EvaluateGuides by Industry6 min readUpdated

Key takeaways

  • EIA found packaged air conditioners in 74% of mercantile buildings, so rooftop units dominate retail HVAC.
  • Each 1% of store electricity is small at one site but meaningful across hundreds.
  • Pilot three to five stores against matched control stores, then roll out in waves with spot checks.
  • DOE and LBNL estimate HVAC faults waste about 0.7 quads of energy a year in U.S. commercial buildings.

For a retail or multi-site chain, HVAC efficiency is a portfolio decision. Each store has a few rooftop units, maybe some refrigeration, and a maintenance contract managed from headquarters. No single store's savings justify much attention, but the same measure repeated across 50, 200 or 1,000 locations can be one of the larger cost lines a facilities team controls. This guide covers the retail energy profile, the portfolio math, where store systems lose efficiency, and a pilot-then-rollout method for testing any measure, including a no-downtime treatment like CRYOGENX4.

The retail energy profile

In EIA's 2018 CBECS, mercantile buildings (stores, strip centers and malls) made up 9% of commercial buildings, 11% of commercial floorspace and 14% of major fuel consumption. They averaged 88.4 thousand Btu per square foot, with enclosed malls and strip centers at 111.0 and other retail at 64.1 (EIA CBECS). On electricity alone, mercantile buildings averaged 16.7 kWh per square foot per year, against 12.6 for all commercial buildings (EIA CBECS Table C22).

The equipment is remarkably uniform: packaged air conditioners, usually rooftop units, served 74% of mercantile buildings, and residential-type central air conditioners served another 21% (EIA CBECS). That uniformity is what makes chains good candidates for standardized programs.

The portfolio math

Illustrative assumptions, not a measured result:

  • A 20,000 sq ft store at the CBECS mercantile average uses about 334,000 kWh of electricity a year
  • At the U.S. average commercial price of 14.53 cents per kWh in July 2026 (EIA), that is about $48,500 per store
  • Each 1% of store electricity is worth about $485 a year at one store, and about $97,000 a year across 200 stores

HVAC is only part of a store's load, so map each measure to the end use it affects. Use your own interval data and tariffs, and the payback, ROI and NPV guide, to compare programs.

Constraints for multi-site operators

  • Trading hours come first. Work that disrupts customers, closes a department or makes a store uncomfortable is a non-starter.
  • Distributed decision-making. Facilities, energy, operations and finance teams all have a say, and regional managers may need to approve site visits.
  • Mixed leases. Some stores are owned, some leased; in some leases the landlord owns the rooftop units.
  • Varied equipment age. Even with standard prototypes, units range from new to near end of life, and maintenance quality varies by region and contractor.
  • Data gaps. Many chains have monthly utility bills for every store but interval data or unit-level metering for only a few.

Where store systems lose efficiency

A DOE-funded study led by Lawrence Berkeley National Laboratory, which analyzed fault data from air handlers, terminal units and rooftop units, cites an estimate that HVAC faults in U.S. commercial buildings waste about 0.7 quads of energy a year, worth nearly $14 billion (DOE/LBNL). In stores, the usual suspects are:

  • Economizer faults: dampers stuck open or closed, failed sensors.
  • Schedules and setpoints that drift away from store hours.
  • Dirty condenser coils and filters, especially where maintenance visits are infrequent.
  • Refrigerant charge errors after repairs.
  • Internal heat-transfer loss as oil carried from the compressor coats the inside of coil tubes, reducing capacity in older units. See what is oil fouling.

Our rooftop units guide covers RTU-specific degradation in detail. For stores with food refrigeration, see grocery and supermarkets.

Practical measures for a portfolio

MeasureRollout methodNotes
Central schedule and setpoint standardsEnergy management system or thermostat policyCheapest; enforce centrally
Economizer repairAdd to PM scopeCheck at every visit
Filter and coil cleaning standardsContract specification with photo evidenceStandardize frequency by climate
Refrigerant-side heat-transfer treatmentPilot, then waves by regionNo shutdown, per the company
RTU replacementCapital plan by age and conditionSee treatment vs replacement

How a no-downtime treatment fits

CryogenX4 describes CRYOGENX4 as a one-time nano-fluid treatment injected into running HVAC-R equipment. The company says it removes insulating oil from internal coil surfaces, returns it to the compressor sump, conditions the metal for better heat transfer and improves the lubricity of the existing oil. It states installation is done while the system runs, with no downtime and no modifications, that most installs take one day, and that the work is done by trained, certified technicians. For a store, that means installs can happen during trading hours without closing the floor.

CryogenX4 reports energy savings of up to 30% and typical payback of 12 to 36 months, and says it has treated more than 50,000 units. Those are company statements, and results vary by equipment condition. For a chain, the right question is what happens on your stores, measured your way.

