How I approach this is I ask myself if the software solution makes sense. as @jeffsu mentioned, I base it on how useful it is. An example of one of these guides I don’t bother with is Windows installations because it’s a 10 minute procedure, nor do I integrate into guides (on purpose by the way) to avoid OS “flame wars” so my guides are OS neutral. Sorry, but a Windows guide is NOT worth my time.
However, I have done it for things like the HP UPD (just updated this) and removed the 1012 Vista reference as I only put that in while I transitioned to the UPD due to the 1012 driver being 4 versions of Windows old, hard drive diagnostics, battery calibration and such just to name a few.
While I have no hesitation with software guides, especially if it can help with hardware in some way like scoring a workhorse color laser printer at the price of a razor and blades unit from Walmart (even with 100k+ pages, especially if the engine is well built) if you’re willing to put in the time to setup the machine with alternate drivers, or tools on identifying failing hard drives or even identifying if a laptop battery has usable life left even if it’s older. However, I probably get a lot more leeway then someone who is new. FYI: The 1012 guide highlights the problem with hardware specific software guides - things change quick, and it may bite you later. It only lasted 4 years before it was on death’s door and it was a complete waste of time.
If it’s too far into the grey, I wouldn’t spend too much time developing it until you have it cleared by admin. Make it usable (and knock out the major bugs), but leave room to improve that’s intentionally planted so you can do a lot with it once it’s clear.