This is a progress report on an ongoing study of pricing information goods
and services on the Internet. We consider the possibility of setting an opt
imum price for an information resource, based on the observed willingness t
o buy at two or more test prices. We show that for simple assumptions about
the dependence of willingness on price, it is easy to determine the optimu
m price with rather high confidence, in a relatively short period of time.
We also show that, in the university library setting, the arrival of custom
ers is not predictable enough to permit us to use variations in the arrival
rate as a surrogate for willingness to pay. We speculate on the implicatio
ns of this result for managers of university research libraries, and for ot
hers doing business in the online environment.