One thing I always do that most procedures skip: measure parallelism at four points, not just two. Front-left, front-right, rear-left, rear-right. I use a magnetic base indicator on the bolster and sweep the slide face in a cross pattern. On Komatsu H1F presses the rear-right corner tends to drift 0.01-0.015mm after 50k cycles because of asymmetric gib loading.
Also +1 on the thermal warmup point. We log BDC position vs time on new setups — typically stabilizes after 15-20 min at production speed. I keep a chart taped inside the control cabinet door for each die so operators know what to expect. Saves a lot of "why are my parts changing" calls.
For the parallelism check itself: if you are using shim stock between slide and bolster, go with brass shims not steel. Steel shims can give false readings if there is any magnetism in the slide casting.