Does your Smartcraft system have any calibration function for GPH and G/U Totalizer?
On my 27'w/ tw/ I/B gas MPI's( non-return) I have 2 FloScan 9000's that use inline flow-meters ...and.. a B&G NetEngSys display (redundant systems that use different kind of sources). Along with other ECM data, the latter reads ECM (algorithm-derived-data-from-injector-pulse-frequency-and-time) GPH and resettable G/U Totalizer same as your Smartcraft, .... point being: both the FloScan and B&G are user-calibratable for accuracy.
The B&G actually proves a bit more accurate of the 2 during most all fill-ups . There is still usually a bit of differential in displayed NGPH between running engines and between the redundant displays (and more at low rpm) but at the pump the actual pumped topped up fill of each independent 100gal tank is generally within .2 to 2.0 gal of the other .
As you mentioned, the only way to really know for sure if your mains really are or aren't burning essentially identical GPH is to top both tanks and keep the genny off and re-top-off after a substantial trip. More than once would be better.