Testing / Prototyping a faceted navigation

19 Sep 2008 - 11:05am
5 years ago
8 replies
1014 reads
Todd Moy
2007

Hi all,

I'm working on a large content-heavy site, which will utilize guided /
faceted navigation to browse the content. Given that there are a lot
of different ways people could browse the system, I was wondering if
anyone had any tips on testing this interaction. The catch is that the
infrastructure isn't built yet.

Right now, my instinct is to prototype (html wireframes) the cases
that I expect people to use--with the knowledge that most of the site
won't be available. The test scenarios, then, would be a bit more
directed than usual. I also suspect that I wouldn't be able to do much
in the way of passionate task testing.

Has anyone had experience doing medium-fidelity testing like
this...and could share some tips or resources?

Thanks,
Todd

--
-----
Todd Moy
todd.moy /at/ gmail /dot/ com
Grand Central # : 919 926 9730

Comments

20 Sep 2008 - 7:45pm
Itamar Medeiros
2006

My experience has been: every time you raise the level of detail of a
prototype, testers tend to get caught up on the aesthetics and lose
focus of the interaction. But I guess it's all a matter of how you
conduct the test and how do you brief the users about the nature of
the prototype.

Anyway, there were several discussions on prototyping (tools,
methods, etc...) on the list already. You might want to check them
out:

http://www.ixda.org/search.php?tag=prototype

{ Itamar Medeiros } Information Designer
designing clear, understandable communication by
caring to structure, context, and presentation
of data and information

mobile ::: 86 13671503252
website ::: http://designative.info/
aim ::: itamarlmedeiros
skype ::: designative

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=33211

20 Sep 2008 - 9:53pm
Jarod Tang
2007

> Right now, my instinct is to prototype (html wireframes) the cases
> that I expect people to use--with the knowledge that most of the site
> won't be available.

It's definitely help by using scenario directed testing, and more
points to be listed
1. the scenario should be you research result, instead of you
expecting, which lays as the key for testing
2. invite the potential user instead of get use of your professional
tester, who are good at testing but maybe lack of using motivation
3. believe and try to interpret what you observed, instead of let user
believe you expected

Cheers,
-- Jarod
--
http://designforuse.blogspot.com/

20 Sep 2008 - 8:32pm
Todd Moy
2007

Thanks for the pointer, Itamar. But, I think my original question
didn't really focus on the issue. The question is more about how to
prototype something that has many different dynamically generated
navigation paths.

Let's assume I gave the participant the task of finding the hardcover
book "Owls of North America" by John P. Author. For simplicity's sake,
the only navigation entries available are the following facets:
Subject, Author, and Format. The participant could start on any of
these paths, and at any point filter down on any of the other facets.

Now, this situation leads me to think that I need to plan for at least
9 different possible interaction paths (3^3). This could really get
out of hand in a non-trivial example, of course. So, I'm interested in
seeing how this situation might be approached.

Thinking more broadly, I would think that testing something like tag
navigation might also incur the same challenge.

Cheers,
-Todd

On Sat, Sep 20, 2008 at 8:45 PM, Itamar Medeiros
<medeiros.itamar at gmail.com> wrote:
> My experience has been: every time you raise the level of detail of a
> prototype, testers tend to get caught up on the aesthetics and lose
> focus of the interaction. But I guess it's all a matter of how you
> conduct the test and how do you brief the users about the nature of
> the prototype.
>
> Anyway, there were several discussions on prototyping (tools,
> methods, etc...) on the list already. You might want to check them
> out:
>
> http://www.ixda.org/search.php?tag=prototype
>
> { Itamar Medeiros } Information Designer
> designing clear, understandable communication by
> caring to structure, context, and presentation
> of data and information
>
> mobile ::: 86 13671503252
> website ::: http://designative.info/
> aim ::: itamarlmedeiros
> skype ::: designative
>
>
> . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
> Posted from the new ixda.org
> http://www.ixda.org/discuss?post=33211
>
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> Unsubscribe ................ http://www.ixda.org/unsubscribe
> List Guidelines ............ http://www.ixda.org/guidelines
> List Help .................. http://www.ixda.org/help
>

--
-----
Todd Moy
todd.moy /at/ gmail /dot/ com
Grand Central # : 919 926 9730

22 Sep 2008 - 5:52am
Jared M. Spool
2003

On Sep 20, 2008, at 9:32 PM, Todd Moy wrote:

> Let's assume I gave the participant the task of finding the hardcover
> book "Owls of North America" by John P. Author. For simplicity's sake,
> the only navigation entries available are the following facets:
> Subject, Author, and Format. The participant could start on any of
> these paths, and at any point filter down on any of the other facets.
>
> Now, this situation leads me to think that I need to plan for at least
> 9 different possible interaction paths (3^3). This could really get
> out of hand in a non-trivial example, of course. So, I'm interested in
> seeing how this situation might be approached.
>
> Thinking more broadly, I would think that testing something like tag
> navigation might also incur the same challenge.

