I've never actually done any analysis on my observing habits - just looked at the final lightcurves and adjusted based on comments from the Pro's that I report to (ie observe with more density or not). So I took a detailed look at one of my latest plans and did some analysis.

At 180 second integrations, an average 3 image run from Pointing image start (end slew to this target) to end of final image (start slew to next target) is 12 minutes (8 minutes for 2 image runs). Slew Start to Slew end typically takes 30 seconds.

So total overhead per target is 3.5 minutes for a 3 image run and 2.5 minutes for a 2 image run (by interpolation it will be 1.5 minutes for a 1 image run)

So, for a 2 target plan, one target gets full attention until overlap, then is revisited every 12.5 minutes during overlap. If I include a 3rd target with 1 image run then this pushes the gap to 17 minutes and a 4th 1 image target pushes the maximum gap to 21.5 minutes (ie if all 4 targets are overlapping) Of course this does not take into account any errors (ie pointing and guiding) which can add up to 1.5 minutes to a target per observation.

Now I pre-plan what targets I choose using Astroplanner as it gives me a nice pictorial representation of what is up when and what the overlaps are like and generally I try to avoid long overlaps. For the long period targets, typically, once I have 3 images I will manually end the plan, comment out the target then start the plan up again.

Regardless of what plans I have uploaded, typically I will just have 2 targets in the plan, it's only in the last month where I have been trying to push things to the limit.