An area of more concern to me, but related to any build up is maldistribution of liquid in the falling film. Maldistribution is a factor that many old school engineers never adequately considered, so it is one of the first things I examine for problems in designs. However a mitigating factor in this design is the nature of the flow and the relatively high conduction rate of the copper should compensate for a fair amount of maldistribution. If some manufacturer wants to test this I've got a good location in mind as long as I get to keep the unit when I'm done. Wouldn't be hard to test off-level installation, some simulated hanging hair balls, or soap build up that create maldistribution.
Contact time, residence time or whatever one calls it is not a relevant criteria for heat transfer. By itself it is meaningless. In fact, this greater contact time the average person think improves heat transfer in fact IMPEDES it. Film heat transfer coefficients increase as velocity increases. So for a given surface area higher velocity = higher heat transfer coefficient = lower contact time/residence time.High flows would not allow the "hot" and "cold" waters to stay in contact long enough to get the temperature differentials the diagram shows.
One uses more surface area/smaller flow channels where possible to increase velocity. Surface area is typically limited by material/fabrication costs or geometry as well as pressure drop. Decreasing flow channel size or increasing length is limited by pressure drop if not by geometry or plugging concerns.
The square quad coil design is attractive because it keeps pressure drop in control in greater overall drain pipe lengths and with higher flowrates (as in several showers or other users drawing water at the same time.) The squarish coils are needed to provide good contact with the pipe, something round coils don't achieve--this I can attest to after seeing too many jerry-rigged tubing coils on pipes in plant settings.