GF,
Basic gameplan is to follow the KISS approach to staying alive ..
(a) climb out on time using typical/expected ROC to achieve a safe turn height
(b) continue an appropriate, but moderate, climb back to the holding pattern to achieve a height somewhat in excess of MSA/LSA/holding altitude ... with no accurate idea of height the aim is to make sure that the aircraft is high
(c) enter the hold, extend as necessary to find a glideslope
(d) reverse and fly the glideslope inbound to check for a spurious glideslope
(e) if the glideslope is correct, then the appropriate pitch and thrust (making a rough allowance for known wind) will keep the aircraft somewhere near the glideslope .. continue the approach
(f) if the glide slope doesn't fit, then turn outbound and descend slowly in the pattern .. repeat from (c)
Usual observation for a typical 3000ft/10nm ILS was that the crew ended up 1000-2000 ft too high entering the hold for the Classic (most of my work was on -200/-300).
If you don't have an ILS or some other slope profile guidance, then the problems become much greater (and perhaps intractable) unless there is a known low terrain region below cloud to which the aircraft can be flown for a cloud break .. I never considered this for an exercise as the aim was to emphasise the importance of the pitch/thrust mantra and build confidence ... the quickest way to destroy confidence is to put a student in an impossible/too difficult situation.
It is essential that the program includes a progressive skill building aspect so that the I/F skills are appropriate to the tasks requested.
I ran my endorsement sessions fairly high workload (but with an emphasis on a fun approach to learning).
In amongst the normal scheduled exercises I used short I/F skills exercises a number of times during each session to provide breaks from the standard program work.
After the usual gen I/F and keyhole target exercises, the most useful is shooting nil vis, hand flown, raw data ILS exercises (repositioning to start from, say, 500 ft AGL or so once the student is comfortable down to Cat 1 minima) and having the pilot progressively fly lower as his skill improves, usually culminating in hand flown, raw data blind landings toward the end of the sim program. It is an essential part of the briefing process to get across very firmly the idea that these things are skills building exercises and have no part in routine line operations.
Another technique which I found very useful in the endorsement scenario was to alternate exercises throughout the session so that the student alternatively had work and support (quasi rest). My guess is that this increases session productivity by, say, 20-30 percent as the fatigue thing can be managed far better than having each student fly a half session program. While this latter is very common in recurrent sim training, I think it is very counterproductive during endorsement training .. the student thinks and learns better if he is not exhausted.
Seemed to work OK for me ...