Data's statement is "We may be able to match hostile's nine point eight, sir. But at extreme risk. " As for warp 9.6 I did not found a quote that validates the 12 hours statement made on Mem Alpha, but I found this quote from Best of Both worlds (season 3-4): "Commander, if the Borg stay at warp nine point six, we'll be forced to discontinue pursuit and power down in less than three hours." . Now this was after combat, so the number of hours in 'optimal' condition might be higher, but probably not by orders of magnitude. In Q Who (season 2), they reach at least warp 9.65 (and probably a bit, but not much, more than that, and I doubt that speed could have been sustained for a very long time). So it seems that 9.2 is safe (and probably can be maintained almost indefinitely), 9.6 can be maintained for some hours, slightly higher is still possible for short bursts and 9.8 is only doable in an all-or-nothing gamble. These statements are for different years so the somewhat coherent picture they seem to paint may be slightly misleading.
All in all, if Voyager's sustainable cruise velocity really is warp 9.975, it sounds like it is very much faster than that of the Enterprise-D. Yet it still would need 75 years to reach the Alpha Quadrant. So that would probably mean (to try to keep it a bit consistent) that Data's maximum warp from Where no one has gone before wouldn't be sustainable top cruising speed.
I need to correct myself... 9000 times LS = 2.7 million LY's in 300 years.
Lt. Paris did establish on screen that Warp 9.9 = 21 473 times LS.
Using Paris statement as a baseline, Warp 9.2 would likely/have to be much slower than 9.9.
Also, Data's statement from 'Where no one has gone before' episode only mentions 'at maximum warp'... whichever factor that may be.
I wouldn't know which factor might equate to 9000 times LS in that case... except to possibly put it in the area of Warp 9.5 perhaps?
Warp 9.2 could therefore be in the area of say 2684 times LS, 9.5 = 5368 times LS, 9.7 or 9.8 = 10736 times LS and 9.9 being 21 473 times LS (sustainable on the USS Prometheus in 2374 which had 4 nacelles and multi-vector assault mode - some other ships might have received engine retrofits like Nebula and Galaxy class allowing them similar or close to those speeds, with ability to use 9.91 for example in a pinch).
So, yes, I would agree that 15 000 times LS might not be a sustainable cruise velocity for Enterprise-D at the time of 'Where no one has gone before'... as that velocity would probably be higher than 9.2 (if 9.9 = 21 473 times LS).
And Voyager, well, the easiest explanation is (as I said before) that the damage from being pulled to the DQ radically reduced its cruising and maximum warp velocities and ended up needing decades to get back as a result.
But, it's conceivable that 9.975 is its regular sustainable cruise velocity (in an undamaged state) which would result in about 9 412 Ly's per day (or 3 435 680 times LS).
Plus, Voyager was launched in 2371 (8 years after the Enterprise-D) and was said to be state of the art (it would at the very least be more advanced at the time than a Galaxy class), so its not exactly without precedent.