We've tested faceted navigation with paper prototypes by creating
"slices" of pages for each facet. Also, in the testing, we knew the
tasks (since we designed them), so were prepared for the likely facets
the participant would choose.

You could do the same with an interactive prototype, but it would take
a bit of coding effort. That's why we start with paper -- quick return
on minimal investment with minimized risk.

Jared

Jared M. Spool
User Interface Engineering
510 Turnpike St., Suite 102, North Andover, MA 01845
e: jspool at uie.com p: +1 978 327 5561
http://uie.com Blog: http://uie.com/brainsparks

22 Sep 2008 - 8:32am
Todd Warfel
2003

On Sep 22, 2008, at 6:52 AM, Jared Spool wrote:

> We've tested faceted navigation with paper prototypes by creating
> "slices" of pages for each facet. Also, in the testing, we knew the
> tasks (since we designed them), so were prepared for the likely
> facets the participant would choose.

Another little trick w/paper is to fold over the items you want to
hide. Something like this:
http://www.flickr.com/photos/zakiwarfel/2879197388/in/set-72157607427633516/

(BTW, I whipped that up in about the same amount of time it took me to
write this email. Seriously, I just did it for this email).

Cheers!

Todd Zaki Warfel
President, Design Researcher
Messagefirst | Designing Information. Beautifully.
----------------------------------
Contact Info
Voice: (215) 825-7423
Email: todd at messagefirst.com
AIM: twarfel at mac.com
Blog: http://toddwarfel.com
Twitter: zakiwarfel
----------------------------------
In theory, theory and practice are the same.
In practice, they are not.

22 Sep 2008 - 9:02am
Eric Scheid
2006

The Oz-IA conference just concluded, and one of the sessions was on exactly
this question =)

<http://www.oz-ia.org/2008/presentations/faceted-search.shtml>

<http://www.slideshare.net/epek/are-users-really-ready-for-faceted-search-pr
esentation>

we'll have a video podcast up sometime in the next few weeks.

e.

23 Sep 2008 - 11:00am
Chris Ratcliff
2008

If you have defined user scenarios, Axure is wireframe prototype tool
that enables you to develop and test user interaction.
http://www.axure.com/

Thanks,
Chris

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=33211

24 Sep 2008 - 12:20am
Fred Beecher
2006

On 9/19/08, Todd Moy <todd.moy at gmail.com> wrote:
>
>
> Right now, my instinct is to prototype (html wireframes) the cases
> that I expect people to use--with the knowledge that most of the site
> won't be available. The test scenarios, then, would be a bit more
> directed than usual. I also suspect that I wouldn't be able to do much
> in the way of passionate task testing.
>
> Has anyone had experience doing medium-fidelity testing like
> this...and could share some tips or resources?

Not to get into the whole "appropriate level of fidelity" argument again,
but in my experience it is *very* important to prototype these sorts of
interactions in the context in which they'll be used, i.e., a Web browser.
The reason this is important is that it's the best way to get truly accurate
information about how your user group(s) respond to that particular
interaction. To address the issue Itamar brings up about fidelity, the
interactive prototype can *look* like a crappy, half-thought-out wireframe,
but the interactions must be there. If you're really really early in the
design process and have easy access to users, then a quick paper prototype
test like what Jared described would be useful too.

So your instinct is a good one. : )

The other important thing you mentioned is prototyping "common scenarios,"
which is right on. Prototyping is all about faking stuff. Smoke and mirrors.
Present your users with a realistic task they'd need to use the system to
complete, prototype the "happy path(s)" (and if you have time one or two
unhappy paths), and use your facilitation skills during testing if people go
down the "wrong" path to find out why, etc.

If you're coding straight HTML, this type of interactive prototyping is way
more time intensive than paper, but there are prototyping tools available
that can help this go a lot quicker. My weapon of choice is Axure, and you
will find a ton about it in the list archives.

Take care,
F.

------------------------------------------------------------------------------------------
Fred Beecher
Sr. User Experience Consultant
Evantage Consulting
O: 612.230.3838 // M: 612.810.6745
IM: fbeecher at gmail.com (google/msn) // fredevc (aim/yahoo)
T: http://twitter.com/fred_beecher

Syndicate content Get the feed