Ideally gibs, especially taper gibs, should be essentially unloaded when adjusting. Fairly obviously any load makes it harder to feel what is going on as the gib is adjusted. By its very nature the loads on the Z axis gib of the WM12 are asymmetric. The weight of the head inevitably imposes a tilting force.
Given the relatively small size of the WM12 it’s not totally impractical to lay the machine down on its back, with suitable supports, to remove most of the offset loading. Which should improve the feel of whats going on although the motor still imposes an offset load. Albeit the other way.
If the machine were mine and I had similar worries about poor gib adjustment possibly contributing to the machine being less stiff than it should be I’d take the nuclear option of removing the head completely at the tilt joint and disconnecting the elevating screw after laying the machine on its back. Having thoroughly cleaned the dovetail slideways and gib I’d re-assemble with copious lubrication before carefully adjusting the gibs whilst sliding things back and forth by hand. A somewhat tedious process. It may take 30 or 40 reciprocations before you are sure that oil has gotten everywhere it should be and the gibs are evenly adjusted to the minimum clearances commensurate with sensibly free movement.
Sliding by hand gives a much better feel as to incipient tightness or attempting to adjust on the tilt. If the elevation screw mountings aren’t properly aligned its pretty easy to feel the extra stiffness when the screw is fitted and take appropriate measures.
Starting from scratch ensures you don’t wind up running down the rabbit hole of trying to correct one error with another. With a used machine of unknown history it’s impossible to be certain that any previous adjustments haven’t introduced a corrective error rather than re-set things properly. It’s possible to go quite a long way with corrective errors generating better, but still not quite right, performance before the house of cards collapses.
Correct set-ups are generally stable over relatively long periods. Corrective errors tend to need tweaking on a regular basis. Accepting that a lightweight machine like the WM12 can ultimately never be as stable as their bigger brethren I’d expect carefully and properly made adjustments to be stable for years in a home workshop. If you feel the urge to fiddle every month or six it’s likely you have corrective error rather than correct adjustment.
Or maybe unreasonable expectations as to the inherent stability limitations of a small machine.
In many ways big machines are mass stabilised! Large lumps of cast iron tend to sit still unless serious efforts to move them are made.
Clive