How to Analyse a Hotel Stock (RevPAR, ARR, Occupancy)

In the September 2024 quarter, Chalet Hotels — which owns its rooms outright — earned an operating margin of about 40% on ₹377 crore of revenue, and still posted a net loss of roughly ₹139 crore (Inve data, Q2 FY25). Three quarters later, in June 2025, the same company did ₹895 crore of revenue at a similar 40% margin and made ₹203 crore of net profit (Inve data, Q1 FY26). Same hotels, same beds, same staff. The difference was room rate and occupancy moving a little — and a high fixed-cost base turning a modest revenue swing into the difference between a loss and a fat profit. (Illustration of how to read the numbers, not a view on the stock.)

That is the whole game in one example. A hotel is among the most operating-leveraged businesses on the exchange: the cost of building and staffing the property is largely fixed, so once you cover it, the next room sold is almost pure margin — and the room not sold tonight is gone forever, unsold inventory that cannot be carried to tomorrow. Get the demand right and a hotel mints cash; get it wrong and the same fixed cost crushes you. The headline you read in the results — revenue up, profit up — tells you the cycle is in your favour. It does not tell you whether you are buying that favour at the top of the cycle and calling it a growth story.

This is how to read a hotel company the way an analyst who has sat through a downturn does: the handful of operating numbers that actually drive the model, where to find the ones the income statement hides, the right way to value a business this cyclical, and the one mistake that turns a great hotel into a bad investment.

A boundary first, said plainly: from the outside you cannot forecast next year's occupancy — that depends on the economy, on supply additions across the city, on whether a G20 or a wedding season lands in your quarter. What you can do is read the operating metrics for what they reveal about pricing power, and check whether you are paying a peak-cycle multiple for it.

What actually drives a hotel's economics?

Strip a hotel company down and it sells one perishable thing: a room-night. Tonight's empty room earns nothing and cannot be stored. So the entire model turns on two levers — how many rooms you fill (occupancy) and how much you charge for each (the room rate) — sitting on top of a cost base that barely moves whether the hotel is half full or full.

That fixed-cost base is the source of both the magic and the menace. Below a break-even occupancy, the hotel bleeds; above it, incremental revenue drops to the bottom line at 60–70 paise on the rupee. It is why the same property can swing from a loss to a 40%-plus margin in a year without doing anything differently — the demand simply crossed the line. Two consequences follow, and they decide how you read everything else. First, revenue growth alone tells you almost nothing — a hotel chain can grow revenue 20% purely because the whole industry's rates rose in an up-cycle, with zero added skill. Second, the question is never "is it growing?" but "is the growth rate-led or room-led, and where are we in the cycle?" A chain raising rates because demand is tight is riding a wave; a chain adding rooms and holding rates is building a business. They look identical on the revenue line.

The metrics that matter — and where they hide

Here is the spine. For a hotel, the income statement is the least informative document; the operating metrics live in the investor presentation and the concall, and that is exactly why most retail investors never see them.

RevPAR — the one number that captures the whole model

RevPAR (revenue per available room) = ARR × occupancy. It blends both levers into a single figure: the average revenue earned per room you have, whether or not it was sold. It is the truest one-line measure of a hotel's commercial health, because it punishes both empty rooms and underpricing at once.

You will almost never find RevPAR on the income statement — it sits in the investor PPT and gets discussed on the call. When IHCL's management said RevPAR would "grow in the high single digits (7–9%) on a sustained basis" (Indian Hotels Q4 FY26 concall, via Inve data), that one sentence told you more about the durability of the business than the entire P&L. Lemon Tree's RevPAR ran ₹5,494 in Q3 FY26, up about 9% year-on-year (Inve data, Q3 FY26); Chalet, an upper-upscale owner, ran far higher at roughly ₹8,000–8,100 in the same H1 FY26 stretch (Inve data, Q1–Q2 FY26). The levels differ by segment — that is normal. What you watch is the trend and the split: RevPAR rising because ARR is rising is pricing power; RevPAR rising only because occupancy is climbing toward a ceiling is a one-time gain that runs out.

