Why are personas written from the third-person perspective so pervasive?

17 Jun 2011 - 9:23pm
3 years ago
61 replies
3936 reads
Byron Stewart
2008

First-person personas are more engaging, and by mimicking tone patterns, grammar usage and other linguistic characteristics that may be possessed by the persona would result in the ability to communicate more with less writing.

 

Rarely have I come across a persona written from the first-person perspective. I’m wondering if there are reasons for this discrepancy, or is it strictly due to design history/protocol? 



Comments

17 Jun 2011 - 11:44pm
penguinstorm
2005

It's somewhat natural for people to fall into third person when they think of someone else's perspective. People tend to forgot that they're supposed to imagine they ~are~ someone else, and so it goes.

It's the same reason people fall into passive voice so often. I work as a technical writer but I put a sticky on my monitor that says "No Passive Voice" as a reminder.

18 Jun 2011 - 5:30am
Charu Ahuja
2008

You bring up a very good point. Using first person perspective seems so logical and yet I just looked back at my last set of personas and realized I wrote them in third person. Most of the Persona samples I've seen include copy in the third person. Even if it is more intuitive to think in first person, many people probably just create something similar to what they've seen. Often times if writing is not your background, voice and perspective are not always top of mind. Thanks for bringing this up!

18 Jun 2011 - 11:13am
Don Dunbar
2009

No, we write in third person to remind ourselves and others that there is bias in our personas. No matter how hard we try, we are not the person the persona represents. Personas are built on conglomerations of traits from a target population they are not real. In this sense personas are really just metaphors for real people. We use them as a short cut to having actual participants because of the difficulty in getting sample populations and to save time in the long run. Writing in first person disguises the bias and is too easily interpreted as real.

18 Jun 2011 - 2:05pm
Lisa Goldberg
2007

During a presentation at work, everyone on the team introduced themselves as a persona and provided some background information about this representative character, with specific details ("I am looking for information about [topic] for my husband, because his treatment isn't working."). This worked very well. Everyone knew who I was but talking in the first person really brought this persona to life.

Lisa

18 Jun 2011 - 2:47pm
Byron Stewart
2008

Hey Lisa,

This sounds like a great exercise.

Were participants working with research based personas, or creating background on the spot?

Thanks,

Byron

18 Jun 2011 - 2:36pm
Byron Stewart
2008

 

Hey Don,

Can you please elaborate on this point.

"Writing in first person disguises the bias and is too easily interpreted as real."

How does writing in the first person disguise bias?

Thanks

Byron

19 Jun 2011 - 12:05pm
Don Dunbar
2009

Byron, my point is really about the abstract : it doesn't really matter which you use "first" or "third" person. But, it is important to remember, even if you have done what Jared recommends below, that there is still a bias. Bias is always present. We can eliminate some bias, but not all. That may not affect the final outcome, but we still need to keep in mind that it is there. Even when you start, as Jared suggests, with real people, the persona you develop has your bias ingrained. In a very simplified thought experiment, try calculating the following:

 2 x 25

2 x 50

2 x 100

Now, you know maths, so you probably can do these calculations in your head, and you recognize that each subsequent calculation holds the answer to the previous one.  Now imagine that you don't know math at this level. What are your reactions to approaching the problems if you just learning this material ( assume that you can interview students just learning this). You should be able to get insight into the student's behavior, their wants and needs etc. But, you are not that student and either is the persona you create from your interviews.  You can only get a sense of what it is like to learn this the first time because the knowledge is still in your head. 

Writing or thinking of the persona in the third person isn't a big deal, but it does, albeit subtly, suggest the underlying bias present. That bias is probably what is called unintentional, and it may not adversely affect your decisions but it is still important to remember it is there.

19 Jun 2011 - 4:06pm
Louise Hewitt
2010

I don't frankly care much about whether persona's are first or third person, and - as Jared said - it's mostly about the thinking behind them.

What does bother me is their role in the design process.

I'm tired of walking into project with a bunch of highly detailed, stylish personas which get packed away. A persona is an expression of a process of thinking about and imagining the requirements/context etc of a user(s), right?

All good, and true the act of doing this creates a residual instinct for the next stages.

But what I do find myself using them for (as documents) is a reference point for creating or validating user journeys. Actually, I prefer 'exposing' user journeys - as if I really done a proper persona, and I understand the context and requirements of the projects goal, then the user journey writes itself. I find it essential that the user journey must be done in the first person..

