Phase a commercial lighting upgrade by area based on what happens in each area, not by fixture count. Start with one area that has a single clear function, a project reason to go first, and a boundary you can evaluate without depending on unfinished work next door.
First, map every candidate area by its function, its priority for your project, and anything it shares with its neighbors, such as controls, circuits, or light spilling across the edge. This is a planning rule of thumb, not a fixed ranking. The right first phase depends on your building and your goals.
How to Map Candidate Zones and Choose a First Phase
A good first phase is an area with one function, a stated reason to go first, and a boundary you can compare cleanly. For each candidate area, note three things:
- What happens there. Name the task, who uses the space, and when it runs. A packing line on two shifts and a storage room used a few hours a week call for different questions.
- Why it goes now. Write down your own reason for the timing, such as a planned shutdown window, a task area the team wants fixed first, or a budget cycle. No standard order of area types exists, so the reason has to come from your project.
- Whether it stands alone. Ask whether you could judge this area's result while the areas around it stay unchanged.
Recording what each area is used for has a practical purpose. A DOE commercial and industrial lighting evaluation protocol includes area description and usage group in the protocol's example area-use field set for evaluating lighting projects. That gives you a ready way to label each area. It doesn't tell you which area should come first.
Equal fixture counts or a neat split down the middle of a floor plan don't make a useful phase on their own. Here's an illustrative case. A facility has a packing area and a break room next to it, each with 20 fixtures. The packing area has a scheduled shutdown week and its own circuits. The break room shares a switch with the corridor. The packing area makes the better first phase because it has a reason to go now and a clean edge, even though the fixture counts match.
If one facility mixes very different tasks, our guide to commercial lighting zone planning includes a zone worksheet for sorting areas before you set the phase order.
Check Whether a Proposed Zone Can Be Evaluated on Its Own
A zone can be evaluated on its own when the area, loads, controls, and operating conditions in your comparison all line up with its boundary. Before you lock in a phase, check each of these:
- Circuits and loads. Do any circuits feed fixtures on both sides of the line, or also power non-lighting equipment? In a PNNL working-office lighting evaluation, circuits that crossed study zones or included non-lighting equipment could not be used to evaluate a zone directly. The researchers used fixture inventory and related estimates for those loads instead.
- Controls. Note any sensors, switches, or schedules that cover more than one area. If the phase changes controls as well as fixtures, DOE's lighting controls protocol ties the choice of method to site-specific evaluation methods and available data, such as whole-premise interval meter data and, ideally, sub-metered lighting data. Our overview of commercial lighting controls can help you sort out which control strategy belongs in which zone.
- Operating patterns. Check whether hours, shifts, or occupancy will be similar before and after the work. A schedule change in the middle of the comparison changes what the result can tell you.
- Light at the edge. Write down any light that crosses the boundary from neighboring fixtures. Treat it as something you observed on your site, not as an automatic failure. In a Philadelphia airport-apron trial, adjacent-pole luminaires contributed light to the measurement area, so the trial included those luminaires to evaluate the full LED system. That was an outdoor apron. It doesn't set a rule or threshold for indoor rooms.
None of these findings ends the plan. They feed the boundary decision in the next section.
Decide Whether to Separate, Combine, or Track an Interaction
Whether you separate, combine, or track comes down to one test: can the planned comparison still answer this phase's question at this boundary? Use the table to turn what you found into an action.
| Observed condition | Boundary action | What the result can claim |
|---|---|---|
| Question answerable at the boundary | Keep the zone separate | A result for that zone |
| Comparison can't tell the areas apart | Combine into one phase | A result for the combined area only |
| Interaction recorded but not isolated | Track it and note it | A qualified result that names the interaction |
The middle row matters most when your only data covers more than the zone. DOE FEMP's federal M&V guidance points out the limit of whole-building verification: it checks the total performance of all measures, including how they interact, but it cannot verify individual measures. A system-level approach that measures the relevant parameters is the separate option in that framework. If a shared meter is all you have, combining the areas keeps your claim honest.
These branches are planning guidance, not an engineering or measurement standard. Light crossing a boundary is not, by itself, a reason to combine. Use the third row when you can describe the effect but can't remove it.
Compare the Completed Phase and Plan the Next One
Set the question for each phase before work starts, record matching before-and-after inputs, then use the answer to continue, adjust, or regroup. An illustrative question might be: "Did the packing-area phase lower connected lighting watts while the shift schedule stayed the same?" That question tells you what to record and what to hold steady.
The DOE protocol's example before-and-after lighting evaluation fields give you a starting set. Keep the ones that fit your question:
- Area description and usage group
- Fixture count and watts per fixture, before and after
- Existing control and new control device
- Baseline and proposed annual operating hours
Keep observed conditions apart from planned ones. The baseline for a finished zone is what you recorded before the work. A future zone has only a proposal until its own baseline is collected. For the next areas, gathering lighting layout measurements up front gives you both a baseline and the dimensions you'll need for our online lighting layout tool.
Don't report a building-wide total as the result of one phase. In DOE FEMP's federal project framework, that is why whole-building totals cannot verify an individual measure.
Then decide the next step. Continue with the same approach when the phase answered its question cleanly. Adjust the comparison or boundary when an input shifted during the work, such as hours or a shared control. Regroup the remaining areas when this phase showed that neighboring zones can't be told apart, and plan them as one combined phase.
FAQs
Can one phase include several rooms?
Yes. A phase can cover several rooms if they share a function and you compare them as one zone. Report the result for the whole group, not for each room, unless each room has its own data and its own comparison.
What can we conclude if a zone has no recorded baseline?
You can describe the zone's current fixtures, controls, and hours after the upgrade. You can't claim a before-and-after change for that zone. Record baselines for the remaining zones before their work starts so later phases don't have the same gap.
Do dimming controls in a phase need an extra check?
They may. DOE's lighting controls evaluation protocol notes that ASHRAE recommends measuring light levels for control measures, especially dimming, to make sure work areas still get enough light. Plan that measurement into the phase if your question covers lighting conditions as well as energy use.
Who decides circuit changes needed to create a clean zone boundary?
A licensed electrician should handle any rewiring or circuit changes. Code questions belong to your local authority having jurisdiction. A phasing plan can show where a boundary would help the evaluation, but it doesn't replace that design or approval.
