Thanks, Sanford. Sales data on shareware (outside the anecdotal) do seem extremely difficult to come by. Your numbers may also be anecdotal, but (a) it's the first time I've ever seen actual NUMBERS presented; and (b) the size of your sample set is significant. Thanks again.
A few questions:
(1) What experience do you have with sales upon upgrade? Is it safe to assume upgrades (for a well-selling product) follow the same "100-200/day in first week, 10-20/day thereafter until new release" sales model of the initial release? Or do upgrades follow a sales model somewhere between "initial release" and "subsequent releases" (eg., 50-70/day in first week, 25-35 thereafter)?
(2) The sales model you describe doesn't seem to fit the typical "product life-cycle" curve of classical marketing described by Levitt (basically a bell-curve skewed to the right). Would you agree? Is it the same curve, only scrunched up in time? In your next article, would you be able to sketch a curve of the life-cycle as you see it?
(3) Any opinions on the "shareware publishers' business model" at interpubbooks.com? Using it to successfully model sales seems dependent, again, on the touchstone of past industry sales data, of which we seem to have none.
(4) Some posters to shareware listservs (eg., alt.comp.shareware., ~authors, ~programmer) have suggested it'd be great for someone to set up an online survey to gather the sales data experience we're missing. Like S&Ps Inndustry Surveys or RMA's similar studies, results could be presented in aggregate so that competitive advantages (ie., individual company sales data) wouldn't be revealed. Would O'Reilly be willing to host such a thing? If not, could you suggest how such a survey might be worded/ constructed (in case I get the gumption)?
Thanks in advance,