It's hard for us to answer questions that are so vague.  

Generally, it is recommended to collect XANES spectra out to at least 100 eV (and better 300 eV) above the edge in order to get a decent normalization.  Well, for "normal" hard X-ray spectra: for edges below about 3000 eV, you can often get by with less energy range.

If you don't have that, it is sometimes OK to pick an energy as far above the edge as possible and pick an edge step so that all the spectra is 1.0 at that energy.  That is definitely prone to systematic biases in normalization.   Furthermore, if your data is particularly noisy, it can add a lot of variation in the normalization to use just a single energy.   

Which just brings me back to the original point: it's hard to say in general how to best handle XANES with a short energy range.