Dears,
As a grad student in economics looking for jobs into economic consulting
here is my 15 minutes outline.
EXECUTIVE SUMMARY
Objective: manuals sold cheaper AND StataCorp make MORE money.
Claim: it is possible to do so with regards to certain groups. Standard
Intro-Micro Type 2 (I think it's type 2, too lazy to check) price
discrimination. Which is the same system as the system StataCorp already
uses for the software: student licenses (possibly poor countries licenses
and so on...).
I do not think it is possible to sell manuals cheaper to everybody and
StataCorp making more money unless you are willing to make assumptions about
Stata software adoptions by other users (an assumption that I see as too
risky and I am unwilling to make)
Hurdle: providing a method to estimate whether this will actually work
before doing it.
Low hanging fruit: it is possible to boost the demand for manuals regardless
of the price
IMPLEMENTATION
Proposition: objective achieved if number of extra manuals sold at lower
price * lower price > (old price -lower price) * number of manuals already
sold to the group
AND unit profit is positive at lower price
Problem: two of the above quantities unknown: number of extra manuals sold
at lower price, number of manuals already sold to the group.
Low cost solution to figure out number of manuals already sold to the group:
maximum of the following two numbers
A) number of manual sets shipped to addresses of people who bought Stata8
who belong to group (either student licenses, or addresses in poor
countries...)
B) give small discount right now to the selected groups (some of these
people already have bought the manuals, this is an extra complication, give
them something that we think they may really want, e.g. 2 free downloads of
past articles in StataJournal, or one lecture in StataCourse)
Estimation of number of extra manuals sold at lower price much more
difficult: Use change in number of manuals shipped to addresses of people
belonging to the group from Stata7 to Stata8 (price changed in that
occasion) to estimate elasticity of demand. Adjust for number of months
software version has been available, and number of Stata software sold to
group. Possibly adjust for other factors. Assume demand is sufficiently well
behaved so that this number applies also to big price cuts.
You now have your nice estimates.
LOW HANGING FRUIT
People not aware of how good manuals are. Free and unobtrusive advertisement
(advertisement that users may WANT to see): anytime that something is left
out of the help (which is basically always) put "this and that may be found
at manual [ ] page... At the bottom of the help file). This notice beside
being advertisement also helps the user asses who useful the manuals are.
I posted two days ago:
1. When I bought Stata, I said to my self, "manuals? I can do without; I'll
use the help"
2. When I wrote my email on this topic, I wrote "manuals? I would seriously
consider buying them if they were cheaper"
3. Today I write "it would be really expensive, but if I could not borrow
them, I would buy them".
What had changed? It had changed that I had been able to borrow the manuals
and gauge with my eyes how useful they are. Of course it is not possible for
StataCorp to let anybody borrow them (although it may become possible with
the new Microsoft self destructing files), but writing in the help what
users may find in the manuals helps in this direction.
Sincerely,
Renzo Comolli
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
