Azimuth: how critical is it?

Dover, I could take issue with some elements of your response to my post, Like for example that when you use a meter, you are measuring the sum of any difference in channel balance plus voltage differences due to crosstalk, but it is not important. When I asked how do you measure azimuth, that question was directed to the OP. I was interested to learn why he felt his cartridge was running with incorrect azimuth. I know how I measure azimuth, and I know that you know how you measure azimuth. I was addressing the OP.The point is, as your response also indicates, there are two different criteria for “correct azimuth”. One is the physical model that you describe, and the other is according to the electrical model, where you measure crosstalk in db or voltage . Those two methods can give very different results. For reasons that I and you and some others have already suggested.

Playing a mono LP with one channel out of phase will tell you more about channel balance, as it pertains to the electronics downstream from the cartridge, than about azimuth, would it not?  At least that is my first reaction; I am thinking about it some more.  How do you define "correct azimuth", with respect to the placement of the stylus in the groove of a stereo LP or with respect to crosstalk (and phase, if you want)?  What do you hear when you do it?  Seems like with a mono LP, you have a mono signal with uniform phase off the LP.  The exact same signal is then fed to two channels in a stereo system. Then when you reverse the phase of one channel, you would still have signal out of that speaker and also you would have signal out of the other speaker.  To some degree the two would acoustically cancel, but not nearly perfectly, because phase is affected by room reflections.  Seems like you would lose SPLs and the tonal balance would be messed up.  I'm just curious.

3 to 5 degrees is not trivial.  Take a look at MC's reference to Peter Lederman's tooth pick trick, and maybe try it. Perhaps you will enjoy your sound even more, by a tiny bit.

What makes you say your azimuth is "slightly off"? Is the headshell not parallel to the platter surface, when viewed from head on?
Simao, Shims could be used on one side of the top of the cartridge, as a wedge between cartridge body and headshell undersurface, to correct for azimuth. Perhaps that is the intent.

The thing about having azimuth adjustable via the tonearm is that it also affords the user the opportunity to get it wrong.  As MC suggests, very often adjusting the cartridge body at an angle to the LP surface, in order to optimize the electrical readouts of channel crosstalk and phase, will also sit your stylus tip at a subotimal angle to the groove walls as regards its physical contact patches.  This is "not good", because it leads to uneven stylus wear and distortions that may be more noticeable than suboptimal crosstalk or phase anomalies.  I do own 2-3 tonearms that permit easy azimuth adjustment, but in recent years, I have settled on the paramount importance of having the stylus sit "square" to the groove walls, over all else. 

But if your headshell is fixed at an angle other than parallel to the LP surface when viewed from the front, I would suggest doing something simple to fix that and then stop worrying.