room consumables. That figure is invented and we flag it as the weakest input here.

Now target a 5 percent RevPAR gain, to $128.52. Two ways to get there.

Route A, rate: hold occupancy and raise ADR from $180 to $189.00. Route B, occupancy: hold rate and raise occupancy from 68 percent to 71.4 percent, being 3.4 additional points.

Four observations.

Both produce the same RevPAR, and both would be reported identically in any benchmarking product.

Route A sells the same number of room-nights at a higher price. Route B sells 1,862 additional room-nights at the same price.

Revenue is nearly identical: $335,070 either way on these figures.

And the costs are not, ours, which is the entire point.

Sixty-Five Thousand Dollars

Our own arithmetic, continued.

Route A adds $335,070 of revenue and no incremental cost, because no additional room is cleaned, no additional guest consumes anything, and no additional utility is used. Net gain: $335,070.

Route B adds $335,070 of revenue and $65,153 of incremental cost, being 1,862 additional room-nights at $35. Net gain: $269,918.

The difference is $65,152 a year on an identical headline metric.

Four observations.

Nothing in a RevPAR report distinguishes those two outcomes. A management report showing RevPAR up 5 percent has concealed a 19 percent difference in what reached the bottom line.

The gap is exactly the variable cost of the extra room-nights, which means it is knowable in advance for any hotel that measures that cost.

Route A also consumes no additional capacity, leaving the rooms available for a higher-value booking, which is a real option the arithmetic does not price.

And the asymmetry runs the other way in a downturn, ours. Losing occupancy costs less than losing rate, by the same $35 a room-night, which is the one comfort in a soft market.

Which suggests where to defend, ours. In a soft market, protect rate and let occupancy take the hit, because the profit lost per unit of RevPAR is smaller that way and the rate is harder to recover than the volume.

Flow-Through Is The Word For It

The concept, and why it belongs on a management report. Ours.

Four observations.

Flow-through is the share of an incremental revenue dollar that reaches profit. On our figures rate flows through at 100 percent and occupancy at 80.6 percent.

Those two numbers should sit beside RevPAR on any report that drives a decision, because RevPAR alone cannot distinguish a rate gain from an occupancy gain.

The rate figure of 100 percent is an idealisation and we say so: percentage-based costs such as credit card fees and commissions do scale with rate, which we address next.

And the occupancy figure is the one to measure rather than assume, ours, since it is one minus the ratio of variable cost to rate, and both of those are specific to the property.

A Third Thing We Got Wrong

A correction to our own working, published rather than quietly dropped.

We expected to show that distribution commission widens the gap between the two routes, on the reasoning that commission scales with revenue while housekeeping does not. We ran it and the gap does not widen. It is constant.

Our own arithmetic. At 0 percent commission the routes net $335,070 and $269,918, a gap of $65,152. At 10 percent: $301,563 and $236,411, gap $65,152. At 18 percent: $274,757 and $209,605, gap $65,152. At 25 percent: $251,303 and $186,150, gap $65,152.

Four observations.

The gap is identical at every commission rate, because commission reduces both routes' revenue proportionally and the revenue is the same on both.

Algebraically the gap reduces to the variable cost of the extra room-nights, which contains no commission term at all.

So our hypothesis was wrong in the form we stated it, ours, and we would have published it had we not computed it.

And there is a true version of the claim, which is the next section.

What Commission Actually Does

The corrected statement. Ours.

Four observations.

The absolute gap is constant and the relative gap widens. At zero commission $65,152 is 19.4 percent of the rate route's net gain; at 25 percent commission the same $65,152 is 25.9 percent of a smaller net gain.

So commission does make the choice between routes matter more, by shrinking both prizes while the penalty stays the same size.

That is a weaker and more precise claim than the one we set out with, and it is the one that survives the arithmetic.

And it has a practical edge, ours. A hotel heavily dependent on high-commission channels has the most to gain from choosing rate over occupancy, because its net gains are small enough that a fixed $65,152 penalty is a large proportion of them.

We would put the general lesson beside it, ours. An error in a hypothesis is cheap and an error in a published claim is not, and the only difference between the two is whether somebody ran the arithmetic before writing it down.

The Discount That Does Not Pay

Our own arithmetic, and the calculation that should precede any rate cut.

The invented hotel's current rooms profit after variable cost is $5,398,350. Cut the rate and ask how many room-nights are needed to hold that.

