I think carbonate porosity is a lot less predictable than people think. I have been on rigs that drilled into cave (paleo Karst) and fractures that would drink every drop you pumped downhole and 'eat' the cuttings, too. The petrofabrics tend to be defined by biological debris and depositional environment as much as dolomitization (which is dependent to some degree on primary porosity). Lovely sucrosic dolomite is not as common as people think and the permeabilities tend to be anisotropic, even without fractures. One of the earlier NUMAR logs I saw (about '92) couldn't get a repeat section--tool centralization was on 120 degree centralizers, the hole was at 45 degrees, and the fossils in the wall were not reading out the same because the tool was riding a little different on each pass.
I think the really outstanding porosity of carbonates is by microfracturing, not endemic matrix porosity. If you can find it, you got it made; however, it is that predictability you mentioned, just tough.
I was leading a well we drilled into the Austin Chalk in the early 90s. Just when we were making the turn at 8000', the bit dropped and we lost all circulation. After fighting it for 8 days and 18,000 bo produced, we called it quits and decided we found what we were looking for with a near vertical well.
It pays to be lucky sometimes.