Audience Segmentation Navigation

29 Jun 2010 - 1:19pm
4 years ago
8 replies
1667 reads
eriklevitch
2008

I know it is a common practice for B2B sites to structure the navigation by audience; however, I think structuring the navigation based on tasks rather than audience type could be a better solution. Would analyzing the tasks for each audience type naturally present a common IA that all audience types could understand? For example, each audience type needs to know the benefits of a product (of course, the benefits for the product differs for each audience type). Would that task result in one section for benefits or, say, 4 navigation options for each audience type where they can learn about the unique benefits?

Any insight into this would be helpful!

Comments

29 Jun 2010 - 4:34pm
thehilker
2010

i attacked this problem by building the navigation around the types of information i had about the products: white papers, case studies, demo videos, etc. each of our personas were looking for very different types of information (technical specs vs. high-level capabilities). the idea was that in product discovery, people are looking for information to inform them as to which product is best for them. each sub-section would display a list of products that fit that criteria. is that clear? it's really just faceted navigation.

ideally, it would also help SEO ("[product name] white papers" or "[product category] demo").

i hope this helps!

29 Jun 2010 - 8:08pm
Gino Rodrigues
2008

The audience scheme has more chances to succeed. Specially if the segments are well defined, as often happens in B2B realm, you have the opportunity to delimit more uniform areas, where you can focus offers and communication. Not to mention SEO techniques to leverage segment-specific landing pages.

A task oriented navigation resembles applications, which may not be the best structure for your content if you need to point exceptions between segments. Of course, SEO can be compromised as well.

Besides, if the tasks are evidently powerful, you can make them cross a segment scheme using a wizard, which can (again, CAN) be a cool interaction opportunity. It recalls me of Fidelity´s "myPlan" (http://personal.fidelity.com/planning/retirement/content/myPlan/index.shtml).

Good luck!

Gino

30 Jun 2010 - 3:09am
Ritika
2010

I agree with Gino...

Normal 0 false false false EN-US X-NONE X-NONE

As different audiences have different needs ,the navigation based on audience provides more insight to the products applicable for the audiences.

This avoids unnecessary navigation to the products which are not applicable to them and they get what they require.

But the navigation by audience can be improved by providing what that audience means to the company. Eg ‘small business’ if they specify the no. of employees they consider for small business then it becomes easier for the customer to select their segmentation.

 

30 Jun 2010 - 8:14pm
Jared M. Spool
2003

Erik,

I'd be anxious about trying to design by some sort of segmentation. In my experience, it's unlikely you can come up with any definitive design that will improve the experience, because you're trying to qualify the user on attributes that are probably unrelated to the task at hand.

Instead, I'd focus on behaviors. (Behaviors would be the scenarios that are behind the tasks you mentioned.)

Of course, once you start researching behaviors and identifying the patterns from that research, you've now entered the world of personas and scenarios. If you design to specific personas and scenarios, I think you'll get the experience improvements you're looking for.

Jared

1 Jul 2010 - 12:06am
Jared M. Spool
2003

"Instead, I'd focus on behaviors. (Behaviors would be the scenarios that are behind the tasks you mentioned.) "

I second the motion.

I'd also say do some tree testing to try on some of your structures and labeling before getting to layout.  During the time I spend focusing on language, labels, etc... I almost always end up considering several approaches, and the only way to really know what is most effective, beyond the obvious, is to experiment.

After all is said and done, and things are up and running, I'd get started with some MVT and CRO with the live stuff to squeeze out a bit more effectiveness.  (Is it silly that I still recommend this ALL the time?  I feel like there's a testing divide, so I never know whether I should expect people are going to call it old news, or call it sci fi.)


--
Mike Caskey
p. 303-500-3245
c. 303-748-7815




On 6/30/10 9:16 PM, Jared M. Spool wrote:

Erik,

I'd be anxious about trying to design by some sort of segmentation. In my experience, it's unlikely you can come up with any definitive design that will improve the experience, because you're trying to qualify the user on attributes that are probably unrelated to the task at hand.

Instead, I'd focus on behaviors. (Behaviors would be the scenarios that are behind the tasks you mentioned.)

Of course, once you start researching behaviors and identifying the patterns from that research, you've now entered the world of personas and scenarios. If you design to specific personas and scenarios, I think you'll get the experience improvements you're looking for.

Jared

(((
1 Jul 2010 - 4:06am
willdonovan
2009

For me, If the information architecture didn't delineate enough of a clear difference in content needs and user scenarios a mental model approach can identify clear differences in navigation requirements.


William Donovan
mobile: 0403 263 284


On 1 July 2010 17:06, Jared M. Spool <jspool@uie.com> wrote:

"Instead, I'd focus on behaviors. (Behaviors would be the scenarios
that are behind the tasks you mentioned.)
"

I second the motion.

I'd also say do some tree testing to try on some of your structures and
labeling before getting to layout.  During the time I spend focusing on
language, labels, etc... I almost always end up considering several
approaches, and the only way to really know what is most effective,
beyond the obvious, is to experiment.

After all is said and done, and things are up and running, I'd get
started with some MVT and CRO with the live stuff to squeeze out a bit
more effectiveness.  (Is it silly that I still recommend this ALL the
time?  I feel like there's a testing divide, so I never know whether I
should expect people are going to call it old news, or call it sci fi.)

--
Mike Caskey
p. 303-500-3245
c. 303-748-7815

On 6/30/10 9:16 PM, Jared M. Spool wrote:

Erik,

I'd be anxious about trying to design by some sort of segmentation. In
my experience, it's unlikely you can come up with any definitive design
that will improve the experience, because you're trying to qualify the
user on attributes that are probably unrelated to the task at hand.

Instead, I'd focus on behaviors. (Behaviors would be the scenarios that
are behind the tasks you mentioned.)

Of course, once you start researching behaviors and identifying the
patterns from that research, you've now entered the world of personas
and scenarios. If you design to specific personas and scenarios, I
think you'll get the experience improvements you're looking for.

Jared

(((

(((
1 Jul 2010 - 8:20am
Gino Rodrigues
2008

You are right about the divide; it´s old news to most of us, but still sci-fi in many corporate contexts.

The good side of it is that it sticks this comunity together. As long as research/testing remain "out of scope", there will be much more discussion here. =)

30 Jun 2010 - 9:34pm
eriklevitch
2008

Great advice everyone! Thank you so much.

Syndicate content Get the feed