Personas or some other technique?

10 May 2007 - 11:36am
7 years ago
11 replies
1574 reads
tdellaringa
2006

Hi there,

I'm working on a complex portal/business tool project for a major home
services (cleaning) company. I have very limited resources however - I don't
have access to the users in order to do interviews for example. All I have
is a business requirement document, and a spreadsheet showing the primary
"actors" on the system (really, roles). I also have the old system (windows
based) installed on my machine for reference (although it is changing quite
a bit)

I wanted to build some personas as a design tool, but I don't think I have
enough information. Here is the most detailed information I have on one of
the actors:

"Owner - Entrepreneur and strong willed. Older owners are use to having
great flexibility in how they maintain their data and systems. Can be
impatient when unknown and known/expected errors occur that prevent them or
their staff from completing a service day. The schedule board is where they
live and breath. The schedule board is very critical to their operations.
Newer and multi-office owners expect flexibility and functionality in the
application. Security is of the utmost importance to them. They are not
typically doing the day to day activities of closing out records and are not
as familiar with where certain features and functions are. There must be a
clear navigation path to give them a clear understanding of what is
happening or should happen. For multi-office owners having the ability to
view rollup metrics for their operations is a priority."

I have 8 other similar paragraphs, none as complete as this one. One problem
is that the above paragraph details *one* type of owner, but it really is a
stereotype. I doubt they all fit that model. I can guess about the mental
models these users have of the current system (and what they expect) but
they would only be guesses.

I came up with 4 behavioral variables based on the actor sheet and some
other very general info I have, which are:

Frequency of use (low to high)
Patience with the system (low to high)
Comfort level with technology (low to high)
Motivation for use (analysis vs. task oriented)

And I also came up with some *assumed* goals based on the same info:

A smooth running business that is profitable
Quickly and efficiently get my work done
Make my short interaction with the app quick and painless
Easily produce accurate and flexible reporting
Clearly teach new users how to use the system and efficiently support
current users
Does this seem like enough info to provide even a skeleton persona? What
would your approach be given the type of information above, what would you
supply or how would you proceed? The client is anxious to get any help they
can, which is a good thing.

(FYI, the reasons behind why I can only get this amount of info are complex,
let's ignore that issue - it's set in stone)

Any tips are appreciated - please CC me at pixelmech at gee mail dot com.

Thanks!

Tom

Comments

10 May 2007 - 11:48am
Mark Schraad
2006

I like to rank each group on a scale of 1-10 like this for each criteria. Even better if you can get the users to do this themselves. Then pair the variables and do x/y plots and look for clusters. This helps me to understand the interactions between the variables/groups and check the credibiltiy of the 'stereotypes'. It is also a great way for designers to look at this as visual information and from a different perspective. For three criteria you will do three charts and so on...

>
>I came up with 4 behavioral variables based on the actor sheet and some
>other very general info I have, which are:
>
>Frequency of use (low to high)
>Patience with the system (low to high)
>Comfort level with technology (low to high)
>Motivation for use (analysis vs. task oriented)
>
>And I also came up with some *assumed* goals based on the same info:
>

10 May 2007 - 12:17pm
Trip O'Dell
2007

I think Mark's idea is a good one, I would just add one thing. A 1-10
scale by itself is fairly subjective - it might be better to provide
the respondent/evaluator a rubric to score against to "define" for
the purposes of your design what a 1 vs a 10 is. E.g. a 3 for
comfort level with technology might be accompanied with a
descriptive statement of usage - "I am comfortable logging in,
launching and using software, saving files, and using the internet"
whereas a 7 might be - "I'm very comfortable using software, and
frequently customize my system and software for the way I use it".

I'm sure I'm stating the obvious, but I've be the recipient of some
pretty garbage "research" in the past when users were asked to rate
themselves without any frame of reference.

On May 10, 2007, at 11:48 AM, Mark Schraad wrote:

> I like to rank each group on a scale of 1-10 like this for each
> criteria. Even better if you can get the users to do this
> themselves. Then pair the variables and do x/y plots and look for
> clusters. This helps me to understand the interactions between the
> variables/groups and check the credibiltiy of the 'stereotypes'.
>
>>
>> I came up with 4 behavioral variables based on the actor sheet and
>> some
>> other very general info I have, which are:
>>
>> Frequency of use (low to high)
>> Patience with the system (low to high)
>> Comfort level with technology (low to high)
>> Motivation for use (analysis vs. task oriented)
>>
>> And I also came up with some *assumed* goals based on the same info:

10 May 2007 - 12:28pm
Vijay Venkatraman
2006

Hi Tom,

Answering your main question about whether you should use a persona - a big
YES. Personally, I would always use a persona - even if the research data is
as limited as you have. It doesn't matter that your persona is not a long
description - what matters is that you use it as a design tool. I have
discovered that I never forget the research data, user goals or user
requirements when I use personas. If I just think of the persona's name, all
the attributes associated to that persona are loaded into my consciousness.
And if I didn't use personas, I'd have to constantly refer to a boring list
of requirements that I can never remember. Even now I remember every minor
detail about the personas I created a year ago!

