Commercial buildings burn roughly 40% of U.S. electricity, and most of the waste is operational, not structural. We're an owner-operator — four Houston office buildings, about 421,000 SF, 19+ years of below-CPI operating expenses — and this is the exact retrofit pipeline we run on our own meters, in the order we'd do it again today.
Most energy-efficiency lists rank measures by how impressive they sound. That's backwards. A Texas commercial electric bill has four separately priced levers, and a measure is only worth doing when you know which lever it pulls:
The ranking below is ordered by payback within that framework: the operational measures that cost nothing come first, the capital measures come later, and verification — the measure that keeps the other seventeen honest — closes the list.
A foundation, fifteen measures, the zone delivery layer, and the daily optimization brain. Savings ranges are typical industry figures within each measure's own category — depth of impact on fan energy, lighting energy, plant energy — not shares of one total.
| № | Measure | What it is | Typical savings | Primary lever |
|---|---|---|---|---|
| ★ | Presence | Occupancy sensing — the foundation the rest builds on | 5–15% + enables the stack | kWh |
| 1 | The Schedule | Runtime discipline — equipment runs when people are there | 10–30% whole-building | kWh |
| 2 | The Cube Law | VFDs on supply & return fans | 20–40% of fan energy | kWh |
| 3 | The Pump Curve | VFDs on pumps & cooling tower | 15–40% of pump/tower | kWh |
| 4 | Trim & Respond | Static-pressure & supply-air-temperature reset | 10–30% of fan energy + | kWh |
| ⌗ | VAV & FPTU | The zone delivery layer | Enables №2, №4, №7 | kWh |
| 5 | The Warm-Up | Optimal & staggered start/stop | 5–15% of HVAC | NCP |
| 6 | The Deadband | Night & weekend setback | 5–12% of HVAC | kWh |
| 7 | The Fresh-Air Meter | Demand-controlled ventilation (CO₂) | 5–15% of heating/cooling | kWh |
| 8 | Free Cooling | Airside economizer | Shoulder-season plant relief | kWh |
| 9 | The Chiller Map | Plant staging & optimization | 5–15% of plant energy | kWh |
| 10 | The 4CP Game | Forecast · pre-cool · curtail the four ERCOT peaks | 10–30% of transmission | 4CP |
| 11 | Lumens per Watt | LED + lighting controls | 40–70% of lighting | kWh |
| 12 | The Power Triangle | Power-factor correction capacitors | kVA / demand charges | PF |
| 13 | Vampire Loads | Plug-load control | 2–5% whole-building | kWh |
| 14 | The Envelope | Window film, roof, air sealing | 2–8% of cooling | kWh |
| 15 | Reading a Model | Continuous M&V — the measure that protects the rest | Protects 5–15%/yr from decay | kWh NCP |
| ◆ | Ada | Daily optimization intelligence, with engineers in the loop | Runs & multiplies all of it | kWh NCP 4CP PF |
Every measure above is executed through building automation and scored against a weather-normalized baseline. The interactive versions — one teaching instrument per measure — live on our main site.
You can't get more efficient than off. Schedules, optimal start, and setbacks need no equipment — just the discipline to enforce them and the instrumentation to notice when they quietly revert. This tier is where every building should start, because it funds everything after it. One school on our platform saved $13,140 per year with zero capital — schedule waste, found and ended.
Fan and pump power scales with roughly the cube of speed, so slowing a motor 20% cuts its energy by about half. VFDs on air handlers, pumps, and cooling towers — plus the reset logic that actually earns the slowdown — are the highest-leverage capital measures in most offices. We run 27 rooftop units at one property alone on this logic; more on the control side in our guide to HVAC optimization for commercial buildings.
Demand measures don't reduce what you use; they reshape when and how cleanly you draw it. Staggered starts stop the 8:00 AM all-at-once spike from setting your distribution charge for thirty days; disciplined 4CP response — forecast, pre-cool, curtail — attacks the transmission line. At one of our buildings, peak demand dropped 36%, from 372 kW to 239 kW. Power-factor correction then multiplies both wins by shrinking billed kVA toward real kW.
Savings decay. Overrides accumulate, schedules revert after holidays, setpoints creep. Traditional measurement and verification is a $10,000–$25,000 study performed once a year — a photograph of a building that changes daily. Ours runs automatically every 15 minutes, scoring actual consumption against a weather-normalized baseline to IPMVP Option C with R² above 0.96. That's not a nicer report; it's the difference between savings that hold and savings that bleed out. If you're evaluating tooling, start with our energy management software (EMIS) guide.
We don't quote other people's case studies. These are our meters, on buildings we own and operate, verified continuously:
The execution layer matters too. Most landlords pay $50,000–$200,000 a year in BAS licensing; we built our own automation fabric on edge microcontrollers for under $50 per controller, zero licensing — which is why every measure on the list gets executed instead of value-engineered out. Why one integrated system beats bolted-on point solutions: integrated building automation systems.
Beyond our own walls, the RMS platform serves roughly 600 K-12 schools plus Harris County, with 15 buildings and 44 validated M&V reports in the verification program.
Runtime discipline — schedules, optimal start/stop, night and weekend setback — because it costs little or nothing and typically cuts 10–30% of whole-building consumption. After that: VFDs on fans and pumps (20–40% of fan energy, 15–40% of pump/tower energy) and LED lighting with controls (40–70% of lighting energy).
Our own portfolio: ~50% at the flagship over a decade (time-normalized, not weather-normalized), ~36% at a second building over 2018–2025, and 22.4% verified in roughly eight months at a building acquired in September 2025. Where you land depends on where you start — but double-digit reductions are normal when measures are executed and then continuously verified.
No. The top of the payback ranking is operational. One school saved $13,140 per year with zero capital spent — purely schedule discipline. Capital measures like VFDs and LED come after the free savings are banked and funding them.
IPMVP Option C, continuously: a weather-normalized baseline model (R² > 0.96 on our portfolio) scored against 15-minute interval data, every interval, automatically. Traditional M&V is a $10,000–$25,000 once-a-year study; continuous M&V catches drift the day it appears, so savings hold instead of decaying.
We'll show you what's hiding in your interval data: where your kWh, demand, 4CP, and power-factor money is going, and which of the 18 measures pays back first on your building. Free RMS trial — all it needs is one recent utility bill.