A 5 percent discount takes ADR to $171.00, contribution per night to $136.00, and requires 39,694 room-nights, or 72.5 percent occupancy. A 10 percent discount: $162.00, $127.00, 42,507 nights, 77.6 percent. A 15 percent discount: 83.6 percent. A 20 percent discount: $144.00, $109.00, 49,526 nights, 90.5 percent. A 25 percent discount: 98.6 percent.

Four observations.

A 20 percent discount requires 90.5 percent occupancy to stand still, from a base of 68.

That is 22.5 additional points, which is not a promotional outcome; it is a different hotel in a different market.

At 25 percent off, the required occupancy is 98.6 percent, which is effectively unreachable and means the discount cannot be recovered by volume at any achievable level.

And the curve is convex, ours. Each additional point of discount requires more incremental occupancy than the last, because contribution per night is falling while the target is fixed.

Which sets a hard ceiling on the strategy, ours. There is a discount beyond which no achievable occupancy recovers the position, and on our figures that point arrives somewhere between 20 and 25 percent off.

Twenty-Two And A Half Points

Why that number is worth carrying. Ours.

Four observations.

Rate decisions are made quickly and occupancy responses arrive slowly, so the discount is live long before anyone can tell whether the volume appeared.

The break-even is also rarely stated as an occupancy target, which is the form a general manager can actually judge. Told to find 22.5 points, most would refuse.

Told to match a competitor's rate, the same manager will agree, because the two framings do not feel like the same decision.

And that is the argument for computing it before the meeting, ours. The break-even occupancy is the discount expressed in a unit the operator has intuition about.

The Number That Decides It

Our own arithmetic, testing how much the conclusions depend on our invented cost.

At a variable cost per occupied room of $20, occupancy flow-through is 88.9 percent and a 20 percent discount breaks even at 87.7 percent occupancy. At $28: 84.4 and 89.1. At $35, our base case: 80.6 and 90.5. At $45: 75.0 and 92.7. At $60: 66.7 and 97.1.

Four observations.

The same decision produces opposite answers across that range. At $20 a deep discount is nearly recoverable; at $60 it is not recoverable at all.

So every conclusion in the second half of this article rests on a number we invented, and a hotel substituting its own would get a different answer.

The range is not exotic. A limited-service property and a full-service one plausibly sit at opposite ends of it, which is why a single industry rule of thumb cannot serve both.

And that makes the measurement the deliverable, ours, rather than any figure in this article.

We would go further on that point, ours. A hotel that computes its own cost per occupied room and ignores every number we have published will be better off than one that adopts ours, because the structure is general and the level is not.

And Almost Nobody Computes It

Our own observation, offered as observation. Ours.

Four observations.

We found no Canadian benchmark for cost per occupied room, having searched for one, while occupancy, rate and RevPAR are published monthly by province and by market.

The asymmetry is telling. The three revenue metrics are measured, benchmarked and reported nationally; the cost figure that determines what they are worth is not.

It is also computable from records a hotel already holds. Departmental rooms expenses divided by room-nights sold, over any period, is the first approximation and takes an afternoon.

And the reason to do it is not benchmarking, ours. It is that the number is an input to every pricing decision the hotel makes, and it is currently absent from all of them.

The refinement worth adding later is seasonal, ours. Cost per occupied room is not constant through the year, since utilities and casual labour move with season and occupancy, so a single annual figure understates winter and overstates summer.

Rate Is Doing The Work Nationally

What the published data shows about which route the industry has taken.

A commercial real estate firm's outlook reports that month-over-month and year-to-date data through July shows the pace is stronger than anticipated, predominantly on average daily rate growth, less so on occupancy, which has been relatively flat, and forecasts national occupancy no higher than 66 percent for 2025, 2026 and 2027 with ADR projected to rise to $216 in 2026 and $221 in 2027[3].

The monthly releases show the same shape. June 2026: occupancy down 3.5 percent, ADR up 5.4 percent, RevPAR up 1.6 percent[5]. July 2026: occupancy up 1.6 percent to 78.9 percent, ADR up 7.2 percent to CAD 267.44, RevPAR up 9.0 percent to CAD 211.01[6]. A quarterly outlook records RevPAR growth of 4.0 percent in 2025 with June year-to-date up a further 6.5 percent[7].

