Personas: how many is too many?

17 Feb 2010 - 8:17am
4 years ago
22 replies
4903 reads
suewah
2008

There is this project I'm working on that is for pet owners, breeders
and vets. They have identified 15 consumer types between the three
main segments I've mentioned. The client is requesting we create
personas for all 15. In my experience I've rarely have had to create
more than 4 on any given project. For this project I believe I can get
away with 3.

Does anyone have any thoughts as too what is too many or too few
personas? Have you come across a similar client request? And how did
you deal with it?

Thanks,
Charles Sue-Wah-Sing
www.suewahsing.com

Comments

17 Feb 2010 - 8:39am
Mark Schraad
2006

One way to aproach this is to have them prioritize them... Look at
cost for all... And for a few, and make the decision together. If it's
worth it to them... And they will pay for them... Then let the process
reveal what works best.

Sent from my iPhone

On Feb 17, 2010, at 6:17 AM, charles Sue-Wah-Sing
<charless at nexklix.com> wrote:

> There is this project I'm working on that is for pet owners, breeders
> and vets. They have identified 15 consumer types between the three
> main segments I've mentioned. The client is requesting we create
> personas for all 15. In my experience I've rarely have had to create
> more than 4 on any given project. For this project I believe I can get
> away with 3.
>
> Does anyone have any thoughts as too what is too many or too few
> personas? Have you come across a similar client request? And how did
> you deal with it?
>
> Thanks,
> Charles Sue-Wah-Sing
> www.suewahsing.com
> ________________________________________________________________
> 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 Feb 2010 - 9:02am
Ari Tenhunen
2010

Absolutely too many personas or user groups. It makes management more
complicated in many ways. I would hav max 5.

You need of course some basic major user roles (Administrator,
Breeder (service provider), Owner (Service buyer), Vet (Expert,
service provider) etc.. Each group has their own access privileges
and there are always things that other groups dont need to or shoul
not see.

Maybe there is some hiearchy, 3 groups wth several sub groups?

Maybe you could let each user customize his/her account by selecting
the services and contents he/swe wants. It means, let each user
%u201Dspecify%u201D his/her own persona.

In planning the service you need to understand the role of all those
personas that your client thinks are needed. But it does not
necessarily mean that these need to be built in the system.

The system needs to serve all those groups by
- giving access to all necessary services and contents
- making possible to organize the services to meet the needs and
pririties of each individual user
- hiding all unnecessary things while keeping them available

Did this help? I hope your project will be a success

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=49443

17 Feb 2010 - 9:40am
alfpooh
2009

I wonder how you could get 15 types at first. Is there any criteria?
or from researcher's pattern recognition.

I think those types can be categorized or filtered by criteria.
For example, 5 of 15 might have similar behavior regarding money.
In that case, those 5 could be in a category. But of course, you may
use another criteria.

Criteria can be general keywords such as money, life, family etc.
I believe some of 15 is quite similar and others will be
distinguished.

With some luck, I think you can find patterns of user types which is
less than 15. I mean more effective number for building persona.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=49443

17 Feb 2010 - 9:52am
Victor Lombardi
2003

One

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=49443

17 Feb 2010 - 10:06am
Mark Schraad
2006

Let me add some clarification...

I agree that 15 is probably too many. I also disagree that 5 is the right
number... the answer will be "it depends". Rather than give you a recipe...
I was trying to suggest a process. I assumed you had talked to the client
and that telling them, "15 is too many" had not been effective.

The next step for me would be to put together a scope of effort for the
personas, ask them to prioritize the groups (you should be asking for this
anyway) and then come to conclusions as a team. Some clients will accept
your recommendation and some will not. Those (not) often need to come to
those conclusions on their own. It's better if they have that realization
while working with you.