1 - I am totally bored of my old phone 2 - I'm going to spend my lunch break using my desktop computer to look at other phones and deals I could get 3 - I'm going to (hey, would she google something here, or visit a site she knows or maybe her contract operator)

Ta da! I don't have to decide because my persona will have the answer.

etc. until I know all the user journey, can spot where it doesn't match the current project and need mitigating, can see opportunities to engage that weren't previously know. Before you know it, the interface/IA has designed itself!

So can I rewrite your question: "Why are personas given so much attention, but not their life-cycle as a project tool" :D

Lou.

On 19 Jun 2011, at 19:44, Don Dunbar wrote:

> Byron, my point is really about the abstract : it doesn't really matter which you use "first" or "third" person. But, it is important to remember, even if you have done what Jared recommends below, that there is still a bias. Bias is always present. We can eliminate some bias, but not all. That may not affect the final outcome, but we still need to keep in mind that it is there. Even when you start, as Jared suggests, with real people, the persona you develop has your bias ingrained. In a very simplified thought experiment, try calculating the following: > > 2 x 25 > > 2 x 50 > > 2 x 100 > > Now, you know maths, so you probably can do these calculations in your head, and you recognize that each subsequent calculation holds the answer to the previous one. Now imagine that you don't know math at this level. What are your reactions to approaching the problems if you just learning this material ( assume that you can interview students just learning this). You should be able to get insight into the student's behavior, their wants and needs etc. But, you are not that student and either is the persona you create from your interviews. You can only get a sense of what it is like to learn this the first time because the knowledge is still in your head. > > Writing or thinking of the persona in the third person isn't a big deal, but it does, albeit subtly, suggest the underlying bias present. That bias is probably what is called unintentional, and it may not adversely affect your decisions but it is still important to remember it is there. > >

29 Jun 2011 - 6:05am
LeeMcIvor
2010

Not sure I agree this improves awareness of bias, even on a basic level. If I write - "he likes bacon sandwiches for breakfast" or alternatively "I like bacon sandwiches for breakfast", how does the former imply bias while the latter doesn't? They're both assertions of fact.
For what it's worth, I tend to write them in the first person, because the document is supposed to represent my participants. Also, my personas are always generated from qualitative research, and I use quotes from the actual research to make up the written narrative on a persona. In this case, the only bias is the natural cognitive bias of the participants themselves in recalling past events, not mine. Obviously how I've combined those quotes and formed the personas has been done subjectively by me, but does use of the first or third person in the narrative really make that clear?
I would say if you use research quotes, then write them as the participant spoke - generally in the first person. Emphasise to your design team, your client, and whoever else at all times that these are representations of research and analysis, not actual people. Don't use your prose to do it for you in such a subtle way.


On 19 June 2011 19:29, Don Dunbar <salemd1ster@gmail.com> wrote:

Byron, my point is really about the abstract : it doesn't really matter which you use "first" or "third" person. But, it is important to remember, even if you have done what Jared recommends below, that there is still a bias. Bias is always present. We can eliminate some bias, but not all. That may not affect the final outcome, but we still need to keep in mind that it is there. Even when you start, as Jared suggests, with real people, the persona you develop has your bias ingrained. In a very simplified thought experiment, try calculating the following:

 2 x 25

2 x 50

2 x 100

Now, you know maths, so you probably can do these calculations in your head, and you recognize that each subsequent calculation holds the answer to the previous one.  Now imagine that you don't know math at this level. What are your reactions to approaching the problems if you just learning this material ( assume that you can interview students just learning this). You should be able to get insight into the student's behavior, their wants and needs etc. But, you are not that student and either is the persona you create from your interviews.  You can only get a sense of what it is like to learn this the first time because the knowledge is still in your head. 

Writing or thinking of the persona in the third person isn't a big deal, but it does, albeit subtly, suggest the underlying bias present. That bias is probably what is called unintentional, and it may not adversely affect your decisions but it is still important to remember it is there.

(((P
29 Jun 2011 - 11:48am
Don Dunbar
2009

"Not sure I agree this improves awareness of bias, even on a basic level. If I write - "he likes bacon sandwiches for breakfast" or alternatively "I like bacon sandwiches for breakfast", how does the former imply bias while the latter doesn't? They're both assertions of fact. 

