Professional Photography Lighting Since 1993 X-System 2.4GHz · Worldwide Dealer Network

The Cheapest Lighting Retrofit Is the One You Get Right the First Time

I'm the quality and brand compliance manager at our company. Every lighting and electrical package crosses my desk before it goes to a customer—roughly 250 builds a year. In 2024, I rejected 31% of first deliveries for spec issues.

If I could pass on one thing to anyone signing off on lighting procurement, it's this:

The cheapest lighting retrofit is the one you get right the first time. The second, 'cheaper' install is the one that eats everything you thought you saved, plus the downtime.

That's not a slide anyone wants to hear at a bid meeting, because it's boring. Boring doesn't win the meeting. But it's what four-plus years of looking at fixtures, switches, and sensors has taught me.

The real cost of 'we'll fix it later'

Every site has a 'later' list—a run of daisy-chained wiring, a switch plate reused from the old fit-out, a sensor that got installed at head height because nobody checked the plan. The thing about a 'later' list is not that it fails. It's that by the time it fails, you're already in stakeholder territory: someone has signed off, tenants have moved in, the scaffolding is gone.

One project I walked in 2024 had a run of Eaton light switch wiring that nobody had verified—maybe 60 plates across a mixed-use floor. We pulled 60, found 14 where the terminal torque was just tight enough to feel right in the hand and not tight enough to pass the spec. The drywall was already closed. When a tenant calls about a circuit tripping, the problem isn't a switch anymore. It's a wall. That wall costs around $4,000 to open and close.

Most of that is written down. An Eaton dimmer and a standard Eaton toggle are not the same component, and the instructions are not vague about the neutral requirement or the torque spec. The failures I see are almost never product failures. They're people-pressure failures—somebody had a deadline, the trim carpenters were waiting, and the wiring got done from memory instead of from the sheet.

Sensor placement beats sensor brand

Most of what we install is occupancy and motion sensing, and the phrase we keep hearing on design calls is 'Eaton speed sensor location.' In lighting, that 'speed' means the seconds between someone walking in and the lights coming on. It's almost entirely a placement question.

The bad placements have a pattern. Too close to an HVAC return, so the coverage drifts by time of day. Angled on the ceiling so the actual detection pattern covers the corridor and not the room. Anywhere within eight feet of the panel, where the control logic has a one-second lag built in. And anything behind a shelving line, which is why warehouse installs drop so many triggers in the first week.

On a 60,000-square-foot warehouse project in 2023, we dropped 28 wall-mounted motion sensors by 18 inches across the whole floor. Mis-triggers fell by about two-thirds, and commissioning time dropped by roughly a week. We didn't change the product. We changed where it sat.

Beach chandelier and the Snapchat Spotlight problem

Two things remind me how often 'later' gets baked into a spec. The first is decorative fixtures—the beach chandelier on the rendering that looked stunning, and by the time it's installed the finish has started lifting before the fabrication warranty even matters. Coastal spec has to be written, not assumed: 316 stainless, polyester powder coat on the housing, and a salt-spray record at sign-off. A good photo and a three-year-old fitting are not the same object.

The second is the short-form video. You've probably seen the Snapchat Spotlight clip where somebody kills the breaker, pulls the trim down, swaps in a retrofit kit, and the lights come on in under a minute. The Snapchat Spotlight format makes it look like a five-minute job. Honestly, for a homeowner with a conventional ceiling, it usually is. What those clips never show is a fixture sitting in an insulated ceiling with no IC rating, or a 12-volt MR16 where someone expects a line-voltage bulb, or a retrofit can that's the wrong size and now sits proud of the ceiling. If you actually search how to replace recessed lighting bulb, you'll get twenty videos and most of them skip the last check—reading the label inside the can, not matching the bulb to the picture.

But volume spec is cheaper

Here's the pushback, and sometimes it's correct. If your building is a temp trailer and the lights only need to work four hours a day, five days a week, a budget-tier fixture is probably fine. I'm not going to fight you on that.

But once a project involves emergency lighting, occupied-hours work at night, or ceilings that are expensive to touch after close-out, cheap spec just moves cost forward. Standard-grade fixtures differ in the parts you're about to never see again—switch contact material, terminal plating, sensor lens clarity, coat thickness on the coastal hardware.

We had to replace 1,800 fixtures in 2022 after the die-cast housings started cracking within a few months of install. The redo was two days of tear-out, two days of reinstall, patching, a rented lift, and six days of delay. That bill was $22,000. The fixtures saved about $4,000 at the time. Adjust that for every dollar 'saved' and the redo cost about five and a half.

What I've become stricter about is not the price of the fixture. It's how close the bid is to the install reality. An expensive fixture installed at the wrong height and tightened to the wrong torque is less efficient than a mid-grade fixture with a checklist. Efficiency isn't the speed of the purchase order. It's the number of times you never have to come back.

Am I overstating it?

Some of it, yes. I can only speak to our slice of the market—a mid-size B2B operation with predictable order patterns. If you're dealing with wildly seasonal demand or one-off sites, the calculus might be different. Established portfolio buyers regularly get budget-tier pricing we never see, and those prices are real.

Honestly, I'm not sure why some contractors consistently get the first install right while others coast to the punch list. My best guess is it comes down to whether they have a pre-close check on terminations and sensor coverage—not a gut feel, an actual signed sheet. If someone has a better read on that, I'd like to hear it.

The lesson I'll take: on projects that repeat—multi-room hotels, multi-site retail, schools, warehouses—standardizing the spec and the install sequence lowers total cost. On projects that need hand-fitting, the traditional way of doing each one individually is probably correct, and I'll keep saying so even when it costs us a bid. The trick is knowing which world you're in before the drywall goes up.

Bottom line: the only place you can actually save money is the first install. The second one is just extra cost on top of the hours you already paid for.

That's the view from someone who spends most of her working day rejecting things that weren't quite right. Take the parts that fit your situation.