ARR — pricing power, the half that compounds

ARR (average room rate, sometimes labelled ADR — average daily rate) is the price half of RevPAR. It matters more than occupancy over time because occupancy is capped — a hotel cannot be more than 100% full — while rate has no ceiling. A chain that can push rate through a cycle has real brand pricing power; one that fills rooms only by discounting is renting demand, not owning it.

ARR is buried in the PPT and the concall, never the income statement. Chalet's ADR rose about 16% to ₹12,170 in Q2 FY26 — the company's own words were "16% growth in average room rates (ADR) to INR12,170" (Chalet Q2 FY26 concall, via Inve data). EIH (Oberoi/Trident) runs at the luxury end, with ARR around ₹17,168 in Q2 FY26 (Inve data, Q2 FY26) — roughly 40% above Chalet's, which is the whole point of a luxury brand. Good looks like ARR rising faster than inflation while occupancy holds — that is genuine pricing power, not a give-away on rate to keep beds warm.

Occupancy — the demand gauge, with a ceiling

Occupancy is the share of available rooms actually sold. It is the demand thermometer, and unlike ARR it has a hard ceiling — which is precisely why it is the more dangerous number to fall in love with. A chain running at 73–77% is near practical full (you need slack for maintenance and check-in churn); pushing higher means either turning away business or the math is flattered by rooms taken offline for renovation.

Occupancy is sometimes in the PPT, often only stated on the call. Lemon Tree ran 73.4% in Q3 FY26 (Inve data, Q3 FY26), EIH about 72% in Q2 FY26 (Inve data, Q2 FY26), Chalet 67–77% across H1 FY26 depending on segment (Inve data, Q1–Q2 FY26). The signal is the combination, not the level. Occupancy and ARR both rising is a tight market with pricing power — the best state. Occupancy rising while ARR falls is discounting to fill beds. ARR rising while occupancy slips can be a deliberate trade-up — or the first sign demand is rolling over. Read them as a pair, always.

Room inventory and the pipeline — where tomorrow's RevPAR comes from

The room count is the capacity; the pipeline is the growth you cannot see in this quarter's revenue. A hotel chain grows two ways — by raising RevPAR on existing rooms (rate-led, high-margin, finite) and by adding rooms (room-led, the durable engine). The pipeline tells you how much room-led growth is coming and how it is being funded.

Room counts and pipeline live in the PPT and the call, never the P&L. IHCL's total portfolio stood at 617 hotels in Q3 FY26 — "361 operating hotels and 256 in the pipeline" in management's own words — with operational keys around 32,300 (Indian Hotels Q3 FY26 concall, via Inve data), against a stated Accelerate-2030 target of 700 hotels (Indian Hotels Q1 FY26 concall, via Inve data). Lemon Tree's inventory grew from about 16,385 rooms in Q3 FY25 to 21,942 in Q3 FY26 (Inve data), with a managed-rooms pipeline of roughly 35,000 (Inve data, Q2 FY26). Good looks like a credible pipeline being delivered on time — and the place to check that is whether last year's opening guidance actually opened, which is where the said-versus-did discipline earns its keep.

Owned vs managed — the same revenue, a completely different business

This single distinction changes how you should value the whole company, and it is invisible on a casual read. An owned hotel puts the property on the balance sheet: the chain takes all the RevPAR upside and all the fixed cost and capital risk — high margin at the property level, but capital-heavy and cyclical. A managed (asset-light) hotel is owned by someone else; the chain just runs it for a fee — far lower revenue per hotel, but almost no capital and steady, annuity-like fees that barely fall in a downturn.