For what it's worth, I tend to write them in the first person, because the document is supposed to represent my participants. Also, my personas are always generated from qualitative research, and I use quotes from the actual research to make up the written narrative on a persona. In this case, the only bias is the natural cognitive bias of the participants themselves in recalling past events, not mine. Obviously how I've combined those quotes and formed the personas has been done subjectively by me, but does use of the first or third person in the narrative really make that clear? 
I would say if you use research quotes, then write them as the participant spoke - generally in the first person. Emphasise to your design team, your client, and whoever else at all times that these are representations of research and analysis, not actual people. Don't use your prose to do it for you in such a subtle way. "

 

I think you are missing the basic point, it doesn't matter which one you use, first or third person. What matters is recognizing that bias exists in creating perosonas.  Using quotes and using first person does not remove bias by itself. 

"Also, my personas are always generated from qualitative research, and I use quotes from the actual research to make up the written narrative on a persona. In this case, the only bias is the natural cognitive bias of the participants themselves in recalling past events, not mine."

Just no. Your bias is still there. It is as Jared says, the persona is in your head not the document. You still have to interpret the participant's quotes and make meaning from them. Again, there are many studies in Science, Philosophy and even Psychology that demonstrate the presence of bias even when we think it is eliminated from our research.  

But I think there is even more. You use the example, "I like bacon sandwiches for breakfast." Well, what does that mean? Is it a suggestion that bacon sandwiches are not normal breakfast fare? Is it quirky, independent minded? Or, does it mean, among other breakfast fare, I like bacon sandwiches? Because, bacon sandwiches are a normal breakfast item? Depending on how you interpret that example leads you to different conclusions about how your persona will behave in different circumstances. It's a different insight. Which one is right? Maybe none of them. 

So maybe, being subtle can be beneficial? Maybe acknowledging bias in anyway we can helps? 

29 Jun 2011 - 5:05pm
cfmdesigns
2004

It clearly means that he wants minimal to no interaction with the touch screen or mouse during the activity, so the screen/peripheral doesn't get greasy!

30 Jun 2011 - 6:05am
LeeMcIvor
2010

Hi Don,
I don't think I made myself clear to be honest. Firstly, my point was indeed that 1st or 3rd person doesn't matter. On the other hand you specifically state that using 3rd person indicates the possibility of bias, and I still doubt that it helps in that way. The fact I write them in the 1st person was just to illustrate that some of us do, as the original poster asked. I'm not saying that either is significantly better, just that neither is particularly useful for indicating bias.
The other point, which I did make but I think you missed, is that I know there is bias in the act of creating a persona, as it's naturally subjective. I said as much - "Obviously how I've combined those quotes and formed the personas has been done subjectively by me". What I also said was that the use of quotes from research participants meant that those quotes contained only the bias of the participants themselves when expressing their points. Again, how I select those quotes and use them in a persona introduces my own bias into the persona, but not the individual quote.
Hope that's clarified my point. Bias is there when you create personas, using 1st or 3rd person narrative doesn't make any difference!



On 29 June 2011 17:48, Don Dunbar <salemd1ster@gmail.com> wrote:

"Not sure I agree this improves awareness of bias, even on a basic level. If I write - "he likes bacon sandwiches for breakfast" or alternatively "I like bacon sandwiches for breakfast", how does the former imply bias while the latter doesn't? They're both assertions of fact. 

For what it's worth, I tend to write them in the first person, because the document is supposed to represent my participants. Also, my personas are always generated from qualitative research, and I use quotes from the actual research to make up the written narrative on a persona. In this case, the only bias is the natural cognitive bias of the participants themselves in recalling past events, not mine. Obviously how I've combined those quotes and formed the personas has been done subjectively by me, but does use of the first or third person in the narrative really make that clear? 
I would say if you use research quotes, then write them as the participant spoke - generally in the first person. Emphasise to your design team, your client, and whoever else at all times that these are representations of research and analysis, not actual people. Don't use your prose to do it for you in such a subtle way. "

 

I think you are missing the basic point, it doesn't matter which one you use, first or third person. What matters is recognizing that bias exists in creating perosonas.  Using quotes and using first person does not remove bias by itself. 

"Also, my personas are always generated from qualitative research, and I use quotes from the actual research to make up the written narrative on a persona. In this case, the only bias is the natural cognitive bias of the participants themselves in recalling past events, not mine."

Just no. Your bias is still there. It is as Jared says, the persona is in your head not the document. You still have to interpret the participant's quotes and make meaning from them. Again, there are many studies in Science, Philosophy and even Psychology that demonstrate the presence of bias even when we think it is eliminated from our research.  