Four observations, ours.

In every period we obtained, rate grew faster than occupancy. In June occupancy fell while RevPAR still rose.

On our own arithmetic that is the higher-flow-through route, so the industry has been growing RevPAR the profitable way rather than the volume way.

Whether that was chosen or simply available is a different question, and the outlook attributes it partly to domestic travel offsetting a decline in United States visitors[3].

And a flat national occupancy forecast through 2027 has a strategic reading, ours. If occupancy is not going to grow, rate is the only lever left, which is a demanding position for a property already at the top of its market's rate.

Which Is Fortunate, Given Costs

The pressure on the other side of the account.

An industry report notes that looking to 2026 and 2027 the sector is expected to move from recovery into moderate growth, with modest GDP expansion of 1.2 to 1.6 percent, and that rising operating costs, particularly labour, utilities and insurance, will continue to pressure profitability even as room rates remain elevated[2].

Four observations, ours.

Labour and utilities are substantially the components of cost per occupied room, so rising costs there push our flow-through figures down and our discount break-evens up.

Which means the case for the rate route strengthens as costs rise, since occupancy becomes progressively more expensive to buy.

Insurance is the exception in that list, being largely fixed rather than per occupied room, so it pressures profit without changing the choice between routes.

And the combination is the operating environment, ours. Flat occupancy, rising variable costs and rate as the only growing line is a description of a market where pricing discipline is the whole game.

Which is an uncomfortable position for one kind of property, ours. A hotel already at the top of its competitive set has the least rate headroom and therefore the least access to the one lever the market is offering.

The Same Pattern A Year Earlier

Whether 2025 was unusual, checked against the year before.

CoStar reported that Canada's hotel ADR and RevPAR were the highest for any year on record on 2024 data, and quoted its director of hospitality analytics that while ADR and RevPAR hit all-time highs, the country's occupancy level was its highest since 2018; that RevPAR growth was strongest in the fourth quarter, bolstered by concert demand in Toronto and Vancouver; that the annual growth rate slowed against 2023, reflecting weaker economic conditions and strong prior-year comparables; and that room rates once again outpaced inflation, while supply and demand were flat year over year[8].

Four observations, ours.

Supply and demand flat, rates outpacing inflation. That is a rate-led year described in the provider's own words, a full year before the one we examined.

The phrase "once again" is doing quiet work, since it implies the pattern predates 2024 as well.

Occupancy at its highest since 2018 is also worth pausing on, because it means the volume recovery had largely completed by then and there was little occupancy left to win.

And that reframes the national picture, ours. Rate has been the only growing lever for at least three years, which makes it a structural condition rather than a strategy anyone chose.

Events Move The Comparator

A caution about month-level benchmarking that the data makes vivid. Ours.

The monthly releases repeatedly attribute large swings to single events: Nova Scotia's plus 15.7 percent ADR and plus 20.2 percent RevPAR helped by a sailing event in Halifax[5]; Montreal's 23.7 percent RevPAR surge in July[6]; and fourth-quarter 2024 growth bolstered by concert demand in Toronto and Vancouver[8].

Four observations, ours.

A twenty-point provincial RevPAR swing from one event tells you the monthly series is not a stable comparator for an individual property.

It also means a hotel in the affected market had an exceptional month it did not earn, and will face an impossible comparable the following year.

Which compounds the period problem documented earlier, since an operator benchmarking against an event-inflated month is measuring against something no ordinary trading produced.

And the discipline is the same one, ours. Compare like periods, know what was in them, and treat any month with a named event in the commentary as unusable for trend.

The annual series is the safer instrument for that reason, ours, since a single event is a smaller share of twelve months than of one, and the national figure of 66 percent has already absorbed every festival and concert in the calendar.

When Occupancy Is The Right Answer

The counterweight, because this article has been one-sided. Ours.

Four observations.

An empty room earns nothing and costs the fixed base anyway, so at low occupancy almost any rate above variable cost is worth taking.

Occupancy also drives revenue the room rate does not capture: food and beverage, parking, spa and meeting space, none of which appear in RevPAR and all of which scale with heads in beds.

A hotel with substantial non-rooms revenue therefore has a materially higher effective contribution per occupied room than our $35 figure implies, which shifts the whole analysis toward occupancy.

And there is a structural argument too, ours. Sustained low occupancy invites a competitor's expansion, and market share defended cheaply now can be expensive to recover later, which is a consideration no single-year arithmetic captures.

