I mean, we could revert to MegaTraveller or TNE and account for mass, volume and surface area. That is the realistic, crunchy way. There is NOTHING wrong with that approach for people that like that stuff. Especially if it's important to the campaign assumptions (i.e. no magic reactionless drives or artificial gravity to smooth the edges)
But as anyone that has done much work with those sequences knows, it's a lot of accounting for not very much difference. Mass vs thrust determines the acceleration in a vacuum. Components are all sorts of mass per volume, but it turns out you have some that are quite dense (drives, full cargo holds) and some that are not (staterooms, control areas, access areas, empty cargo holds). And the 10-60% of the ship that's liquid hydrogen storage turns out to be a fair average. In the end you have a kilogram value and a cubic metre value that often agrees with a density of liquid hydrogen.
CT and Mongoose ship design is overtly a simplified process to let you put components together and end up with a vessel with usable characteristics. Of course a ship has mass; it's put together using volume, but it's also sufficient to say that a drive of sufficient size can propel the average mass for a given ship volume, at a particular rate.
And deckplans? You can't usually work out exact volume or mass from a diagram of where the crew can walk around. Most of that is just air, boxed in by sturdy but thin metal. Except for the simplest hulls (and especially for streamlined ones) there's going to be varying heights between deck and ceiling, curved space that might be useful to store fuel or machinery in, or maybe not. The old rule of thumb is still used - 2 squares per dTon, within plus or minus 10%... but maybe don't bother showing the fuel tanks or volume that's unusable due to streamlining.
As I said earlier, mass is the most important parameter for a spacecraft (and it's probably more important than volume for aircraft. Volume is only more important for watercraft). One approach anyone can take is to remove the "displacement" from dTons and think of the ships just in terms of mass most of the time. When volume is needed, such as for deckplans, use two 1.5m squares, or approximately 14m3, per metric ton (with a +/-10% variance). Which I suspect was precisely what was intended when the idea was first published around 1978 (Snapshot?)