But I think there is even more. You use the example, "I like bacon sandwiches for breakfast." Well, what does that mean? Is it a suggestion that bacon sandwiches are not normal breakfast fare? Is it quirky, independent minded? Or, does it mean, among other breakfast fare, I like bacon sandwiches? Because, bacon sandwiches are a normal breakfast item? Depending on how you interpret that example leads you to different conclusions about how your persona will behave in different circumstances. It's a different insight. Which one is right? Maybe none of them. 

So maybe, being subtle can be beneficial? Maybe acknowledging bias in anyway we can helps? 

(((Please
30 Jun 2011 - 9:05am
Alan James Salmoni
2008

Is there room for both 1st and 3rd person?

So with a 'feature' of a persona, there is something in the 3rd person like "Regularly shops online with trusted sellers" which is then expanded upon in the first person to provide a 'flavour': "Yeah, I'm happy to buy at Amazon or Ebay but it has to be someone I know 'cause I got stiffed once buying something dodgy".

The aim of a persona is to communicate an archtype to project team / clients / stakeholders, etc, and make them more 'imageable' and this seems a richer to do it. It begins with a dry but factual description and follows with something more likely to generate empathy.

Apologies for ignoring the bias argument and focus on the original question (which is a bum as I quite like reading about measurement error)

Alan

On Thu, Jun 30, 2011 at 12:45 PM, LeeMcIvor wrote: > Hi Don, > I don't think I made myself clear to be honest. Firstly, my point was indeed > that 1st or 3rd person doesn't matter. On the other hand you specifically > state that using 3rd person indicates the possibility of bias, and I still > doubt that it helps in that way. The fact I write them in the 1st person was > just to illustrate that some of us do, as the original poster asked. I'm not > saying that either is significantly better, just that neither is > particularly useful for indicating bias. > The other point, which I did make but I think you missed, is that I know > there is bias in the act of creating a persona, as it's naturally > subjective. I said as much - "Obviously how I've combined those quotes and > formed the personas has been done subjectively by me". What I also said was > that the use of quotes from research participants meant that those quotes > contained only the bias of the participants themselves when expressing their > points. Again, how I select those quotes and use them in a persona > introduces my own bias into the persona, but not the individual quote. > Hope that's clarified my point. Bias is there when you create personas, > using 1st or 3rd person narrative doesn't make any difference! > > On 29 June 2011 17:48, Don Dunbar wrote: > >> "Not sure I agree this improves awareness of bias, even on a basic level. >> If I write - "he likes bacon sandwiches for breakfast" or alternatively "I >> like bacon sandwiches for breakfast", how does the former imply bias while >> the latter doesn't? They're both assertions of fact. >> >> For what it's worth, I tend to write them in the first person, because the >> document is supposed to represent my participants. Also, my personas are >> always generated from qualitative research, and I use quotes from the actual >> research to make up the written narrative on a persona. In this case, the >> only bias is the natural cognitive bias of the participants themselves in >> recalling past events, not mine. Obviously how I've combined those quotes >> and formed the personas has been done subjectively by me, but does use of >> the first or third person in the narrative really make that clear? >> >> I would say if you use research quotes, then write them as the participant >> spoke - generally in the first person. Emphasise to your design team, your >> client, and whoever else at all times that these are representations of >> research and analysis, not actual people. Don't use your prose to do it for >> you in such a subtle way. " >> >> >> >> I think you are missing the basic point, it doesn't matter which one you >> use, first or third person. What matters is recognizing that bias exists in >> creating perosonas.  Using quotes and using first person does not remove >> bias by itself. >> >> "Also, my personas are always generated from qualitative research, and I >> use quotes from the actual research to make up the written narrative on a >> persona. In this case, the only bias is the natural cognitive bias of the >> participants themselves in recalling past events, not mine." >> >> Just no. Your bias is still there. It is as Jared says, the persona is in >> your head not the document. You still have to interpret the participant's >> quotes and make meaning from them. Again, there are many studies in Science, >> Philosophy and even Psychology that demonstrate the presence of bias even >> when we think it is eliminated from our research. >> >> But I think there is even more. You use the example, "I like bacon >> sandwiches for breakfast." Well, what does that mean? Is it a suggestion >> that bacon sandwiches are not normal breakfast fare? Is it quirky, >> independent minded? Or, does it mean, among other breakfast fare, I like >> bacon sandwiches? Because, bacon sandwiches are a normal breakfast item? >> Depending on how you interpret that example leads you to different >> conclusions about how your persona will behave in different circumstances. >> It's a different insight. Which one is right? Maybe none of them. >> >> So maybe, being subtle can be beneficial? Maybe acknowledging bias in >> anyway we can helps? >> >> (((Please >> > (((

30 Jun 2011 - 10:57am
Don Dunbar
2009

Hi Lee, this response was much more clear to me than your original post. I think we are on the same track.

