> The problem here is that as a general principle: You're not smarter
> than Amazon. (I don't mean you personally, Kyle.) Amazon is not a
> small bookstore, but far larger than any library, so by your reasoning
> it should have even fewer covers than a library. And Amazon and other
> ecommerce sites do *exhaustive* testing. They do A/B tests all the
> time. Make a cover smaller—do sales go up or down? Make it larger—up
> or down? Round and round.
I do not doubt that covers increase sales, and for bestsellers where
recognition plays a greater role, there's an argument that they
perform a useful function as well. However, in research libraries
where people are accessing old literature, for soil surveys, committee
reports, technical texts, and the like, the utility factor for covers
drops.
> Amazon knows what works. And what works selling a book ought to have
> some relationship to what works browsing it.
I wouldn't be surprised to hear that showing covers also causes people
to borrow more books. However, I think this has more to do with them
just enjoying the experience of using the catalog more the same way
that drawing cool graphics on a can causes more people to drink the
soda pop inside.
Whenever a library has a renovation project, use goes up
significantly. From a functional point of view, nothing has changed.
But people are more likely to ask for services if they think your
building is nice. Same rules apply in the virtual world. This is not
the same as improving service.
If you're wondering why I would accept that book covers increase use
but I still think they are a waste of time, it is because I think the
demand for them is implicitly based on the questionable assumption
that greater demand for services is good.
The goal must be to get people what they need, and this is not the
same as maximizing use. If we can assess the quality of our services
by the demand for them, we can reasonably infer that McDonald's serves
the best food on earth.
Not that we can't learn plenty from Amazon. Many of their methods seem
very appropriate for popular collections consisting mostly of high use
materials. Even research and special collections can learn from them
-- for example, the search inside function is a powerful discovery
tool. Even if all Amazon cares about is increasing sales, that
particular utility makes it easier to determine if you're looking at
the book you actually need.
kyle
Received on Fri Jun 27 2008 - 20:05:31 EDT