Mark

On Wed, Feb 17, 2010 at 8:39 AM, mark Schraad <mschraad at gmail.com> wrote:

> One way to aproach this is to have them prioritize them... Look at cost for
> all... And for a few, and make the decision together. If it's worth it to
> them... And they will pay for them... Then let the process reveal what works
> best.
>
>
>
>
> Sent from my iPhone
>
>
> On Feb 17, 2010, at 6:17 AM, charles Sue-Wah-Sing <charless at nexklix.com>
> wrote:
>
> There is this project I'm working on that is for pet owners, breeders
>> and vets. They have identified 15 consumer types between the three
>> main segments I've mentioned. The client is requesting we create
>> personas for all 15. In my experience I've rarely have had to create
>> more than 4 on any given project. For this project I believe I can get
>> away with 3.
>>
>> Does anyone have any thoughts as too what is too many or too few
>> personas? Have you come across a similar client request? And how did
>> you deal with it?
>>
>> Thanks,
>> Charles Sue-Wah-Sing
>> www.suewahsing.com
>> ________________________________________________________________
>> 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 Feb 2010 - 10:25am
zakiwarfel
2004

On Feb 17, 2010, at 7:02 AM, Ari Tenhunen wrote:

> Absolutely too many personas or user groups. It makes management more complicated in many ways. I would hav max 5.

Base it on the data. Look at the overlaps and differences in their behaviors, wants, goals and needs. When you find a significant difference in behavior that warrants a unique design case, then and only then do you create another persona.

There's no magic number of too many or too few. There's only what the data tells you. Coming in and saying "there should never be more than five" without being able to back it up with data will get you into trouble and is frankly inaccurate, which equates to bad design.

Cheers!

Todd Zaki Warfel
Principal Designer, messagefirst
Author of Prototyping: a practitioner's guide http://bit.ly/protobk
----------------------------------
Contact Info
Voice: (215) 825-7423
Email: todd at zakiwarfel.com
Blog: zakiwarfel.com
Twitter: @zakiwarfel
----------------------------------
In theory, theory and practice are the same.
In practice, they are not.

17 Feb 2010 - 11:11am
Anonymous

I agree with Todd to base it on the observed data.

I used your email as an educational note to our internal team. Here is my
email to my team cut and pasted here.

Below – in the attached email “Personas: how many is too many?” - is a very
common problem in using personas. People naively differentiate them on
marketing terms or consumer types. That is not a persona. Personas aren’t
stratified by variables like income (imagine a 0-50K, 50-100K, 100k+ ) or
race, or even task (heart surgeon, bone surgeon, brain surgeon). Marketing
departments try to differentiate target markets based on these variables,
and that is a different methodology which we won’t go into.

Personas are discovered through ethnographic (observational) research, not
made up. They are only differentiated by goals – same goals, same persona.
Personas are a deep finding, kind of like Jungian archetypes (ENTP etc) if
you are familiar with those. Usually primary personas need a specific
product design dedicated for them, because their goals are so different from
the others. Our goal with personas is to ultimately design a product for
each one, and thereby move from making one all-inclusive product trying to
serve all needs for everyone, to specific products for each “Primary
Persona”.

Navid

On Wed, Feb 17, 2010 at 12:17 AM, charles Sue-Wah-Sing <charless at nexklix.com
> wrote:

> There is this project I'm working on that is for pet owners, breeders
> and vets. They have identified 15 consumer types between the three
> main segments I've mentioned. The client is requesting we create
> personas for all 15.

17 Feb 2010 - 11:21am
Sara Durning
2009

I've experienced a similar situation where the client wanted a person
for each segment.

In our case, we were able to convinence the client post-interviews
that the core goals of all segments were satisfied by 3 primary
personas.

The develoment and use of personas is partly an education process as
it is a shift in mindset for many clients from marketing
(demographics) to understanding how people actually use the site
(goals, attitudes and behaviours).

A great resource is Steve Mulder's "The user is Always Right". He
has a digital version on his website:
http://www.practicalpersonas.com

Sara

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=49443

17 Feb 2010 - 11:30am
Anonymous

I think Navid hits it on the head.

Too often personas get mixed up in that crazy world of "market
segmentation" where they really don't belong (at least not
traditional age/income/race market segmentation).

Especially when dealing with web pages, I tend to think of personas
in terms of pathways and IA. There are only so many tasks a visitor
can perform at your site, and ideally you lead them through the
logical steps to their goal. Each pathway is built from the persona.