You mention that you can make some educated guesses about the users. Given
that you are not going to get any real user data, incorporate those educated
guesses to beef up the persona. I would do that if I strongly believed the
educated guesses were would add significant value. I would use the goals
created out of those guesses to see how the design would mutate if I chose
to use them or not.

- Vijay

On 5/10/07, Tom Dell'Aringa <pixelmech at gmail.com> wrote:
>
> Hi there,
>
> I'm working on a complex portal/business tool project for a major home
> services (cleaning) company. I have very limited resources however - I
> don't
> have access to the users in order to do interviews for example. All I have
> is a business requirement document, and a spreadsheet showing the primary
> "actors" on the system (really, roles). I also have the old system
> (windows
> based) installed on my machine for reference (although it is changing
> quite
> a bit)
>
> I wanted to build some personas as a design tool, but I don't think I have
> enough information. Here is the most detailed information I have on one of
> the actors:
>
> "Owner - Entrepreneur and strong willed. Older owners are use to having
> great flexibility in how they maintain their data and systems. Can be
> impatient when unknown and known/expected errors occur that prevent them
> or
> their staff from completing a service day. The schedule board is where
> they
> live and breath. The schedule board is very critical to their operations.
> Newer and multi-office owners expect flexibility and functionality in the
> application. Security is of the utmost importance to them. They are not
> typically doing the day to day activities of closing out records and are
> not
> as familiar with where certain features and functions are. There must be a
> clear navigation path to give them a clear understanding of what is
> happening or should happen. For multi-office owners having the ability to
> view rollup metrics for their operations is a priority."
>
> I have 8 other similar paragraphs, none as complete as this one. One
> problem
> is that the above paragraph details *one* type of owner, but it really is
> a
> stereotype. I doubt they all fit that model. I can guess about the mental
> models these users have of the current system (and what they expect) but
> they would only be guesses.
>
> I came up with 4 behavioral variables based on the actor sheet and some
> other very general info I have, which are:
>
> Frequency of use (low to high)
> Patience with the system (low to high)
> Comfort level with technology (low to high)
> Motivation for use (analysis vs. task oriented)
>
> And I also came up with some *assumed* goals based on the same info:
>
> A smooth running business that is profitable
> Quickly and efficiently get my work done
> Make my short interaction with the app quick and painless
> Easily produce accurate and flexible reporting
> Clearly teach new users how to use the system and efficiently support
> current users
> Does this seem like enough info to provide even a skeleton persona? What
> would your approach be given the type of information above, what would you
> supply or how would you proceed? The client is anxious to get any help
> they
> can, which is a good thing.
>
> (FYI, the reasons behind why I can only get this amount of info are
> complex,
> let's ignore that issue - it's set in stone)
>
> Any tips are appreciated - please CC me at pixelmech at gee mail dot com.
>
> Thanks!
>
> Tom
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> List Guidelines ............ http://listguide.ixda.org/
> List Help .................. http://listhelp.ixda.org/
> (Un)Subscription Options ... http://subscription-options.ixda.org/
> Announcements List ......... http://subscribe-announce.ixda.org/
> Questions .................. lists at ixda.org
> Home ....................... http://ixda.org/
> Resource Library ........... http://resources.ixda.org
>

10 May 2007 - 12:46pm
Chaunsee at aol.com
2007

In a message dated 5/10/2007 1:29:10 P.M. Eastern Daylight Time,
vijay.discussion at gmail.com writes:

Answering your main question about whether you should use a persona - a big
YES. Personally, I would always use a persona - even if the research data is
as limited as you have.

------------------------------------------------------------------------------
-------------------------------------------
Has your organization used personas before? If not, what is your plan for
implementing personas and getting the product design team to apply them to
various design and evaluation activities. The Personas book by Pruitt and Adlin
(The Persona Lifecycle: Keeping People in Mind Throughout Product Design)
discusses the groundwork needed for teams to accept and apply personas.
Personas generally require some education and public relations for them to be
successful.

