Another important issue is proper determining of binding energy, which may be tricky for non-conducting samples.

See for example:

http://srdata.nist.gov/xps/DataDefinition.aspx

point 7b.

Anyway, you should start with a thoroughful literature search for the systems related to those you are interested in, probed by XPS.

All the above is very good information.

Also remember that you can get X-ray induced damage, even with monochromatic sources so for some systems (e.g. Cl, F , Cu2+, Au3+ etc) containing samples then multiple measurements should be made to look for changes in chemical state.

With respect to curve fitting only fit the minimum number of peaks that you can realistically justify in a chemical way (e.g. perhaps oxide, hydroxide and water for a defined O(1s) envelope etc).

The reference energy is important too. Most users take around 284.7 (+/- 0.2 eV) for catalysts, 285.0 eV for polymers/organics and the fermi level for conductors. Any literature source should include their reference points and you can adjust your reference point accordingly. Remember too that for a lot of systems you will not know how well their energy scale is calibrated (it should follow the ISO guidelines, but many users don't do these routine checks) so there can be some variance

Assuming that good care was taking for referencing your binding energy scale, the analysis of the high resolution scans is crucial to extract those information. A Shirley background needs to be subtracted before peak fitting. There are many software for for the analysis, I use Igor Pro. My first guide in the interpretation is the XPS handbook (Moulder, J. F.; Stickle, W. F.; Sobol, P. E.; Bomben, K. D. Handbook of X-ray Photoelectron Spectroscopy. Eden Prairie, Minnesota 55344, USA: Perkin-Elmer Corporation, Physical Electronics Division; 1995. ). Together with digging into the literature would help achieve a reliable peak assignment.