The mix is disclosed in the PPT and discussed on the call, never derivable from the income statement alone. IHCL has tilted hard toward asset-light: "68% of our operating portfolio and 93% of our pipeline under managed or asset-light" formats (Indian Hotels Q4 FY26 concall, via Inve data), with a long-term target of about 70% capital-light, which Inve's Promise Tracker shows management has revised upward over time (Inve data, Q4 FY26). Chalet sits at the opposite pole — a predominantly owned, upper-upscale portfolio, which is exactly why its hospitality assets carry healthy property-level economics (its hospitality-segment EBITDA margin was about 46% in Q3 FY26, before corporate and interest costs — Inve data, Q3 FY26), while its annuity/Commercial Real Estate (office-leasing) segment runs far higher at about 83.5% (Inve data, Q3 FY26) — and why its consolidated result still swung to a loss in a soft quarter. Neither model is "better." Asset-light deserves a higher, steadier multiple because the earnings are less cyclical; owned deserves a cyclical multiple and a hard look at the balance sheet. Confusing the two is the most common valuation error in the sector.

F&B revenue and the margin line — the quality check

Food & beverage (restaurants, banqueting, weddings) is the second revenue stream, and it tells you something rooms don't: a hotel with a strong banqueting and F&B business has demand that is less tied to the room-occupancy cycle — a wedding books regardless of business-travel softness. It rarely gets its own line in the summary P&L; you pull it from the segment note or the PPT.

Which brings the threads together at the EBITDA margin — the one place the operating leverage finally shows on a statement you can pull straight from the filings. Read it at two levels. Property-level (or hotel-segment) EBITDA margin strips out corporate overhead and shows the raw economics of the rooms — Chalet's hospitality assets at ~46% (Inve data, Q3 FY26), IHCL's hotel-segment margin around 40.7% in Q3 FY26 (Inve data). Consolidated EBITDA margin is after everything — IHCL ran about 34.9% for FY26 (Inve data, Q4 FY26), Lemon Tree about 50% at the EBITDA line in Q3 FY26 given its owned-heavy model (Inve data, Q3 FY26). The gap between the two is the corporate drag; the direction of the consolidated margin across a cycle is the operating leverage made visible.

How do you value a business this cyclical?

The standard mistake is to value a hotel on a P/E ratio. Don't — and the reason is the operating leverage you just read about. Earnings at the bottom of the cycle collapse, so trailing P/E looks insanely high near the trough (tiny denominator) and deceptively cheap near the peak (fat denominator that is about to fall). A P/E tells you where you are in the cycle, not what the business is worth.

Two lenses fit a hotel far better:

EV/EBITDA is the workhorse, because EBITDA is less distorted by the heavy depreciation and interest that owned-hotel balance sheets carry, and enterprise value captures the debt that funds the rooms. But it comes with a cyclicality warning: a hotel near a cycle peak trades at a low EV/EBITDA on peak earnings that flatters it, and a high one on normalised earnings. IHCL's market cap was about ₹1,03,000 crore against roughly ₹2,769 crore of FY25 operating profit (Screener.in, FY25) — an EV/EBITDA comfortably in the high-30s, and a P/E near 60 (Screener.in). Those are not value multiples; they are the price the market pays for a high-quality, increasingly asset-light franchise at a strong point in the cycle. The discipline is to ask what the multiple looks like on mid-cycle EBITDA, not the last good quarter.

EV per room (or per key) is the cyclically robust cross-check, because the room count doesn't lurch with the cycle the way earnings do. Dividing enterprise value by the number of operational keys gives a per-room price you can compare across chains and against the replacement cost of building a comparable hotel. If EV/room sits far above what it would cost to build the same rooms, you are paying for brand and location premium — sometimes justified, sometimes the cycle talking. The right way to read both: normalise. Value a hotel on what it earns through an average year, not its best one — the same instinct that separates reported profit from real cash earnings in any business, applied to a sector where the gap is widest at exactly the wrong moment.

A worked case: said versus did

Take Chalet Hotels through FY25–FY26 — an owned-heavy, upper-upscale chain — because it shows both the operating leverage and the said-versus-did test in one company. (Illustration, not a view on the stock; figures from Inve data unless noted.)

The operating story first. ADR climbed from "almost INR 13,000" in Q3 FY25 (Chalet Q3 FY25 concall, via Inve data) — an 18% jump — and held in the ₹12,000–12,200 band through H1 FY26 (Inve data, Q1–Q2 FY26). RevPAR ran from ₹9,000 in Q3 FY25 to a Q4 FY25 peak of ₹10,909, up 21% (Inve data). Hospitality-segment EBITDA margin sat around ~40–48% across the period (Inve data). On the strength of that, revenue went from ₹377 crore (Q2 FY25, a loss quarter) to ₹895 crore (Q1 FY26, ₹203 crore profit). The pricing power was real, and the operating leverage did exactly what the model predicts.

