Been a while since I last used procedural separation as a primary tool, and that was as an area controller when the radar went down (literally - the head was blown off) for about three weeks or so.
IIRC, climb rates were perfectly acceptable, vertical separation having been first established.
Practically one would not usually use it over any kind of large climb (or to a lesser degree, descent) as it was known that climb rates/performance could change considerably over several thousand feet (or several hundred for a normally aspirated piston.) But as a way of maintaining separation for, say, initial climbout until the routes were separated, nothing wrong with it.
Normal caveats apply, of course. Monitoring needed, forget about it in turbulence, and if the frequency is busy enough that one of the crews might be delayed in getting the "Unable to maintain climb.." call out, forget it.
I usually considered it a handy stop-gap measure.