Correction removes material; it does not add any
A polish does not fill a scratch. It levels the surface around the scratch by removing clearcoat until the floor of the defect is flush with everything next to it. Every correction therefore spends a consumable you cannot replace without a repaint. Aerospace clearcoat is built to roughly two to two and a half thousandths of an inch, and it is not laid down evenly across an airframe. Each aggressive correction takes a share of that, no record of it exists on any aircraft we have gauged, and the paint gives no warning before it burns through at an edge or a rivet line.
The first step is a gauge, not a machine
We measure film thickness across the airframe before anything is committed, and we take a test section rather than trusting an average. The reading decides the process. Where the defect sits inside the clearcoat, it can be removed. Where it has gone through into the basecoat, it cannot be polished out by anyone, at any price, and anyone telling you otherwise is about to make it worse. Edges, fairings and rivet lines are read separately, because that is where the film is thinnest and where a burn-through happens first.
What choosing wrong actually costs
A burn-through is not a polishing error you fix on the next pass. It is a paint repair, often a panel repaint, with a color match to the rest of the scheme attached to it and the aircraft out of service while it happens. The other way round, the cost of leaving defects in place is that the aircraft looks tired. Defects are ugly. Burn-throughs are work. That asymmetry is why the decision leans conservative whenever the readings are marginal, and why we would rather lose the correction than sell you one the paint cannot support.
Paint is the corrosion barrier, not the trim
Clearcoat carries the UV load and the gloss, and the system underneath it keeps water and acidic airport contamination off the structure. FAA AC 43-4B, the FAA's guidance on corrosion control, leaves inspection frequency and treatment with the operator and treats washing acidic airborne contaminants off the surface as part of that program. The same guidance names two controls for filiform corrosion, which works under paint where the film is compromised: keeping the aircraft below seventy percent relative humidity, and washing the acidic contamination off. Spending film for a cosmetic gain on an aircraft two years from a strip is spending the barrier to buy a look that goes into the stripping bay.
What changes the answer
Repaint schedule first. Manufacturers generally recommend a strip and repaint at five to six years, industry practice runs closer to seven, ten is the outside limit, and a coastal, salt-laden environment can pull it in to around four. Polishing history second, because previous work has already come out of the film and nobody logged it. Then the paint system: single-stage schemes, common on older piston and turboprop aircraft, have no clearcoat at all, so polishing takes pigment and the color transfers to the pad. Composite panels are their own case again, since they flex, the paint on them needs very specific film thicknesses, and finishes there can crack or etch within a year of application.
Buy the finish once
Correction that is not protected afterwards is correction you will buy again. Swirls in a clearcoat come from wash technique, whether that is dirty media, the wrong product or someone in a hurry on an open ramp, and the same technique puts them back within a season. Where the film supports correction, we follow it with a coating and hand over a wash regime the next crew can actually follow. Where the film does not support correction, that decontamination and protection discipline becomes the whole job, specified and quoted in its own right, and on a tired scheme it does more than most owners expect.