Aldo Sassi, the well-known cycling coach, dedicated many pages of his book "Dalla parte del ciclismo" to redefine the concept of "anaerobic threshold", eventually concluding that it does not exist, at least in its classic definition. 

Tim Noakes, doctor and physiologist expert in evaluating runners and triathletes, maintains in his book "Lore of running" that lactic acid production is not related to an anaerobic condition in the muscles, but it is simply a result of the metabolism of carbohydrates: if the intensity of the effort increases, more carbohydrates are utilized and lactate concentration rises too, indipendently from the oxygen availability in muscles. (If this is the case, why if Hb increases, lactate values decrease for each intensity?...) 

It's not my intention now to "sink myself into mud" of theoric-scientific evolutions, but simply to confirm the utility and importance of measuring the so-called "anaerobic threshold", or, if you prefer, to determine the curve that relates intensity - lactate - heart rate.  

I have been testing for many years now this particular relationship on cyclists of all levels and I can confirm that, in my opinion, this measurement is the most closely related with the performance of the cyclist on the road. 

It is an essential practice in order to adapt the training intensities to the progresses of the athlete in his preparation, but also to evaluate the level of his performance potential reached by the cyclist, as the "anaerobic threshold" is strictly related to racing results, especially for climbing and time-trialing, but not only for this. 

Portable lactate-measuring machines have been available for purchase for some years now, which from a single drop of blood and in a few seconds allow accurate measurements and easily repeatable, anywhere and anytime. 
It is therefore much easier nowadays to evaluate the relationship effort-lactate, also thanks to the utility of power meter devices. 

Personally I am used to evaluate the "anaerobic threshold" on a climb, repeating a certain uphill distance several times at increasing efforts, with a difference in height of 80-120m, checking at the end of each effort the heart rate, the lactate, the tim e(therefore the VAM), the average watts (if a power meter is available). 
A good professional rider of 70kg of body weight starts with about 300 watts in the first effort, increasing by 25 watts each time, doing the last stretch at around 450-500 watts, so to reach a sufficiently high lactate concentration (usually higher than 6 mM/l). 
For each effort, the athlete has to try and keep an intensity as much constant as possible (it is easier to do with the help of a power meter device). 

The data we obtain allows us to draw a power (expressed in watts or VAM)-lactate curve, which usually is esponential, and a power-heart rate relation, which usually is linear. 
Conventionally I identify "anaerobic threshold" at a value of lactic acid of about 4 mM/l (variable from athlete to athlete though), corresponding to a certain power output (always expressed in watts and/or VAM) and a certain heart rate. 

Repeating this test every 2-4 weeks makes it possible to adjust the training intensities and to single out possible improvements: if the cyclist at 4 mM/l of lactate is capable of developing more watts or a higher VAM, then his "anaerobic threshold" has improved.