Chauncey

************************************** See what's free at http://www.aol.com.

10 May 2007 - 12:54pm
tdellaringa
2006

on 5/10/07, Chaunsee at aol.com <Chaunsee at aol.com> wrote:
>
> In a message dated 5/10/2007 1:29:10 P.M. Eastern Daylight Time,
> vijay.discussion at gmail.com writes:
>
> Answering your main question about whether you should use a persona - a
> big
> YES. Personally, I would always use a persona - even if the research data
> is
> as limited as you have.
>
>
> -------------------------------------------------------------------------------------------------------------------------
> Has your organization used personas before? If not, what is your plan for
> implementing personas and getting the product design team to apply them to
> various design and evaluation activities. The Personas book by Pruitt and
> Adlin (The Persona Lifecycle: Keeping People in Mind Throughout Product
> Design) discusses the groundwork needed for teams to accept and apply
> personas. Personas generally require some education and public relations for
> them to be successful.
>
>
Good question - and the answer is no, they have not. The company is also not
what you would call "up to speed" on UX design. They are pretty much old
school people trying to move in the right direction. I have Cooper's book
which I have been using for my information (Ess. of ID 2). If I provide the
information/tools, they will do their best to use it - but I understand the
point that they may not really be equipped to do so. What would you suggest
I do along those lines? I cannot go there and provide training, and much of
this work may end up being remote freelance work.

Tom

10 May 2007 - 1:51pm
Chaunsee at aol.com
2007

Here are a few good references that bear on the personas that you mention.
These are available on the ACM Digital Library.

Pruitt, J. and Grudin, J. 2003. Personas: practice and theory. In
Proceedings of the 2003 Conference on Designing For User Experiences (San Francisco,
California, June 06 - 07, 2003). DUX '03. ACM Press, New York, NY, 1-15.

Blomquist, Å. and Arvola, M. 2002. Personas in action: ethnography in an
interaction design team. In Proceedings of the Second Nordic Conference on
Human-Computer interaction (Aarhus, Denmark, October 19 - 23, 2002). NordiCHI '02,
vol. 31. ACM Press, New York, NY, 197-200.
This is a case study about implementing personas into the design process.
Here is a quote from the abstract:
"This paper reports a twelve-week participant observation in an interaction
design team with the purpose of learning what really goes on in a design team
when they implement personas in their process. On the surface it seemed like
they used personas, but our analysis show how they had difficulties in using
them and encountered problems when trying to imagine the user."

Khalayli, N., Nyhus, S., Hamnes, K., and Terum, T. 2007. Persona based rapid
usability kick-off. In CHI '07 Extended Abstracts on Human Factors in
Computing Systems (San Jose, CA, USA, April 28 - May 03, 2007). CHI '07. ACM Press,
New York, NY, 1771-1776.

Chauncey

************************************** See what's free at http://www.aol.com.

10 May 2007 - 2:20pm
Vijay Venkatraman
2006

Hi Chauncey,

Well, the answer to that question is both a Yes and No. There are many
product lines in my company (Tektronix) and one product line is using
personas successfully. And that's thanks to Leo Frishberg - he's been
instrumental in that change. You may know him or have read his posts - he's
reasonably active here.

I have worked here for about 10 weeks now and have not been presented with a
formal opportunity to introduce personas. I have joined projects well into
the implementation phase where design means make some minor UI changes based
on user requirements that were gathered by Marketing.

The product lines I work for do not use personas - most have never heard of
that term. I have no strategic plan to popularize personas, but I am
introducing the concept to my clients in conversations. My boss is working
on communicating the value of UX, UX processes, among many things, so I just
try to catalyze that process with my own small steps.
The problem, as I have encountered, is that most don't understand how
personas help the design process. They don't understand why a persona is
needed when they have the set of user requirements. I have asked that
question myself, and through using them, I have realized that their benefits
are greatly in terms of what I said in my response to Tom. I tell them that
and clients respond, "Yea?" and still don't get it.

I have heard of that book by Tamara Adlin - not read it, will look into it.

