Utility cost benchmarking: comparing cost per square foot across buildings without fooling yourself

Every portfolio has the spreadsheet: buildings down the side, annual utility spend divided by area across the top, sorted. It is used to decide which buildings get attention, and it is usually misleading, because the number mixes four things that move independently: how much energy the building uses, what it pays per unit, how the year's bills happened to fall, and how the area was measured.

Utility cost benchmarking done properly separates those four before comparing anything. This article describes the method used in the monthly energy review, where the result is called the utility cost intensity (UCI), and shows how it sits next to energy use intensity so a reader can see whether a building is expensive because it uses a lot or because it pays a lot.

Start with intensity, not spend

ENERGY STAR defines energy use intensity as a building's energy use divided by its gross floor area, expressed per square foot or square metre per year. Utility cost intensity is the same idea applied to cost: total utility cost for the year divided by the same gross floor area, in dollars per square foot or square metre. The two only make sense together. A building with an average EUI and a high UCI has a price problem, a tariff problem or a billing problem. A building with a high EUI and a high UCI has an energy problem.

• EUI: equivalent kWh per square foot per year, all fuels converted to a common unit.
• UCI: dollars per square foot per year, all commodities, before or after tax, stated consistently.
• GEI: kilograms of carbon dioxide equivalent per square foot per year, from consumption and published factors; see GHG emissions from utility data.
• WUI: cubic metres of water per square metre per year.

Four things to strip out before comparing

1. Calendar

Bills straddle month ends and some arrive late or doubled up after an estimated read. An annual cost that simply sums the bills dated in the year may contain eleven or thirteen months of consumption. Assign every bill to a reporting month by one rule, then sum twelve reporting months. The validation article describes the rule and the estimated-read case.

2. Tax and pass-throughs

Decide whether UCI is before or after HST and apply it to every building. A public-sector owner that recovers HST and a private owner that does not will make different choices, but a single portfolio must make one. The same goes for one-time adjustments, security deposits and late-payment charges, which belong in a separate line, not in the intensity.

3. Price

Two Ontario buildings with identical electricity consumption can have different bills because they are in different rate classes, because one is Class A for global adjustment and the other is Class B, or simply because their bills were issued on different days and carried different published rate estimates. The IESO's global adjustment pages explain why the rate moves monthly. Keep unit cost (dollars per kWh, per cubic metre) as its own metric so price effects are visible rather than buried in UCI.

4. Weather and area

A colder year raises gas cost in every building, but not equally. Weather normalization using heating degree-days, which Environment and Climate Change Canada defines as the number of degrees the daily mean temperature is below 18 degrees Celsius, is applied to the energy side; the weather normalization article covers the method. Area is the quieter error: gross floor area, rentable area and conditioned area differ by ten to twenty percent in many buildings. Use one definition, record which one, and use it for every intensity.

What public benchmarks exist

For energy, Natural Resources Canada publishes benchmarking snapshots from Portfolio Manager data. Its office snapshot, based on about 7,500 buildings as of December 2022, reports a median site energy use intensity of 1.1 GJ per square metre and a median source EUI of 1.5 GJ per square metre, with source EUI ranging from 0.9 GJ per square metre at the 10th percentile to 2.5 at the 90th. That spread, nearly three to one between the best and worst deciles of ordinary Canadian offices, is the reason benchmarking is worth doing at all.

For cost there is no equivalent public Canadian data set, because cost depends on tariffs, province and rate class. Cost benchmarking is therefore mostly internal: each building against its own history and against the rest of the portfolio, with the price components isolated. Where a portfolio has several buildings in the same utility territory and rate class, the comparison across them is meaningful; across provinces it is mainly a comparison of tariffs.

A practical way to use the public data is as a sanity check rather than a target. If a building's site EUI is near the NRCan median for its type, the benchmark is telling you the building is ordinary; if it is near the 90th percentile, the benchmark is telling you to look at the equipment and the schedule before looking at the tariff. The internal comparison, across your own buildings and against each building's own history, is where the monthly work happens.

How the review presents it

The monthly review shows, for each building, consumption and cost for the month and year to date against the prior year, the intensities, and the unit cost, with the portfolio average as a reference line. The annual benchmark report ranks the buildings on EUI and UCI, prints the floor area definition and the tax treatment, and lists which buildings moved materially from the previous year and why. The portfolio view article shows how the data gets there.

FAQ

Should UCI include water?

Include every commodity the owner pays for, and also report per-commodity UCI. A building with cheap energy and expensive water is a different problem from the reverse.

Which floor area should we use?

Gross floor area, because it is the basis for ENERGY STAR Portfolio Manager and for the municipal reporting programs that use it. Record the source of the figure with the building. See the Portfolio Manager guide.

Can we benchmark with bills alone?

Yes. Twelve validated months of bills per meter produce EUI, UCI, unit cost and the annual comparison. Interval data adds the operational detail in the heat map view but is not required for the benchmark. Request a review to see it on your buildings.