Why EU and GCC hotel benchmarks are rarely directly comparable

The first question is whether the two properties deliver the same service from the denominator being used. Occupied room-nights is the conventional denominator, and it captures a very different proportion of total activity in different property types.

Consider two properties with identical room counts and identical occupancy. The first is rooms-only with a breakfast offer. The second has a large restaurant open to non-residents, extensive conference facilities, a spa, a pool, and on-site laundry servicing both the hotel and an outside contract. The second delivers materially more service per occupied room-night, and consumes more resource doing so. Its intensity figure will be higher, and the difference reflects its service mix rather than its efficiency. The same applies to covers, spa and leisure usage, events hosted, and any outsourced service provided to third parties.

The second question is what the reported figure includes. Two properties can report the same nominal scope and differ substantially in what falls inside it.

Laundry is the clearest example. A property with on-site laundry consumes water, energy and chemicals directly, and those appear in its consumption. A property that outsources laundry has moved that consumption off-site, into the supplier's figures rather than the property's. Comparing the two on reported consumption is comparing a property that does the work with one that has contracted it out — and the second will usually report lower figures for reasons that have nothing to do with how efficiently it is operated.

Other boundary differences compound this — whether the figure includes common areas and grounds, tenant-operated outlets, back-of-house, purchased steam or chilled water, and multi-use building shares. In the GCC context, the source of water — municipal supply, desalination, or on-site production or treatment — is a further difference, because desalinated water carries a substantially different energy and cost profile from water abstracted and treated locally. Contracted services — cleaning, laundry, maintenance, landscaping — may sit outside the property's reporting boundary entirely, even though the property caused the consumption.

Where a comparison genuinely needs to be made, the workflow is sequential and each step can fail.

Establish the purpose first: a comparison for internal target-setting has different requirements from one intended to support an investment case, and one intended for publication has higher requirements still. The purpose determines how much reconciliation is warranted.

Establish that the service outputs are alike, or adjust for the difference. Where one property serves significantly more non-resident covers, the denominator may need to incorporate that output rather than counting only room-nights. Where the mix cannot be reconciled, the comparison is not available.

Establish that the boundaries match. Add back outsourced consumption where the data exists, or record the difference explicitly, confirming what is inside each figure area by area. Normalise for the drivers not under management control: cooling degree days, and occupancy where the property is not full.

Then compare — and state the residual uncertainty. What remains is a comparison between two properties as alike as the data allows, with the remaining differences named rather than assumed away.

Small samples are the most common case. A comparison between two or three properties is a comparison between two or three individual properties; it is not a benchmark, and presenting it as one implies a population that does not exist. The differences between individual properties — site orientation, building age, plant condition, service mix, local climate — are frequently larger than the effects being studied.

Where boundaries cannot be reconciled, the same applies. If one property outsources laundry and the other does not, and the outsourced consumption cannot be obtained, the two figures are not comparable, and an adjustment based on assumption would produce a number without a defensible basis.

The useful alternative is a structured description: each property's consumption, its service outputs, its system boundary and its drivers, side by side without a computed ranking. This lets a reader understand each property on its own terms and see where genuine differences lie.

This matters particularly across the EU and GCC markets, where the differences in climate, water source and service expectations are systematic rather than incidental. Treating properties in different markets as though they shared a common context produces a comparison whose differences reflect the climates and the infrastructure, not the management. Published labelling criteria [A] and national conservation reporting [F] describe their own contexts, and neither supplies a country-level emission factor that would make the two commensurable.