If you are responsible for a facility and its safety, it helps to be clear on a distinction that trips up a lot of operations. An arc flash study is a document. An arc flash program is what actually keeps your people safe. The two get treated as the same thing, which is how a facility ends up with a completed study in a binder and a safety manager who is still exposed.
The four parts, and what happens when one is missing
A current study
What it is. The hazard analysis itself, updated every five years and again whenever the electrical system changes meaningfully, such as new equipment, added feeds, or reconfigured protection.
The trade-off. Treating a study as a permanent purchase is the most common way a program drifts. Without a defined update cycle and a trigger for system changes, the analysis on the label stops matching the equipment behind it.
Accurate labels on the equipment
What it is. The study results communicated where the work actually happens, on every piece of equipment.
The trade-off. Relabeling is tedious, and it is the step that turns analysis into protection. A current study with no updated labels leaves your team guessing at the panel, which is the only place the guess matters.
PPE that matches what the study calls for
What it is. Protective equipment on hand at the level the analysis requires.
The trade-off. Correct labels with the wrong PPE are worse than no labels, because people follow the label and believe they are covered. This is the part that is easiest to verify and easiest to let slide as inventory turns over.
Trained people
What it is. Everyone who works on or near the equipment trained on the hazard and on the right work practices.
The trade-off. Training takes production time. Good PPE with no training means the gear is available and the safe work practices are not, and behavior at the equipment is what actually decides whether someone gets hurt.
Why all four have to be true at once
These parts depend on each other, which is the piece that is easy to underrate. Each one on its own produces the appearance of a program and none of the protection. The program only works when every part is in place, because what decides the outcome is behavior at the equipment, and behavior depends on all four.
A real program has a defined update cycle, a clear trigger for re-evaluating whenever the system is modified, and someone who owns keeping it current between studies. Without an owner, there is nobody whose job it is to notice when the program has gone stale.
So the useful question is not who can produce a study for the lowest price. It is whether the approach covers all four parts, keeps them current on a set schedule, and leaves one person accountable.