Ah, so you did a sort of normal-mode 50-50 ohm insertion loss measurement?
Yeah, that seems reasonable. I wouldn't expect a teeny one like that to be gangbusters below the low MHz, but still enough to take the edge off if that's what you need. Attenuation over 100MHz is largely a matter of wiring anyway, you'd need feedthrus and shielding to deal with that.
Reminds me, I should measure my 24V charger supply; it's a variant of this design:
https://www.seventransistorlabs.com/Images/100WPowerSupplies.jpgI put it in a metal box, with filtered IEC receptacle, in addition to the filtering that's already onboard (which does a good job, but not usually good enough to meet regs, based on the informal / precompliance setup I have here). I also put a line filter (just discrete parts) on the output terminals, so it should be pretty quiet now, in both directions.
Not a big priority, because it lives in my lab, and I've got a bit of an ecosystem with pluggable headers, IEC cords and so on, hanging off my LISN (off an isolation transformer, off a variac), which is an effective absorber anyway.
Tim