Ranking is a related mechanism, ours. Booking platforms weight conversion and volume, so a property that stops filling rooms can become harder to find, which is a cost of the rate strategy that does not appear in any of our figures.

If You Run A Hotel

Practical, and not accounting or pricing advice. Ours.

Four points.

Compute your cost per occupied room. Departmental rooms expense over room-nights sold, and every conclusion here changes with it.

Put flow-through beside RevPAR on the management report, because RevPAR cannot distinguish a rate gain from an occupancy gain and the two are worth different money.

State every proposed discount as a break-even occupancy before approving it. On our own figures a 20 percent cut needed 22.5 additional points.

And check the period on any benchmark before comparing yourself to it, since we found a single month's national data republished as an annual forecast.

If You Advise One

For our own profession. Ours.

Four points.

Split rooms expense into fixed and per-occupied-room components, which most hotel charts of accounts do not do and which is the precondition for everything above.

Reconstruct last year's RevPAR movement into its rate and occupancy parts, and price each at its own flow-through. The profit story is frequently different from the RevPAR story.

Ask which benchmark the client uses and for what period, and verify it against a source that states one.

And bring non-rooms revenue into the contribution figure. A hotel with meaningful food and beverage has a different answer from our worked example and should not use ours.

What To Do

Stop treating RevPAR as a single number. It is a product of two factors with different economics, and the report conceals which one moved.

Measure cost per occupied room, because it determines flow-through and every discount break-even in the business.

Prefer rate to occupancy at the margin, on our own arithmetic, unless non-rooms revenue or a low occupancy base says otherwise.

Convert every discount into a required occupancy before approving it. A 20 percent cut needed 90.5 percent occupancy on our figures.

Note that commission does not change the absolute gap between the two routes, which was our own error, but does widen it in proportion.

Verify the period of any benchmark, since we found June 2026 national data presented as a 2026 annual forecast.

Use the right geography. Vancouver at 78.4 percent, British Columbia at 70.4 and Canada at 66 are three different comparators.

And remember occupancy earns the ancillary revenue, which sits outside RevPAR entirely and belongs in the decision.

The Limits Of This Analysis

Several caveats matter. This article discusses hotel operations and is not accounting, tax or business advice. We obtained no underlying dataset: every performance figure here comes from a press release or a summary of a paid subscription product, and we could not inspect any methodology. Our cost per occupied room of $35 is invented, we found no Canadian benchmark for it despite searching, and it is the single input on which every conclusion in the second half of this article depends; the sensitivity table shows the results reversing across a plausible range. All arithmetic is ours: the 150 rooms, the $180 rate, the 68 percent occupancy and every derived figure are invented to demonstrate a structure. Our rate flow-through of 100 percent is an idealisation, since percentage-based costs including commissions and card fees do scale with rate; we address commission separately and do not model card fees at all. Our discount break-even calculation holds all else constant, and in practice a rate cut changes channel mix, guest profile and ancillary spend in ways we have not modelled. We did not model non-rooms revenue, which materially favours the occupancy route for full-service properties and which we flag rather than quantify. The period mismatch we document is our own finding from comparing two sources; we did not obtain the underlying quarterly report and it is possible the summary we quote misrepresents it rather than the report being wrong. And one of the three findings in this article is a correction to our own hypothesis, published because we ran the arithmetic expecting commission to widen the gap and found it constant.

Related Reading

Niche Industry Finance Larger Than The Profit Another thin-margin business where the headline metric hides where the money went. Niche Industry Finance The March Trench The other article this session where our own arithmetic disproved what we set out to show. Niche Industry Operational Finance The Breach Nobody Caused Section 130.1 caps any holder at 25 percent of a class. Because that is a ratio, one investor redeeming can put another over the line without them...

Frequently Asked Questions

Are rate and occupancy equally valuable?

No. On our own arithmetic for an invented 150-room hotel, a 5 percent RevPAR gain is worth $335,070 through rate and $269,918 through occupancy, a difference of $65,152 on an identical headline metric. The gap is exactly the variable cost of the additional room-nights.

What occupancy does a discount need to break even?

More than most operators expect. On our own figures, from a 68 percent base, a 5 percent discount needs 72.5 percent, a 10 percent discount needs 77.6 percent, a 20 percent discount needs 90.5 percent and a 25 percent discount needs 98.6 percent, which is effectively unreachable.

