[Sigia-l] Research and Search Results

[Chris Chandler]

"Derek R" wrote:

> To introduce 'categories' is to run away from the *product sale* in
> favor of window-shopping/browsing environments!
>
> Why not *sell* the user what they asked for?


How is a web site, any web site, going to sell me the "video camera" of my dreams when I didn't know basic information
like the difference between mini-dv or digital 8 formats?

Not all searching is "known item" searching.



> The user has arrived on your Web site, money-in-hand, and the IA
> walks-them-away from the cash register to correct their vocabulary!!


Uh huh, so now you're FOR making money and AGAINST correcting vocabulary...?

It seems to me the only time a shopper arrives at your store with money in hand ready to check out is when they know
EXACTLY what they are looking for, which presupposes research done elsewhere or an existing expertise. The principle
that every transaction/experience with an eccomerce web site could lead to a sale is popular with management, but highly
unrealistic.

Secondly, if you haven't noticed, the world is chock full of technical vocabulary --  if you don't happen to speak the
dialect of digital video, you can search for "digital video camera" till the cows come home and not find what you are
"looking for."


> >| browsing using categories is successful
> >| twice as often as using search
>
> Successful at what? Successful at USING categories! Agreeing on the
> category-name! Why go through all that mumbo-jumbo?


How could I not? That "mumbo jumbo" as you say is involved in pretty much every communicative act.


> I'll sell you whatever you want and you can *call it* anything you like.


This slogan has all the makings of one of those new IBM commercials -- somewhere between "fairy dust" and the looking
glass that shows the "future of business."

But really, I don't see that you've presented an alternative! You don't even have a black box or "magic happens here"
note to tell me how searching alone is going to solve the basic problems I've presented in mundane examples.

I'm all for search. I hope one day soon to have "intelligent agentz" (minimally, a consumer reports clipping service, a
market estimator/auction agent and an viable electronic wallet service) at my disposal to handle most of this mumbo
jumbo. I think google is awesome and fervently hope that someday, there will be no bad search implementations. I just
don't believe that it is the "correct" answer, and "categories/browsing" is the incorrect answer. Perhaps I've
oversimplified your position.


> >| Your conclusion that "if you want to make money,
> >| you'll do it with search and not categories"
> >| flies-in-the-face of these [UIE] results
>
> Jared's study was of *task* completion -- not sales or revenue
> generation. You are not even aware of the UIE results, are you?



Uh, I'm not the one discussing UIE's results in terms of sales, you are.



> Obviously *pre-defined* categories are going to 'complete tasks' if the
> users 'submit themselves' to using them. To do so they must *commit
> time* and over-bearing *effort* to browse *and maintain* sustained
> interest.
>
> Unfortunately, the user was already *ready-to-purchase* when they
> arrived onsite !!! Why do you think they are there?
>
> Category-introduction just *complicates* things beyond original concern
> (i.e. the product purchase). Do you understand yet?


Not really, but I appreciate your patience.

If my "original concern" is that I want to purchase a Sony DCR-TRV25, then what you say makes perfect sense. But, in the
case of "I want a digital video camera" it does not. I believe in the utility of both search and browse. The decision as
to which is more effective "depends" on the situation.




> Why browse categories -- which is abstraction/dreamy-window-shopping --
> when you can sell product (the real) ??
>
> The whole point of selling is to supply what is *asked* for.
>
> Even http://ebay.com/ is beginning to realize the objective of business
> (online or offline) is to *make sales* ($$$). That means 'fast and
> simple' (i.e. a 'search' focus) is made available to users:
>
> eBay Test Home Page -->
> http://derekrogerson.com/images/ebay.jpg
> http://pages.ebay.com/community/news/hptest2info.html
>
> Why ignore 'what the user wants to buy' and try to convince them of
> something else?
>
> Screw categories! Make the sale, damnit!


Excellent choice of example -- very few companies have spent as much time as ebay thinking about how to complete the
maximum number of sales. Needless to say, I don't believe it bolsters your case as much as you seem to think.

On their current homepage they have category listings and a browse button above the fold on the home page. Under browse,
they've recently added a section called "themes" (another known alias of the dastardly "category") to show merchandise
related to "sports" or "entertainment."

On the test interface, they've moved closer to the google model -- but I notice that "see all categories" and "browse"
are still there.

Search and browse can and do work in tandem you know. If I search for "digital video cameras" I get 10 pages of results
(468) and the "featured item" is a digital microscope. On the other hand, using the "related categories" navigation area
on the left I can quickly navigate to a list of 106 items, all directly relevant to my original desire.



> >| as usual, your inflammatory rhetoric -- pro-
> >| eugenic? fascist?? who do you think you'll
> >| convince with that kind of nonsense?
>
> The UIE article *explicitly* labels humans as 'Garbage' -->
> http://world.std.com/~uieweb/searchar.htm
>
> As usual, I am pointing to 'background assumptions' which are just
> sitting there, without attention. These are clear *observations,* not
> personal accusations.


I'm glad to hear you don't view them as such, but I have to wonder if you've fully considered how the rhetorical force
and typical uses of those words might lead others to read differently.


> 'Garbage-in/garbage-out' is a garbage user-philosophy, and I have
> indicated as much.

Reading the abstract, I see what you are referring to, but think it is possible you've drawn the wrong conclusion. I
read that passage, and indeed the entire abstract as an indictment of typical web search engine implementations -- they
are the real garbage -- not an indictment of searchers.

This conclusion resonates with some of my recent practice. The consumer goods site I mentioned previously had an
incredibly atrocious site search -- it was actually incapable (poorly installed/configured/customized, whatever) of
retrieving results from the database of products!!! In one meeting, I told the client that if the search performance
could not be improved in time for the relaunch of the site, they would be better off without search. The marketing
director over ruled me, because she had read Nielsen's piece and believed it a "usability rule" to have search on every
page, especially the home page.

On the other hand, some students in my summer class performed discount usability studies on sixteen OPACs (online
library access systems) and their most important conclusion was that libraries would do well to implement a single,
google-esque search box on their start pages and do whatever they could to make the complexity of the underlying data
sources invisible to the user.

It depends.

-cc