Start with the situation, not the product
I've been handling commercial lighting orders for nine years—hotel lobbies, offices, retail rollouts. I've personally made and documented 23 significant mistakes, totaling roughly $46,000 in wasted budget. That's why I keep a checklist now, and that's why I answer lighting questions differently than I did in 2017.
Nearly all of that wasted money came from one habit: treating similar-sounding questions as if they were the same question.
Three questions get mixed up constantly. “Which lens do I need for a tighter beam?” “Can I replace this chandelier with an LED fixture?” “Where can I cut LED strip lights?” They all sound like product questions. In practice, they are three different scenarios, each with its own decision path.
Scenario 1: The beam looks wrong and you're ready to blame the lens
In my first year, I selected a spotlight for a retail display wall based on the spotlight graphic on the data sheet. The graphic looked like a crisp, narrow cone. The real fixture produced a beam almost twice as wide. The wall looked washed out, not dramatic. The vendor's answer: “That graphic is just for illustration.”
A spotlight graphic is not photometric data. If a fixture spec doesn't include an IES file or an LM-79 report, treat the visual as decoration. That lesson cost me one client's patience.
Then I over-corrected and started adding converging lenses to everything. That's when I learned why the converging lens focal point actually matters. A converging optic only produces a tight, controlled beam when the LED is positioned correctly relative to the lens. If the emitter sits inside the focal point—closer to the lens—the beam doesn't get tighter. It widens, and you can also see artifacts of the chip instead of a clean pool. I ended up throwing away $1,800 worth of optics on a single order.
When someone asks about the converging lens focal point, they're usually trying to focus light into one small area. That's the right instinct. But the emitter position and the lens need to be matched. The math in textbooks uses a point source; real LEDs have a chip surface. Change the chip and the same lens will give you a different beam.
Now for the wording that confuses people: When someone asks about a diverging lens object inside focal point, they're probably picturing a magnifying glass. A magnifying glass is a converging lens. A diverging lens does the opposite. It spreads the light, creates a virtual image that remains upright and reduced, and does not make a tight spotlight. If your goal is a smooth wall wash, a diverging optic can be excellent. If your goal is an accent beam, you've come to the wrong scenario.
At Focal Point, we make both precision optics and complete fixtures. That mix means I get to watch this physics-versus-purchasing confusion weekly. The rule I use now: decide whether you need a beam or a wash before discussing lens type.
Scenario 2: You are replacing a chandelier and expecting a simple swap
The chandelier replacement mistake happened in September 2022. It involved 16 fixtures, six weeks of delay, and $3,400 in extra logistics.
The client asked for a “same look, LED” chandelier replacement. We compared dimensions, finish, and lamp count. Looked perfect. What we didn't compare: driver location, dimming method, and ceiling structure.
The old fixture used candelabra-base bulbs and was controlled by an existing incandescent dimmer. The new chandelier had integrated LED modules, a remote driver, and 0-10V dimming. There was no 0-10V wire in the ceiling and no room for the driver behind the canopy. It was not a fixture problem. It was an infrastructure problem.
Before you approve a chandelier replacement, inspect three things:
- Mounting and weight. What is the ceiling structure? A 4-inch octagon box cannot hold every chandelier. Public spaces often need a safety cable.
- Driver and controls. Where will the driver be mounted? Does the existing wiring support the control signal, or do you need to add a control wire?
- Serviceability. Integrated LED modules are convenient, but if one fails in five years, how much will replacement labor cost? Ask for replaceable modules or a documented spare plan.
A room's chandelier is its visual focal point, so people choose it like furniture. But it's also electrical infrastructure. Legacy dimmers and non-dimmed switch legs do not automatically work with an LED fixture's driver. The chandelier itself is rarely the reason for change orders; the ceiling and controls are.
Bottom line for chandelier replacement: you are not buying a light bulb. You may be buying a driver relocation project, a dimming system update, or a mounting reinforcement. The fixture photo should be the last thing you approve, not the first.
Scenario 3: You're cutting LED strip lights
“Where can you cut LED strip lights?” At the printed cut zone. Only at the printed cut zone.
This sounds too easy, but it caused two rejected retail installation batches in Q1 2024. The team trimmed strips to match measured lengths and ignored the markings. Cutting through the middle of a series group kills that section, or the strip simply doesn't light after the cut point. The redo cost was $890 plus a three-day delay.
A 12V strip often has a cut mark every 3 LEDs. A 24V strip often has cut marks every 6 LEDs. Some constant-current strips can be cut every single LED. You cannot assume from a product photo. The manufacturer's printed cut line tells you where the series groups end. Copper pads in other places may be for soldering or power injection, not for cutting.
If your measurement doesn't match the cut spacing, change the layout. I know it's tempting to shave off 5 mm to fit the last bracket. That 5 mm can cut through a circuit and ruin the whole run.
After cutting, protect the exposed end with the proper clip, heat shrink, or an end cap. And while you're planning, check the manufacturer's maximum run length. A strip can be cuttable every inch and still have voltage drop issues at the far end if you run one extremely long line in a cove. The cut question is only part of the installation question.
Which scenario is this, really?
Here is how the decision tree works for me now:
- If a fixture is already installed, but the light pool is too wide, too narrow, or shows hot spots and artifacts, start with photometry and optics. That's scenario 1.
- If you are removing an old chandelier and hanging a new one, start with the ceiling, the driver, and the controls. That's scenario 2.
- If you are running a flexible LED strip inside a cove, display case, or sign, start with the cut marks and the datasheet. That's scenario 3.
Once you identify the scenario, the practical solution usually becomes obvious. The expensive mistake is to combine all of them into one generic lighting answer.
If you start with the wrong scenario, every “solution” makes the project look worse. That's the expensive lesson I documented 23 times.
My checklist now starts with one question: Are we controlling a beam, replacing a fixture, or modifying a linear run? Answer that first, and the rest has a much better chance of going right.