Does distribution commission change the choice?

Not in the way we expected. We thought commission would widen the absolute gap between the routes and it does not: the gap is $65,152 at every commission rate, because commission scales both routes' revenue equally. What does happen is that the gap widens in proportion, from 19.4 percent of the rate route's net gain at zero commission to 25.9 percent at 25 percent.

What is Canadian hotel occupancy?

National occupancy for full-year 2025 is reported at 66 percent with ADR of $216 and RevPAR of $143, and forecast to go no higher than 66 percent through 2027. British Columbia reported the highest provincial figures at 70.4 percent, and Vancouver the highest major market at 78.4 percent, so the right comparator depends on geography.

Why check the period on a benchmark?

Because we found one being quoted from the wrong one. A trade summary presents national occupancy of 73.0 percent, ADR of CAD 252.63 and RevPAR of CAD 184.33 as 2026 forecasts. A CoStar release gives the identical three figures as June 2026 monthly data. Benchmarking an annual result against a June figure would tell a hotel at the national average that it is failing badly.

When is filling rooms the better answer?

When occupancy is low enough that rooms are earning nothing, and when the hotel has substantial non-rooms revenue. Food and beverage, parking and meeting space scale with occupancy, sit outside RevPAR entirely, and raise the effective contribution per occupied room well above the $35 in our worked example.

What single number should a hotel compute first?

Cost per occupied room, being departmental rooms expense divided by room-nights sold. On our sensitivity analysis, moving it from $20 to $60 changes occupancy flow-through from 88.9 percent to 66.7 percent and the break-even on a 20 percent discount from 87.7 percent occupancy to 97.1 percent. The same decision reverses across that range.

