Design Strategy: Archetypes vs. Analytics

16 Jan 2009 - 1:09pm
5 years ago
6 replies
847 reads
Paul Bryan
2008

Hi,

A colleague of mine and I were having coffee recently. I was telling him
about my user archetype (persona) development project. He snickered and
said, ³My team is delivering an individualized design experience based on
hard data. You¹re stuck in design yesteryear.² After this discussion I was
wondering: Is the future of interactive design strategy in the hands of
statisticians? What do you think?

/pb

Paul Bryan
Director, User Research and Experience Design

Comments

16 Jan 2009 - 1:50pm
Traci Lepore
2008

I'd say that he is quite mistaken to think that stats alone are going
to be the best way to do user experience design. While they can tell
you many great things and can help the design, if you don't actually
talk to the user you will never understand the reasons and intents
behind why they are doing anything that results in the stats. Real
transformational design and great experience happens when you have
that deep understanding.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=37301

16 Jan 2009 - 2:42pm
Katie Albers
2005

Ah yes, the "Our information is better because it has numbers!" trope.
You know, if (and I speak very broadly here -- not just about our
field) we weren't so hung up on numbers we could get some real work
done. I warn you, that the following will not be an argument, per se,
but rather a series of reactions.

There was a series of ads about 10 years ago which showed things like
a little old lady riding a Harley and asking why she kept get getting
direct mailings offering Playboy subscriptions and the like. The
general point is that you can't assume that any one person is anything
like your aggregated data set, and the more you rely on the data set
the more likely you are to actually offend some large number of them
with various apparent certainties (e.g., if you like the articles on
Harleys, you'll love this article on the WWF).

It's inconvenient, but people are not the sum of their numbers (and
yes, the pun was intended) and they hate things that make them feel
like you think that they are.

The use of data sets encourages a false sense of knowledge on the part
of people using them. As T.S. Eliot pointed out (and this is a
paraphrase), we have lost wisdom in knowledge, lost knowledge in
information, and lost information in data.

Our job is not *only* functional; if we do it well it is not *only*
measurable; experience is a fundamentally immeasurable thing. If you
use only numerically-based input and methodologies, your work will be
Just Fine...It will also consistently lack that certain, indefinable
something that makes it extraordinary.

The acquisition of the zero, the definition of calculus, the practice
of rigorous statistical analysis, mathematical modelling, all these
things are very important to our world and our culture today. But you
can be fluent in all of them and you still can't use them as the basic
tools of developing a strikingly good interaction, or experience or
interface.

Numbers are attractive because they offer a sense of
Correctness...there is only one right answer (although, as we used to
say at MIT, 2+2=5 for very large values of 2 and very small values of
5). That still doesn't mean that they're always the right tool.

It's been my experience through many stages of the development of
aspects of interactive technology, that when the numbers come in,
everyone in the "class" starts working really hard to make C's. It
certainly isn't something to cite as proof of the advanced nature of
your own development efforts.

Perhaps your colleague was seeking sympathy?

Katie

Katie Albers
Founder & Principal Consultant
FirstThought
User Experience Strategy & Project Management
310 356 7550
katie at firstthought.com

On Jan 16, 2009, at 11:09 AM, Paul Bryan wrote:

> Hi,
>
> A colleague of mine and I were having coffee recently. I was telling
> him
> about my user archetype (persona) development project. He snickered
> and
> said, “My team is delivering an individualized design experience
> based on
> hard data. You’re stuck in design yesteryear.” After this discussion
> I was
> wondering: Is the future of interactive design strategy in the hands
> of
> statisticians? What do you think?
>
> /pb
>
> Paul Bryan
> Director, User Research and Experience Design
>
> ________________________________________________________________
> 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

16 Jan 2009 - 3:04pm
Steve Baty
2009

Speaking as someone with a) a real passion for numbers; b) a degree in
mathematical & statistical modelling; and c) a bit of experience in this
thing we call user experience design - I have to wholeheartedly agree with
Katie's arguments below.

We can learn *a lot* about people either individually or collectively by
surveying, recording, measuring, collating, analysing, aggregating and
reporting. But around the hard knot of clusters of similar-appearing people
such an undertaking might produce, is a whole fuzzy world of real
individuals.

We humans are messy creatures: very hard to pin down, and sometimes prone to
get all contrary when you try. We have emotions, hormones, moods, good days
& bad; we get obsessed, easily distracted; we change our minds. The numbers
won't tell that whole story - ever.

The moment you become solely reliant on numerical data to describe people
you're making a very large and dangerous abstraction that is not justified
by the observed world.

