...
the avoidance of "thickness effects". The optically thicker a sample is, the more the measured energy-dependent absorption will be affected by imperfections in sample homogeneity. At a total absorption of 5 absorption units, 1 photon in 150 will make it through the sample. If regions of the sample illuminated by the beam have varying absorption, because your sample was poorly mixed, or worse, there are areas where the element of interest is absent, the measured absorption signal will be dominated by the contributions from those regions. More information can be found in the following papers:
Stern, E. A. & Kim, K. (1981). Thickness effect on the extended-x-ray-absorption-fine-structure amplitude. Physical Review B23, 3781–3787.
Heald, S. M. (1988). Design of an EXAFS experiment. In: Koningsberger, D. C. & Prins, R. (eds) X-Ray Absorption: Principles, Applications, Techniques of EXAFS, SEXAFS and XANES. John Wiley & Sons, 119–162.
Goulon, J., Goulon-Ginet, C., Cortes, R. & Dubois, J. M. (1982). On experimental attenuation factors of the amplitude of the EXAFS oscillations in absorption, reflectivity and luminescence measurements. Journal de Physique 43, 539–548.
optimal signal to noise. See here for further explanation.
...