Ultimately, I agree that 15 sounds like a lot, and that those 15
probably share a number of goals that could be grouped into fewer
ultimate personas.

You can't be everything to everyone, but ignoring potential
customers is a hard mindset to instill.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=49443

17 Feb 2010 - 11:53am
Susan Oslin
2010

Keep in mind that personas are based on goals and not traditional
market segmentation. It sounds to me like the goals of the site have
not been clearly defined and the client is in the world of the
traditional marketing mentality. Criteria for defining the user
experience is distinct from criteria for determining a products
feasibility in the market and how/who to distribute to which is the
goal of market segmentation. Market segmentation can inform goal
definition and is useful information but says nothing about why a
user makes certain choices. Answering why will provide you with the
goals.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=49443

17 Feb 2010 - 4:37pm
alfpooh
2009

Why we can say "5" ? If you are experienced researcher, 15 will be
too many for most of project. Actually there is another reason from
the other side: the client.

Persona is a tool to understand users and to make a model from it. So
as a tool, it should be effective.

Once we had 4 major personae. When we add more, we couldn't remember
all persona's name often.

Yes, it's the magic number. 10 personae will be hard to remember
than 5 for your client and you. So 5 can be more effective than 15 in
general.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=49443

17 Feb 2010 - 4:56pm
Vicky Teinaki
2008

If you go back to the source (Cooper's the Inmates Are Running the Asylum)
Cooper suggests 3-12, but also points out that many personas have
overlapping needs.

The cast of characters (3-12) - page 135
http://books.google.co.nz/books?id=04cFCVXC_AUC&lpg=PP1&ots=jfzhX3u_qN&dq=the%20inmates%20are%20running%20the%20asylum%20personas&pg=PA135#v=onepage&q=personas&f=false

Sample list - page 143
http://books.google.co.nz/books?id=04cFCVXC_AUC&lpg=PP1&ots=jfzhX3u_qN&dq=the%20inmates%20are%20running%20the%20asylum%20personas&pg=PA143#v=onepage&q=personas&f=false

17 Feb 2010 - 5:28pm
Hilary Bienstock
2009

On reading this discussion, my thought is that it might be best to focus the design on one segment each, and then make the personas necessary just for that segment.

However, 15 is still probably too many. The key to condensing them would be to focus on goals. If persona A's can be satisfied any time B is satisfied, but B has additional needs, then A is unnecessary. An analysis of this type would hopefully help you and you stakeholders whittle down the list. Perhaps you can include them in the goal comparing meeting so that the idea of condensing the personas down seems to come from them?

Hilary

Hilary User Experience
Hilary Bienstock, Principal

hilary at hilaryue.com :: 310.883.5818 :: fax 310.829.2839

________________________________
From: Alf <alfpooh at gmail.com>
To: discuss at ixda.org
Sent: Wed, February 17, 2010 6:37:03 AM
Subject: Re: [IxDA Discuss] Personas: how many is too many?

Why we can say "5" ? If you are experienced researcher, 15 will be
too many for most of project. Actually there is another reason from
the other side: the client.

Persona is a tool to understand users and to make a model from it. So
as a tool, it should be effective.

Once we had 4 major personae. When we add more, we couldn't remember
all persona's name often.

Yes, it's the magic number. 10 personae will be hard to remember
than 5 for your client and you. So 5 can be more effective than 15 in
general.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=49443

________________________________________________________________
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 Feb 2010 - 7:28pm
Petteri Hiisilä
2004

On 18.2.2010, at 0.56, Vicky Teinaki wrote:

> If you go back to the source (Cooper's the Inmates Are Running the Asylum)
> Cooper suggests 3-12, but also points out that many personas have
> overlapping needs.
>
> The cast of characters (3-12) - page 135
> http://books.google.co.nz/books?id=04cFCVXC_AUC&lpg=PP1&ots=jfzhX3u_qN&dq=the%20inmates%20are%20running%20the%20asylum%20personas&pg=PA135#v=onepage&q=personas&f=false
>
> Sample list - page 143
> http://books.google.co.nz/books?id=04cFCVXC_AUC&lpg=PP1&ots=jfzhX3u_qN&dq=the%20inmates%20are%20running%20the%20asylum%20personas&pg=PA143#v=onepage&q=personas&f=false

Kim Goodwin, VP Design & General Manager @Cooper, gives a complete description about how to create and work with personas in her 2009 book: Designing for the Digital Age (foreword by Alan Cooper). This is the latest and most exhaustive source for the topic right now. Tons of examples, images and exercises.

