I tried the FoCal Software and wasn't very impressed with it. that focustune sounds interesting, but I just use something like what BCsteve has with a ruler, it's worked out much better, much faster, and the results are repeatable, unlike what I was getting from FoCal.
the camera will store a number of different combinations whit fine tune settings and auto-detect them (so like a lens, and lens + teleconverter can both be stored separately since it is different equipment, but otherwise its 1 AF setting per lens).
however, with a zoom lens you may have different calibration requirements at different zoom levels, so its best to do tests over a variety of zoom ranges, then choose the setting that is the best average correction value since only 1 setting can be input for the lens....so for example, if you're finding that at 24mm you need -6 and at 85mm you're needing -1....you can only put in one value for the lens, so you'd average them out and put in something like -3....or if you know you shoot at a specific zoom range the majority of the time, then you can target the settings for that range, like if you're getting the values in the example, and you know you're almost always shooting at 24-35mm, then you can set it to something like -4 or -5 to make it more biased towards that range.
if that makes sense...
also make sure you're placing your test target at a appropriate distance for the focal length you're testing, you don't want to be too close to the minimum focus distance, nor too far that you're at infinity either...I think most recommend 25-50x the focal length you're shooting at. though that number can vary, and you want to take into account how far you are from the subjects you normally are photographing too...
EDIT- also make sure when you're doing the test shots to use plenty of light so you can keep the shutter speed up, ISO low, and shoot with the lens wide open....also use mirror up mode and either a remote release or timer or something, with the tripod on a very stable surface (not on padded carpet or something like that), that way you eliminate as may variables as you can....and once you've input your calibrations, re-test it at the settings you've input to make sure you're getting repeatable and consistent results.
EDIT again - if you end up using something with a ruler like BCsteve or I do, make sure the ruler is metal and is straight, and is finely graduated (not just every 1/4 inch or something, you want good fine gradations so you can really examine and pick out the best settings, probably at least 1/32 inch, or full mm gradations on a metric scale) since the difference between 1 point and the next in the AF fine tune is very small you'll need to be able to SEE the difference between them.