if you measure the output of 1 speaker and the meter says it's 75 dBSPL, and you check a second speaker and the meter reads 73 dBSPL, then when you increase the gain for the second speaker to the same reading as the first one, then they both will truly output the same SPL. This is what a low cost meter can do well: repeatability.

this does not mean that the reading on the meter is "really" 75 dBSPL though, but it can vary from one meter to another, and also the readings can vary according to the frequency that's measured.
This is called manufacturing tolerances.

most of the time, the auto calibration on an AVR is more precise than what a low cost meter can show.
the low cost meter will be able to bring all speakers to the same level, but will not necessarily show a true/exact meter reading: it will not tell you whether your SPL is, for example, 74 or 75 dBSPL, and there are additional errors when one changes frequencies, as those meters don't have a flat frequency response, so a sound measuring 75 dBSPL at 1 kHz will not give the same reading at other frequencies, specially for big chunks at the ends of the spectrum.

if you were to check a signal - with a low cost meter - of a specific strength at 1 kHz and then at 32 Hz, your meter would show very different readings; with a precision meter, the difference between the two would be minimal, as the tolerances are much tighter.

in short, with a low cost meter, all speakers can be adjusted to the same SPL, but you don't know what the true SPL reading is;
with a precision meter, an added benefit is that the display will give you a more precise, truer SPL reading, and this, over a much larger chunk of frequencies.
See: http://www.blu-ray.com/community/gallery.php?member=Gelli
or: Axiom Gallery