Now the said-versus-did, which is where you separate a good operator from a good story. In Q3 FY25, management guided to "5,000 rooms in next few quarters." In Inve's Promise Tracker that room-inventory guidance is marked ghosted (Inve data) — the room count stayed at 5,000 (Inve data, Q1 FY26), the incremental additions did not arrive on the original timeline, and the specific commitment quietly stopped being repeated. The Delhi Airport hotel opening, guided for H1 FY27, shows as delayed (Inve data, Q2 FY26). Against that, the discipline held where it mattered most: the net-debt-to-EBITDA guidance of "3.0x–3.5x at peak debt" is marked achieved (Inve data, Q4 FY25), and the three-year ₹2,000 crore capex plan was revised up (Inve data) — capex rising is not a broken promise, it is a growth decision, but you want to see it land as rooms.

Here is the lesson the pairing teaches. The pricing power was genuine and the balance-sheet discipline was kept — but the room pipeline, the durable growth engine, slipped. A reader watching only the revenue and margin line would have seen an unbroken up-quarter and missed that the capacity additions underpinning the next leg were running late. The RevPAR told you the cycle was good; the ghosted room guidance told you to discount the growth narrative until the keys actually opened.

Red flags specific to hotels

• A peak-cycle P/E dressed as a growth story. When occupancy is near its ceiling and ARR is at record highs, earnings are at their most flattering and least repeatable. A "cheap" trailing P/E in a hot demand year is the classic cyclical trap.
• Revenue up, but entirely rate-led with occupancy already near full. That growth has no second act — you cannot raise rate forever once beds are full and you have nowhere to add more.
• Aggressive owned-hotel expansion at the top of the cycle. Hotels built near a demand peak open into the next downturn, carrying their full fixed cost into the softest years. Watch when the capex was committed, not just how much.
• A pipeline that is announced but never opens. Guided rooms that slip quarter after quarter — like Chalet's ghosted 5,000-room timeline above — mean the durable growth you are paying for is not arriving. Check openings against prior guidance.
• Occupancy flattered by rooms taken offline. A hotel under renovation can post a high occupancy on a shrunken available-room base. Read occupancy against the room count, not in isolation.
• F&B and other income propping up a soft rooms quarter. A one-off banqueting season or asset-sale gain in "other income" can mask weak underlying RevPAR.

Frequently asked questions

A repeatable workflow, then. Anchor on RevPAR and split it into ARR and occupancy across six to eight quarters — rate-led growth with occupancy holding is the durable kind. Read the owned-versus-managed mix to know which multiple the business deserves. Check the room pipeline against prior guidance to see whether the growth is real. And value it on mid-cycle EBITDA, not the last good quarter. Inve's KPI Screener lines up RevPAR, ARR, occupancy and margins across hotel companies with the trend and a data-confidence flag per number, and the concall summaries pull every forward commitment — room openings, RevPAR guidance, capex — into one guidance table per quarter, with the speaker and the quote. For a contrasting sector where the balance sheet, not the room rate, is the business, see how to analyse an NBFC.

So invert the question you bring to a hotel's results. Don't ask "was this a good quarter?" — in an up-cycle, every quarter looks good. Ask: if demand softened for two years, which of these earnings survive? The managed fees largely do; the owned-hotel margins largely don't. That single inversion tells you what you are actually buying.

And the owner's question, the one to sit with before buying any hotel stock: at what point in the cycle am I buying — and what must I believe about occupancy and rate three years out, not next quarter, for these rooms to still be earning what I am paying for them today? If the answer leans on the cycle staying hot, you have bought the peak and called it growth.

Related reading

• How to Analyse a Bank Stock (NIM, CASA, GNPA, ROA)
• How to Analyse a Building Products Company (Pipes, Tiles)
• How to Analyse a Cables and Wires Stock