By Jason Charles Guck, Business Development, Delta Edge CI
Lighting has a credibility problem in commercial energy work. It was the first measure everybody did, it has been pitched to every facility manager in the country several times over, and most operators believe they already handled it. The usual response to a lighting conversation in 2026 is that the buildings were done years ago.
Usually that is half true, and the half that is not true is where the money is.
What the national data actually says
According to the Commercial Buildings Energy Consumption Survey, LED bulbs were reported in 9% of commercial buildings in 2012 and in 44% of commercial buildings in 2018. That is fast adoption, and it is the number usually cited to argue the opportunity has closed.
Read the measure carefully though. It counts buildings where LEDs were reported, not buildings converted to LED. A building with LEDs in the lobby, the exterior wallpacks and the conference rooms, and linear fluorescent everywhere else, counts in that 44%. In the same survey, standard fluorescent lighting remained the most common type in commercial buildings, though it was reported in 16% fewer buildings in 2018 than in 2012.
Partial conversion is the normal condition, not the exception. Most buildings that “did lighting” did the visible, accessible, easy-to-justify fraction and stopped when the obvious payback ran out or the budget cycle ended.
The scale is worth holding in mind. The Department of Energy’s 2020 Lighting Market Characterization counted 1.6 billion lighting installations in the commercial sector, about 20% of all installations in the country, but roughly 69% of the electricity consumed by lighting in U.S. buildings, because commercial fixtures run at higher wattage for far more hours. DOE’s forecast has LEDs reaching 84% of installations by 2035, which is another way of saying the conversion is not expected to be finished for another decade.
Where the remaining load hides
The unconverted fraction is predictable, because the reason it was skipped is always the same. It was harder to reach, harder to justify, or nobody looked.
- High-bay and warehouse space. Metal halide and high-output fluorescent fixtures at twenty feet require a lift and a shutdown window, so they get deferred. They are also among the highest-wattage fixtures in the building, running the longest hours.
- Back of house. Kitchens, prep areas, storage, mechanical rooms, loading docks. No customer sees them, so they never made a renovation scope, and they frequently run continuously.
- Exterior and parking. Often converted at the pole heads and missed at the building perimeter, or converted years ago with early LED product that has degraded.
- Stairwells and corridors. Code-required continuous operation at 8,760 hours a year. A single fixture here uses more annual energy than several in an office that runs fifty hours a week.
- Refrigerated cases and walk-ins. Fluorescent in a cold space performs poorly and the waste heat is removed by the refrigeration system, so the fixture is charged twice.
The pattern is that the fixtures left behind are disproportionately the ones with the most operating hours. That is not a coincidence. They were skipped because access was difficult, and difficult access usually means a space that runs on its own schedule rather than an occupied schedule.
The controls half, which is mostly unclaimed
The larger missed opportunity in 2026 is not the lamp. It is that the retrofits done between 2014 and 2020 replaced fixtures one for one and left the switching exactly as it was.
A fluorescent fixture with no occupancy sensor becomes an LED fixture with no occupancy sensor. The wattage dropped, the hours did not. In a storage room or a rarely used corridor, the hours are the larger term. Occupancy sensing, daylight harvesting near glazing, and scheduled dimming in low-traffic periods are all available on current LED product natively, and in many buildings they are installed and never commissioned, which is the same as not having them.
There is a second, less obvious return. Lighting is a coincident load: it is on during the same hours the building sets its monthly peak. Reducing connected lighting load reduces billed demand in kilowatts, not only consumption in kilowatt-hours. On a rate with a meaningful demand charge, that is a second line of savings that a simple kilowatt-hour payback calculation misses entirely, and it is the reason a lighting project sometimes pencils better than the spreadsheet says.
Auditing what you have in an afternoon
Before anyone quotes a project, an operator can establish whether there is anything here worth pursuing. Walk the building with a phone and answer four questions per space.
- What technology is in the fixture. A linear tube with visible pins and a ballast hum is fluorescent. If you are unsure, hold a phone camera to it; fluorescent and older LED often show visible flicker on the sensor that the eye does not catch.
- How many hours a year it actually runs. Not the occupied schedule, the fixture schedule. Continuous spaces are where the value concentrates.
- Whether anything turns it off other than a person. If the answer is no in a space that is empty most of the time, the controls opportunity is larger than the lamp opportunity.
- Whether it was converted long enough ago to have degraded. Early LED product in exterior and cold applications has often lost meaningful output, which shows up as complaints about dark parking areas rather than as an energy question.
Rank what you find by annual operating hours, not by fixture count or by how visible the space is. A hundred office troffers on a fifty-hour week are worth less than a dozen high bays running continuously in a warehouse.
Why this still gets done last
Lighting is usually the measure with the shortest payback in a building and the one operators are least interested in discussing, because it feels like old news and because the sales history around it has been poor. That combination is exactly why it survives as an opportunity. Measures that are boring and previously oversold do not get revisited.
It is also one of the measures that can be structured without capital, which removes the budget objection that stopped the first attempt. I covered how that structure works in an earlier post on reducing energy costs without capital expenditure.
The test is simple. Walk the spaces nobody shows a visitor, count what is running continuously, and check whether anything but a human switch controls it. If the answer is a room full of fluorescent tubes on a light switch, the easy money is still on the table in your building, whatever was done in 2016. If you want that walk done properly and the numbers put against your actual rate structure, that is work we do at Delta Edge CI.