It's also the practitioner's guide to Goal-Directed Design and in my opinion should be standard reading in any IxD curriculum. Highly recommended. On a scale of one to five, I'd give it six stars.

http://www.cooper.com/insights/books/

As for the original question: Todd's answer is spot on. This article explains some of the differences between personas and marketing segments: http://www.cooper.com/journal/2002/03/reconciling_market_segments_an.html

Also read this one: http://www.cooper.com/journal/2001/08/perfecting_your_personas.html

Cheers,
Petteri

--
Petteri Hiisilä
Senior Interaction Designer, owner / iXDesign

Mobile: 408-256-0430 (USA), +358505050123 (FIN)
Twitter: http://twitter.com/petterihiisila
LinkedIn: http://www.linkedin.com/in/petterihiisila

17 Feb 2010 - 10:04am
Challis Hodge
2003

It sounds as they're looking for market segmentation rather than
personas. Personas should be built in a way that the fewest possible
can represent the needs of all. This facilitates efficiency in design.
Market segmentation and resulting profiles can be quite complex,
detailed and specific to facilitate marketing needs.

So one way to answer the question is: no more than you can effectively
and efficiently design for. Which for me tends to max out somewhere
around 5, 6 or 7.

-challis

>
> On Feb 17, 2010, at 6:17 AM, charles Sue-Wah-Sing <charless at nexklix.com>
> wrote:
>
>> There is this project I'm working on that is for pet owners, breeders
>> and vets. They have identified 15 consumer types between the three
>> main segments I've mentioned. The client is requesting we create
>> personas for all 15. In my experience I've rarely have had to create
>> more than 4 on any given project. For this project I believe I can get
>> away with 3.
>>
>> Does anyone have any thoughts as too what is too many or too few
>> personas? Have you come across a similar client request? And how did
>> you deal with it?
>>
>> Thanks,
>> Charles Sue-Wah-Sing
>> www.suewahsing.com

17 Feb 2010 - 11:35am
Anindita
2010

Hey.. This is an interesting topic since I have been in this position
myself before. I was able to convince my client that these many
personas would mean additional expenses in terms of development,
maintenance and management of requirements. But if your clients
aren't too concerned about their budget (which is a novelty I might
add.. enjoy it :)) what you could consider is to identify why they
would consider 15 personas to start with. The way we do user groups
is to consider all possible user groups and identify which are the
primary user groups and which are secondary. Primary users would the
ones that cover a majority of the tasks/requirements. This will
really help identify which user groups are a mild "version" of each
other and which ones really need to developed. Since we would need to
tie back design decisions with each persona to justify decisions it
would be good to design to the x number of personas that cover close
to 90% of your requirements. This almost always winds down to being
3-5 personas for me. Hope this helps!

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=49443

17 Feb 2010 - 11:20am
Bryce Johnson
2007

I work on a project where there is over 60 now and to be honest they
are all useful. It really depends on what you need them for.

If you are interested you can find a PDF poster of most of them here:
http://www.microsoft.com/dynamics/en/us/familiar.aspx
Look for the The Microsoft Dynamics Customer Model. Each Persona on
the poster has about 6 pages of background. It's totally overwhelming
but useful when you need it.

Bryce Johnson
Tw: @brycej | http://powerpivotor.com

On Tue, Feb 16, 2010 at 10:17 PM, charles Sue-Wah-Sing
<charless at nexklix.com> wrote:

> Does anyone have any thoughts as too what is too many or too few
> personas? Have you come across a similar client request? And how did
> you deal with it?
>
> Thanks,
> Charles Sue-Wah-Sing
> www.suewahsing.com

17 Feb 2010 - 9:46am
Sue Chase
2010

I don't believe there is a such thing as too many or too few
personas. Each project has it's own unique requirements. My first
suggestion is to be thorough in your analysis of each role. The
system needs to meet everyone's needs.