Steve

2009/1/17 Katie Albers <katie at firstthought.com>

> The use of data sets encourages a false sense of knowledge on the part of
> people using them. As T.S. Eliot pointed out (and this is a paraphrase), we
> have lost wisdom in knowledge, lost knowledge in information, and lost
> information in data.
>
> <snip>
>
> The acquisition of the zero, the definition of calculus, the practice of
> rigorous statistical analysis, mathematical modelling, all these things are
> very important to our world and our culture today. But you can be fluent in
> all of them and you still can't use them as the basic tools of developing a
> strikingly good interaction, or experience or interface.

Don't forget negative numbers, the irrationals (pi etc), a complex numbers
(sqrt(-1) = i) ! None of which help explain people any better; although very
good for engineering.

>
>
> Numbers are attractive because they offer a sense of Correctness...there is
> only one right answer (although, as we used to say at MIT, 2+2=5 for very
> large values of 2 and very small values of 5). That still doesn't mean that
> they're always the right tool.

<snip>

>
>
> Katie
>
> Katie Albers
> Founder & Principal Consultant
> FirstThought
> User Experience Strategy & Project Management
> 310 356 7550
> katie at firstthought.com

--
Steve 'Doc' Baty | Principal | Meld Consulting | P: +61 417 061 292 | E:
stevebaty at meld.com.au | Twitter: docbaty | LinkedIn:
www.linkedin.com/in/stevebaty

Blog: http://docholdsfourth.blogspot.com
Contributor - UXMatters - www.uxmatters.com
UX Book Club: http://uxbookclub.org/ - Read, discuss, connect.

16 Jan 2009 - 3:10pm
Cindy Alvarez
2004

Numbers are great. Most designers should embrace them more, and claw and
scream for more metrics where there are none.

BUT - they offer no starting point. You have to put something out there for
people to react to in the first place. And it's more efficient to have a
well-reasoned "first guess" based on personas and general good usability
heuristics. If both "A" and "B" are mediocre, you're not going to convert
them to awesome through incremental tweak-and-test sessions.

Cindy
--
The Experience is the Product - http://www.cindyalvarez.com

On Fri, Jan 16, 2009 at 11:09 AM, Paul Bryan <paul at usography.com> wrote:

> Hi,
>
> A colleague of mine and I were having coffee recently. I was telling him
> about my user archetype (persona) development project. He snickered and
> said, ³My team is delivering an individualized design experience based on
> hard data. You¹re stuck in design yesteryear.² After this discussion I was
> wondering: Is the future of interactive design strategy in the hands of
> statisticians? What do you think?
>
> /pb
>
> Paul Bryan
> Director, User Research and Experience Design
>
> ________________________________________________________________
> 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
>

17 Jan 2009 - 7:26am
Joe Lamantia
2007

Paul

Sounds a bit like a hostile colleague to me. In these sorts of cases,
it's often best to step aside, and let them proceed straight on at
full speed :)

As long as his team's failure (and if they really are looking only at
numbers, and he isn't just pulling your chain, then they will likely
fail in some notable fashion) doesn't adversely affect you, experience
recommends letting them jump off of the quantitative cliff without too
much interference.

If your work is decent, it will only make you look better by
comparison...

Cheers,
Joe Lamantia
On Jan 16, 2009, at 8:09 PM, Paul Bryan wrote:

> Hi,
>
> A colleague of mine and I were having coffee recently. I was telling
> him
> about my user archetype (persona) development project. He snickered
> and
> said, “My team is delivering an individualized design experience
> based on
> hard data. You’re stuck in design yesteryear.” After this discussion
> I was
> wondering: Is the future of interactive design strategy in the hands
> of
> statisticians? What do you think?
>
> /pb
>
> Paul Bryan
> Director, User Research and Experience Design
>
> ________________________________________________________________
> 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

joe at joelamantia.com | www.joelamantia.com

"I have not failed. I've just found 10,000 ways that won't work." -
Thomas Edison

18 Jan 2009 - 9:52am
Paul Bryan
2008

Thank you for these thoughtful replies.

I don%u2019t think my colleague was wrong. He works for a different
company that has been focused on personalized experiences that are
dynamically generated based on historical data.

I was just wondering how quickly, and how thoroughly, you think
quantitative data will displace qualitative data and
%u201Ctraditional%u201D conceptual design processes that are well
known in the web world.

I think you%u2019ve done a great job in pointing out the limitations
of quantitative data as the *sole* basis for design strategy.

Kind regards,
/pb

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=37301

Syndicate content Get the feed