Pilot a few stores, then roll out

  1. Select three to five pilot stores that represent your most common prototype, climate and equipment age. Pick a matched control store for each: similar size, hours, climate and equipment, left untreated.
  2. Get the data in place. Ideally, interval meter data or unit-level loggers on rooftop units. If you only have monthly bills, extend the measurement period and use weather normalization.
  3. Service before baseline. Bring pilot and control stores to the same maintenance standard first, so the pilot measures the treatment, not a cleanup.
  4. Baseline for several weeks, treat the pilot stores, then measure through comparable weather.
  5. Compare pilot against control using a difference-in-differences approach: the change at pilot stores minus the change at control stores. Normalize for cooling degree days. See baselines and weather normalization.
  6. Set rollout criteria in advance: which result justifies a wave of 20 stores, then 100, and which unit ages or conditions are excluded.
  7. Roll out in waves by region, with continued measurement at a sample of stores in each wave to confirm results hold across climates and contractors.

The CryogenX4 pilot program page describes how a small pilot is structured, and CryogenX4's commercial page gives an overview for commercial buildings.

Use the data you already have

Before buying loggers, look at what the chain already collects. Many retailers have an energy management system or smart thermostats that record rooftop unit run time, space temperatures and setpoints by store. Utility bill management services often hold years of monthly data for every location, and some utilities provide 15-minute interval data on request. Run time per cooling degree day, by unit, is a surprisingly good screening metric: stores whose units run far longer than identical units elsewhere in the same climate are candidates for maintenance attention first, and for testing efficiency measures second.

Choosing control stores well

The control group is what makes a multi-site pilot credible, so choose it carefully. Good control stores share the pilot store's prototype and floor area, climate zone, trading hours and rough sales volume, and their rooftop units are of similar age and model. Avoid stores with planned remodels, equipment replacements or schedule changes during the test period, since any of those would distort the comparison. If you can, choose controls before selecting pilot stores, or assign stores to pilot and control at random from a matched list. That removes the temptation to pick "good" stores for the pilot.

Common pitfalls in multi-site pilots

  • Measuring too short a period, so results depend on a few hot or mild weeks
  • Servicing pilot stores more thoroughly than control stores
  • Relying on monthly bills that cover different numbers of days
  • Rolling out before results are reviewed by someone independent of the vendor

What to put in the rollout specification

Once a pilot justifies rollout, the program lives or dies on consistency across hundreds of contractor visits. Write a specification that covers:

  • Eligibility: which unit types, ages and conditions qualify, and which are excluded (for example, units scheduled for replacement within two years or with open repair tickets).
  • Pre-treatment checks: filters, coil condition, charge and economizer operation recorded for every unit.
  • Documentation: unit nameplate, before and after pressures and amp draw, and photos, uploaded to your work-order system.
  • Scheduling: work windows that respect trading hours and store manager sign-off.
  • Measurement sample: a fixed share of stores in each wave that keep logging equipment for verification.

Reporting results to leadership

Executives want three things: what was spent, what was saved, and how confident you are. Report savings per store and across the treated portfolio, separate measured stores from those estimated from the pilot, and show the control-store comparison. If your company reports Scope 2 emissions or tracks ENERGY STAR scores, include those effects too, using the same measured kWh.

Next step: Pull a list of stores with interval data, prototype and rooftop unit age. Choose three to five pilot stores and matched controls from that list before talking to any vendor.

Frequently asked questions

How many stores do we need in a pilot?

Three to five pilot stores with matched controls is usually enough to see whether a measure works on your prototype. More stores, or longer measurement, help if you only have monthly bills.

Why use control stores?

Weather, sales and operating changes affect every store. Comparing treated stores against similar untreated stores over the same period separates the measure's effect from everything else.

What if the landlord owns our rooftop units?

Work out who pays the electric bill and who is responsible for maintenance. If you pay the bill, you capture savings, but the landlord may need to approve work on their equipment.

Should we fix economizers before testing a treatment?

Yes. Bring pilot and control stores to the same maintenance standard first, so the pilot measures the treatment rather than repairs.

Sources

  1. Mercantile Buildings (2018 CBECS) — U.S. Energy Information Administration
  2. 2018 CBECS Table C22: Electricity consumption totals and conditional intensities by building activity subcategories — U.S. Energy Information Administration
  3. Electric Power Monthly, Table 5.6.A: Average Price of Electricity to Ultimate Customers by End-Use Sector — U.S. Energy Information Administration
  4. Evaluation of Fault Prevalence in Commercial Buildings (BTO Peer Review 2021) — U.S. Department of Energy / Lawrence Berkeley National Laboratory

Keep reading

See what your equipment could save

CryogenX4 is a one-time treatment installed while your system runs. Start with a pilot on a few units, measured against a baseline, before you commit to a building or a portfolio.