If you believe you can effectively use only 3 personas then start
building your argument for it. Do a cost analysis and in detail
explain how 3 personas can be just as effective as 15. If they are
not willing to go down to 3, try a compromise with 5 or 7. Just keep
working from there until you reach an effective but most of all
efficient number of personas.

Also, be open to using all 15 personas, it can actually be beneficial
and may help you create a great product instead of a good product.

I hope this helps.
Gala

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=49443

17 Feb 2010 - 11:46am
Nick de Voil
2009

What in the world have Jungian archetypes got to do with pet websites?

I don't think Victor is very wide of the mark.

http://www.devoil.com/papers/PersonasConsideredHarmful.pdf

Nick

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=49443

18 Feb 2010 - 6:40am
William Hudson
2009

I've not been following the whole conversation, but of course even 15
personas would be preferable to 15 market segment/user needs analyses.
The point of personas is that they promote empathy - or would do if we
make them 'believable', hence the need for seemingly unimportant
details.

I wrote a blog about this a few months back for a different reason, but
I think it's relevant here, too.

http://www.syntagm.co.uk/design/blog/?p=89

Regards,

William Hudson
Syntagm Ltd
Design for Usability
UK 01235-522859
World +44-1235-522859
US Toll Free 1-866-SYNTAGM
mailto:william.hudson at syntagm.co.uk
http://www.syntagm.co.uk
skype:williamhudsonskype

Syntagm is a limited company registered in England and Wales (1985).
Registered number: 1895345. Registered office: 10 Oxford Road, Abingdon
OX14 2DS.

Attend our courses on Ajax design & usability, card sorting and web
usability:
CHI 2010 Conference (Atlanta, Georgia) http://www.chi2010.org
UPA 2010 Conference (Munich, Germany) http://bit.ly/8x9NU

18 Feb 2010 - 11:44am
suewah
2008

That's for everyone's input. All valid.

Navid, I really like your summation of personas you used with your
group.

The problem here as with most online initiatives is that the
marketing team is at the table. And it is an education challenge.
Personas are not a means to understand marketing opportunities within
new audiences. Though that can be a pleasant side effect - it is not
the intention.

I'm still of belief 15 personas is too much for most projects I've
been on. But could be very necessary if the scale and complexity of
the project demands so. And I like some of your suggestions about
weeding out personas based on overlapping goals, intentions and
needs.

Thanks All
Charles

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=49443

5 Mar 2010 - 1:49am
koblavi
2010

As far as the issue of personas go, I’d stick to Allan Cooper’s guidelines to creating and using personas in the book, The inmates are Running the Asylum. I must however point out that he does not  actually recommend between 3 and 12, rather he sites this rage as typical in the projects that he or he consulting firm has worked on.

“We give every project its own cast of characters, which consists of anywhere from 3 to 12 personas.”

First of all I also think that for the kind of project you described 15 personas is a pretty large number. However depending on how these personas relate to the product, they’ll be either more or less important in the design of your product. Here are some things you might want to consider

·         Identify your direct personas (i.e. those who will be directly interacting with your product) and focus the design on them. Personas that do not have any direct interaction with the product will mostly be helpful in setting up scenarios. This should reduce the number of personas you will actually consider during the design of the product

·         Identify your primary personas: That persona whose requirements are very different from that of the others. The design of your product around their requirements and consider those of your secondary personas throughout the process.

·         Cancel out the commons: Some of your personas’ attributes may intersect. If this tends to occur with your chosen personas you might want to cancel out these common attributes that they share with other personas, this should eventually bring you down to a more appreciable yet representative number.

·         Do not be too generic:  As far as user centered design goes, every user has specific needs. If your choice of 15 personas is to satisfy 15 different types of direct users then you might be stretching it a bit too much. Chances are that all 15 types would be dissatisfied with an overly generic product. Restate the goals of your product and let those steer the design. Not the personas.

Personas as you may know can be very helpful for design. However their misuse may lead to complications. I hope the tips are helpful.

Syndicate content Get the feed