[Sigia-l] Determining Users' Mental Model to Drive Site Architecture?

[Livia Labate]

> Jodi Bollaert said about open card sorting:
: my client is wary of that approach since it may totally
: negate weeks of time they've spent deliberating over two
: potentially good architectures.

If the two architectures are really good as they say then there is nothing
to worry -the open card sort will reveal just how compatible they are with
the users needs.

: What they really want to do is decide which of the two
: architectures is best, and THEN do a close-ended card
: sorting exercise with the one version (keeping their
: high-level categories intact, and letting users decide
: where a stack of subcategories should go).

I'm very skeptic about top-down architecture decisions, specially on an
Intranet. I wonder if the reason is a) management thinks they know best b)
you're using some constraining software that imposes on your architecture.
The primary intranet users are the workers and there is a great chance the
top-down decisions like that do not reflect what these users need. But like
I said, if your client trully believes in its vailidy, why not put it up to
the test?

Regardless, the closed card sorting will also be useful, even if you can't
do the open card sorting before it - because even here the users might
identify problems with the fixed high-level categories.

: To narrow their choice to one architecture, I've suggested that
: they spend a bit of time interviewing users

They haven't done any of that to make their initial decisions?

: to learn more about their mental models -- do users tend to think
: about their intranet more in terms of products/divisions or tasks?

I don't know how your co-workers do their jobs (which is the reason to
survey them), but I don't think that this is an answer you can get by asking
anywhere other than that particular work environment.

: I'm hopeful that the results of the interviews will reveal which
: would be the more intuitive architecture. I suppose it may also suggest
: that a combination is best, however, the client has already communicated
: that the constraints of their CMS may make such a strategy very difficult.

Ah there it is, CMS, the reason for making upfront choices (damn boxed
tools). I can understand your dilema; even though you are genuinely
concerned with addressing the user's needs and how they deal with the
information, you have a tool that will only allow you to change so much.
Yuck.

: Questions I would ask in the user interviews include:
: - What information do you typically look for?
: - Where would you expect to find this information?

I would ask:

A) What tasks you generally work on?
B) What information is inolved in these tasks?
(your 1st Q but targeted/segmented by task)
C) Where would you expect to find this information?
D) How often do you develop these tasks and for how long?
(helpful in expressing users expectations re intranet flow)

: I'm thinking that their answers will have a product- or
: task-oriented slant, and that may be enough to choose one
: approach, and move on to more testing. Since I've not
: experienced this situation before, I'm feeling on shaky ground.

Given your constraints I'd do the same thing, good luck :)



- Livia Labate
_______________________________