Summary: how to verify HVAC energy savings
You can't verify HVAC savings just by comparing total usage or bills before and after installation, because outdoor temperature, foot traffic, and the number of units running differ every day. The credible approach is "condition-matched comparison": first filter for data with the same time of day, similar outdoor temperature, and the same number of HVAC systems running, then compare average usage before and after, covering different seasons where possible. Reports should clearly separate measured results from estimates and list unfavorable data points.
Why bills mislead
HVAC energy use depends on many factors, and they're usually different before and after installation:
| Factor | How it misleads |
|---|---|
| Outdoor temperature | Cooler months after install overstate savings |
| Hours and traffic | Seasons and promotions shift load |
| Units running | 10 units before, 8 after: the drop in usage isn't the product's doing |
| Setpoint | A raised setpoint muddies the result |
How matching works
Measure first
Start recording energy use before installation, covering at least one full operating period. Without pre-installation data, there's no baseline to compare.
Fix setpoints
Keep the setpoint the same before and after, so savings from "changing the temperature" aren't counted as the product's effect.
Filter on 3 conditions
Compare only data with "the same time of day, a similar outdoor temperature range, and the same number of HVAC systems running."
Compare averages
Within the filtered data, compare average usage before and after to calculate the savings rate. You can also compare by outdoor temperature band to see whether the effect is consistent across temperatures.
Across seasons
HVAC load varies a lot by season; cover at least two seasons for representative results.
Whole-system view
Besides the main test units, you can compare usage for the whole system as supporting evidence, noting that it wasn't condition-matched.
In practice: "Retail store HVAC energy savings, measured" uses exactly this method, verifying across spring and summer and listing results for each of 13 outdoor temperature bands.
Estimating yearly savings and payback
Test periods usually last only a few months, so annual benefits are "estimates," not measurements. A common estimation method:
| Item | Calculation |
|---|---|
| Yearly energy saved | Rated kW × PF × load factor × aging × hours × saving rate |
| Yearly cost saved | kWh saved × price |
| Yearly CO₂ cut | kWh saved × emission factor |
| Payback period | Cost ÷ yearly savings |
Every parameter affects the result: the load factor should match the region and site, the aging factor depends on equipment age, and the electricity price and emission factor should state which year's figures are used. Exaggerate any one parameter and the estimate is distorted.
5 things to check in a report
A baseline?
Without a baseline, there's nothing to compare.
Same conditions?
Were time of day, outdoor temperature, units running, and setpoint controlled?
How long, how many seasons?
A test of just a week or two, or a single season, has limited representativeness.
Bad data shown?
Be wary if every band shows a decrease; an honest report lists data points where usage went up.
Estimates labeled?
Annual savings and payback are estimates; the formula and parameters should be stated.
FAQ
How do you verify HVAC energy savings credibly?
Use condition-matched comparison: first filter for data with the same time of day, similar outdoor temperature, and the same number of HVAC systems running, then compare average usage before and after installation, keeping setpoints the same and covering at least two seasons. Comparing only electricity bills is easily skewed by weather and foot traffic.
Why didn't my electricity bill drop noticeably after installing an energy-saving product?
Higher outdoor temperatures in the same period, or longer business hours or more units running, may have offset the savings. Only by comparing data under the same conditions can you see the product's own effect.
How long should an HVAC energy test run?
We recommend covering at least one full operating period each before and after installation, spanning two seasons. HVAC load varies a lot by season, so results from a single season have limited representativeness.
Is the annual energy savings figure in a report measured?
Usually not. Test periods last only a few months, so annual energy savings, cost savings, and payback are mostly estimated by multiplying the measured savings rate by parameters such as load factor and operating hours. The report should state the formula and parameters.