References

1. CoStar press release, dated 21 January 2026, reporting full-year 2025 Canadian hotel performance. States that Canada's hotel industry reported its highest annual top-line performance on record; that among the provinces and territories British Columbia reported the highest absolute levels in each of the key performance metrics, being occupancy of 70.4 percent, ADR of CAD 257.03 and RevPAR of CAD 180.92; that Quebec at negative 1.3 percent and Ontario at negative 0.6 percent were the only provinces to see a decline in occupancy while all provinces and territories posted increases in ADR and RevPAR; and that among the major markets Vancouver registered the highest absolute performance levels at occupancy of 78.4 percent, ADR of CAD 284.44 and RevPAR of CAD 223.05. Note: a press release from the data provider. The underlying dataset is a paid subscription product we did NOT obtain, and no methodology was available to us. costar.com
2. Hotel industry news site's summary of a hotels and chains report for 2026, April 2026, drawing on CoStar data and a consultancy's proprietary research. States that the Canadian hotel industry achieved a record year in 2025, with national occupancy stabilised at 66 percent, ADR risen to $216 and RevPAR at a historic high of $143; that growth was primarily driven by strong pricing strategies and resilient travel demand although regional disparities persisted, with Western Canadian markets outperforming; that looking ahead to 2026 and 2027 the industry is expected to move from recovery into a more stable phase of moderate growth with GDP expansion of 1.2 to 1.6 percent; that demand is expected to normalise with domestic travel the primary driver; and that rising operating costs, particularly labour, utilities and insurance, will continue to pressure profitability even as room rates remain elevated. Note: a trade news site summarising a paid consultancy report we did NOT obtain. Our source for the national annual figures, which are load-bearing in the period-mismatch finding. hotelnewsresource.com
3. Retail trade publication's report on a commercial real estate firm's Canadian hotels outlook, November 2025. Reports that Canadian hotels are on pace to exceed expectations largely because Canadians are choosing to travel domestically, offsetting a drop in United States tourists; quotes the firm's senior vice president that month-over-month and year-to-date data through July shows the pace stronger than anticipated, predominantly on average daily rate growth and less so on occupancy which has been relatively flat; and states that the firm forecasts RevPAR to stay positive in most Canadian markets in 2026 at between 2 and 4 percent over 2025, with national occupancy forecast to go no higher than 66 percent for 2025, 2026 and 2027 while ADR is projected to rise to $216 in 2026 and $221 in 2027. Note: a retail trade publication reporting a commercial real estate firm's forecast, NOT the forecast document itself, flagged. retail-insider.com
4. Hospitality trade site's summary of a quarterly Canadian lodging outlook for the second quarter of 2026. States that Canadian hotel markets show divergent 2026 forecasts with national occupancy slightly below 2025 at 73.0 percent versus 75.6 percent, while ADR rises to CAD 252.63 from CAD 239.72 and RevPAR edges up to CAD 184.33; that Halifax leads gains with occupancy of 89.3 percent and ADR of CAD 278.95 while Calgary sees occupancy decline to 78.4 percent and ADR drop to CAD 209.39; and that each report includes occupancy, ADR and RevPAR for six major markets, with data for all of Canada available by annual subscription. Note: a trade site's summary, NOT the quarterly report itself. THIS IS THE SOURCE THIS ARTICLE IDENTIFIES AS PRESENTING JUNE 2026 MONTHLY DATA AS A 2026 FORECAST; its three national figures are identical to the cent to those in ref 5. We did not obtain the underlying report and it is possible the summary misrepresents it rather than the report being wrong. hotel-online.com
5. Hospitality trade site's report of a CoStar press release dated 23 July 2026, giving June 2026 Canadian data with percentage change from 2025: occupancy 73.0 percent, down 3.5 percent; average daily rate CAD 252.63, up 5.4 percent; revenue per available room CAD 184.33, up 1.6 percent. Records that this was Canada's first monthly occupancy decline since December 2025; that Nova Scotia reported the largest gains in ADR at plus 15.7 percent to CAD 270.46 and RevPAR at plus 20.2 percent to CAD 228.22, helped by a sailing event in Halifax; and that Newfoundland and Labrador saw the highest occupancy lift at plus 4.6 percent to 86.4 percent. Also records November 2025 occupancy of 61.6 percent, down 1.0 percent. Note: a trade site reporting a CoStar release. The MONTHLY identification of the three figures that ref 4 presents as an annual forecast, and the basis of this article's period-mismatch finding. hotel-online.com
6. Hospitality trade site's report of a CoStar press release dated 24 August 2026, giving July 2026 Canadian data with year-over-year change: national occupancy 78.9 percent, up 1.6 percent; ADR CAD 267.44, up 7.2 percent; RevPAR CAD 211.01, up 9.0 percent. Records that Quebec led provincial performance with occupancy up 7.8 percent to 81.7 percent and RevPAR up 17.1 percent to CAD 213.11, while Montreal topped major markets with an 11.8 percent occupancy increase and a 23.7 percent RevPAR surge. Note: a trade site reporting a CoStar release. Cited to show the rate-led pattern continuing and to illustrate the size of monthly variation against the 66 percent annual figure. hotel-online.com
7. Hotel industry news site's summary of a consultancy's Canadian lodging outlook for the second quarter of 2026, stating that despite ongoing global geopolitical and economic uncertainty the Canadian hotel industry continues to deliver results, with RevPAR growth of 4.0 percent in 2025 and June year-to-date performance up a further 6.5 percent; and that each quarterly report includes occupancy, ADR and RevPAR for six major markets, with data for all of Canada available by annual subscription. Note: a trade news site summarising a report we did NOT obtain, flagged. Note its explicit statement that the quarterly product covers SIX MAJOR MARKETS rather than all of Canada, which is relevant context for the period-mismatch finding. hotelnewsresource.com
8. CoStar press release dated 23 January 2025, reporting that Canada's hotel average daily rate and revenue per available room were the highest for any year on record on 2024 data; quoting the provider's director of hospitality analytics for Canada that while ADR and RevPAR hit all-time highs the country's occupancy level was its highest since 2018, that RevPAR growth was strongest during the fourth quarter bolstered by concert demand in Toronto and Vancouver, and that the annual growth rate slowed compared with 2023 reflecting weaker economic conditions and strong prior-year comparables; that room rates once again outpaced inflation while supply and demand were flat year over year; and that the lowest market occupancy was reported in Edmonton at 58.5 percent. Note: a press release from the data provider, cited for the prior-year baseline and for the observation that rate rather than volume drove growth in 2024 as well. costar.com

This article discusses hotel operations and is not accounting, tax or business advice. No underlying dataset was obtained; every performance figure comes from a press release or a summary of a paid subscription product. The cost per occupied room used throughout is invented and no Canadian benchmark for it could be found. All arithmetic is the authors' own and every hotel figure is invented.