I have used heat to get a neck straight temporarily then install carbon fiber rods to lock it down. Has worked successfully several times. Of course it requires the removal of the fretboard. I've got a couple of extensive threads over at the MIMF detailing these jobs. If anyone is interested I can provide a link.

So, how common are twisted necks? I get that they happen, but I just had a Taylor 110CE in with a slight twist, and I'd actually never seen that before. I actually thought it was an optical illusion at first because the nut was physically shorter on the treble side, but with the nut off it was clear that there was a slight twist. Nothing like what you show in those pictures though Barry.

I was able to make the guitar playable by dressing and levelling the fretboard, but it will be something for the customer to keep an eye on.

In my experience they are just the normal result of aging and the wood moving. Twist is something that you just deal with like any other aspect of owning and caring for a guitar.

On average I see 4-8 every 2 weeks. Sometimes mild sometimes insane. Lots of times it's just on instruments that are cheaper or just not maintained well. Usually the owners are told at GC that it just needs a truss rod adjustment which doesn't fix the issue and then they come by my shop.

So would a proper fix on a minimally twisted neck be dealing with it through levelling and dressing the frets taking the twist into account? Then milling the board flat when it's time to re-fret? But if its really bad ultimately dealing with it as Barry suggested above?

I don't see how you can heat the neck and not replace the frets.

Twisted neck is pretty major and if the twist is slight I prefer to just plane the fingerboard, but if you heat it to remove the twist it's hard to control where the neck will end up, and to ensure the guitar is fully playable I think it's still prudent to remove the frets and true the fingerboard before calling it good. Not to mention who knows if the frets will pop back up when heated?

In the MIMF post it seems clear the problem may in fact be caused by compression rods... it's not hard to over tighten them and they will eventually distort the neck especially if the neck needed a lot of correction (heavy strings for example, or bass necks). Not a bad idea to just use a double rod since they're not all that expensive and is easier to make/install if you really have to.

In new construction, careful stock selection and use of a strongback when gluing up the fretboard to neck seems to avoid the issues that some builders report using waterbased glues. I know the boss briefly used epoxy glue for fretboard-to-neck glue-up a decade or more back, but the switch to strongbacks and hide or fish for the job yields a straight neck without the epoxy-related repair issues.