Best,
Vijay

On 5/10/07, Chaunsee at aol.com <Chaunsee at aol.com> wrote:

> Has your organization used personas before? If not, what is your plan
> for implementing personas and getting the product design team to apply them
> to various design and evaluation activities. The Personas book by Pruitt
> and Adlin (The Persona Lifecycle: Keeping People in Mind Throughout Product
> Design) discusses the groundwork needed for teams to accept and apply
> personas. Personas generally require some education and public relations for
> them to be successful.
>
> Chauncey
>

13 May 2007 - 12:37am
Bernie Monette
2005

> From: Tom Dell'Aringa <pixelmech at gmail.com>
> Date: Thu, 10 May 2007 11:36:35 -0500
> To: <discuss at lists.interactiondesigners.com>
> Subject: [IxDA Discuss] Personas or some other technique?
>
> I have 8 other similar paragraphs, none as complete as this one. One problem
> is that the above paragraph details *one* type of owner, but it really is a
> stereotype. I doubt they all fit that model. I can guess about the mental
> models these users have of the current system (and what they expect) but
> they would only be guesses.
Are you working by yourself? One of the great things about personas is they
act as a means for the developer to express their assumptions and then test
these with the client. So if you have no data or limited data then you and
your team can start to make stuff up (horrors) and use these with your
client or stakeholders.

At some point someone has decided that this thing has a user and now it is
vital to get that assumption out on the table.

Cheers,

Bernie

--
Bernie Monette
InterActive Arts
Internet Presence Management
http://www.iaai.ca monette at iaai.ca 416 469 4337

14 May 2007 - 10:13am
tdellaringa
2006

On 5/10/07, Mark Schraad <mschraad at mac.com> wrote:
>
> I like to rank each group on a scale of 1-10 like this for each criteria.
> Even better if you can get the users to do this themselves. Then pair the
> variables and do x/y plots and look for clusters. This helps me to
> understand the interactions between the variables/groups and check the
> credibility of the 'stereotypes'. It is also a great way for designers to
> look at this as visual information and from a different perspective. For
> three criteria you will do three charts and so on...

Let me see if I have this right. Currently I have 5 variables. I took two
of them and plotted them out on a x/y and got the following, which seems to
give me 3 clusters and one edge case:

http://www.pixelmech.com/rev2/plots.gif

Am I on the right track here? Here I can see that the Franchise owner,
Branch manager and regional manager are all frequent users that are doing a
lot of analytical work on the system, while teammate and captain are
basically one time daily users doing a single simple task. So I could use
both of those clusters as a behavioral trait of a persona?

Tom

14 May 2007 - 10:59am
Mark Schraad
2006

You are very much on the right track.

If you think conceptually in terms of segmentation, grouping your audience into desired attributes, you are cfinding clusters that share similar needs. When you plot all of your cariable you will find similarities (which helps you) and conflicts (which identify some of the more dificult challenges). Hopefully this will help you make resource allocation decisions.

Mark

On Monday, May 14, 2007, at 11:13AM, "Tom Dell'Aringa" <pixelmech at gmail.com> wrote:
>On 5/10/07, Mark Schraad <mschraad at mac.com> wrote:
>>
>> I like to rank each group on a scale of 1-10 like this for each criteria.
>> Even better if you can get the users to do this themselves. Then pair the
>> variables and do x/y plots and look for clusters. This helps me to
>> understand the interactions between the variables/groups and check the
>> credibility of the 'stereotypes'. It is also a great way for designers to
>> look at this as visual information and from a different perspective. For
>> three criteria you will do three charts and so on...
>
>
> Let me see if I have this right. Currently I have 5 variables. I took two
>of them and plotted them out on a x/y and got the following, which seems to
>give me 3 clusters and one edge case:
>
>http://www.pixelmech.com/rev2/plots.gif
>
>Am I on the right track here? Here I can see that the Franchise owner,
>Branch manager and regional manager are all frequent users that are doing a
>lot of analytical work on the system, while teammate and captain are
>basically one time daily users doing a single simple task. So I could use
>both of those clusters as a behavioral trait of a persona?
>
>Tom
>

14 May 2007 - 11:07am
tdellaringa
2006

On 5/14/07, Mark Schraad <mschraad at mac.com> wrote:
>
> You are very much on the right track.
>
> If you think conceptually in terms of segmentation, grouping your audience
> into desired attributes, you are cfinding clusters that share similar needs.
> When you plot all of your cariable you will find similarities (which helps
> you) and conflicts (which identify some of the more dificult challenges).
> Hopefully this will help you make resource allocation decisions.
>
> >http://www.pixelmech.com/rev2/plots.gif
> >
> >Am I on the right track here?

Ok, so I have 5 behavioral variables I came up with (I added one more) and
they are:

Frequency of use (low to high)
Patience with the system (low to high)
Comfort level with technology (low to high)
Motivation for use (analysis vs. task oriented)
Proficiency with the system (inexperienced to very experienced)

For the first plot I did frequency vs. motivation. Do I just plot pairs that
seem to naturally fit together? Seems like patience with the system and
comfort level with tech would be a good one. And again my patience and
proficiency. I assume I don't have to cover every single pairing?

Syndicate content Get the feed