The magnets work, too. The level of execution in body repairs has increased a good bit over the past 40 or so years, but there are usually some contours of the factory sheet metal which many "bondo artists" don't always duplicate well, so some visible cues would be there when more scrutiny might be desired. Radius on sheet metal curves and such, down low. UNLESS they section in a piece of stamped metal meant for "patch panel" purposes. In which case, little evidence might be there, by magnet test. In those lower panel areas, "tapping" with a soft/hard blunt object might be necessary to detect the additional panel rigidity from the flanged sheet metal of the patch panel rather than what the OEM metal thickness/rigidity is. But to me, if it takes that much scrutiny, unless it's really something I would want, I'll pass.
Reason is that you don't know what sort of metal prep (corrosion-related) was done or how good it was as the panel was repaired. Always the possibility that it might start to "bubble up" in later years of ownership, which degrades the potential value the vehicle might have once had. Yikes!
So . . . if my tap test might indicate something has been done, there have usually been OTHER areas that I discovered on the vehicle that were not what I wanted to see. As they say, "One thing can lead to another . . ." To me, just part of "due diligence" past CARFAX. But then I was doing this 50 years ago, well before CARFAX was around. Collision damage is one thing, which has its own visible cues, just as hail damage does too. Knowing how some body shops (and techs) operate helps, sometimes. BUT when you can tell they've done a good and comprehensive job in what they did, more confidence in the repairs can result.
CBODY67