19 Jun 2011 - 4:06pm
monkeyshine
2010

I agree with the statement that writing in first person disguises the bias. I'm sure a great writer could pull off a 1st person persona well to great affect, but if you've ever tried to write a piece of fiction in first person you understand how incredibly difficult it is to do well. I think the risk is high of having the format/tools get in the way of the persona goal. I've used quotes from actual users/customers to strengthen the impact of the persona but that's as far as I'd go.
Deanna


On Sat, Jun 18, 2011 at 12:19 PM, Don Dunbar <salemd1ster@gmail.com> wrote:

No, we write in third person to remind ourselves and others that there is bias in our personas. No matter how hard we try, we are not the person the persona represents. Personas are built on conglomerations of traits from a target population they are not real. In this sense personas are really just metaphors for real people. We use them as a short cut to having actual participants because of the difficulty in getting sample populations and to save time in the long run. Writing in first person disguises the bias and is too easily interpreted as real.

19 Jun 2011 - 8:09am
Johann Sarmiento
2008

Interesting discussion.  The last set of persona documents that I have been involved with are for internal use at a large organization and they are all in the first person (with sections like:  "Things you will hear me say,"  "What is critical to my success is..."  etc.)   I think that they are very compelling this way, but I very much agree with Jared: at the end, it is all about the sustained conversations you promote around these and other artifacts with designers, researchers, stakeholders and others.

--Johann

(((Pl

19 Jun 2011 - 10:05am
Byron Stewart
2008

Hi Johann,

What did you find most compelling about personas written in the first person?

Thanks,

Byron

18 Jun 2011 - 7:57pm
Jared M. Spool
2003

We've done a lot of research on effective personas. I haven't seen many first-person personas, but well-executed third-person personas do fine. I can imagine that the same persona descriptions, written in the first person, if done well, would also be fine.

The thing to remember is that personas are not the document. The document — and thus the first- or third-person description — are just a souvenier.

For a persona to be effective in the design and development process, everyone needs to internalize who that person is. They need to understand that person and relate to them. It doesn't matter what the souvenier document looks like. It only matters what's in the head of the designer or developer.

A few years back, I wrote a piece on this, called Personas Are NOT A Document. It goes into a little more detail.

Jared

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

18 Jun 2011 - 10:26pm
Byron Stewart
2008

Hi Jared,

It sounds like your saying, process vs product is that right?

Thanks for the link to additional resources.

Byron

 

19 Jun 2011 - 7:34am
Jared M. Spool
2003

It sounds like your saying, process vs product is that right?

Only partially.

The process is important for persona creation. When you look at the teams that get the most from using personas, their members are intimately familiar with the real people who made up each persona. Each team member has met the real users, watching them work in their real environments.

In our research, the teams that fail to glean anything useful from creating personas have something in common: they just made up the personas from their own imaginations without having done any research to see who the real users behind them should be.

You're correct in that the persona creation process is very important to the success of persona usage.

However, the product is also important. At the end of the persona creation process, the team should have a very clear understanding of who the personas are, what the scenarios are, and how the personas and scenarios will influence the design.

The product is in the minds of the designers and developers. When the persona process has worked, you can walk up to any designer or developer on the team and ask them who those personas are. They should have no trouble listing and describing them, just like they could list and describe the other members of their own team. And everyone should have the same list and descriptions. That's how you know you've succeeded.

The document (whether written in the first person or third person) is just a tool along the way to getting to the desired outcome, just like a wireframe doesn't make a design exist. Where teams fail is when they think they are done because they've finished the document. In fact, they've only really begun.

Hope that helps,

Jared

19 Jun 2011 - 10:01am
Byron Stewart
2008

 Hi Jared,

The process…

How would you say the role-play approach that Cooper used fits into the persona process? (It was unclear in Origin of Personas)

Is role-play used in developing the persona, or once the persona is complete you then role-play with it?

Would personas written in the first person,and used in role-play illicit empathy in a way that third person can't?

Thanks again for the feedback.

Byron

19 Jun 2011 - 11:24am
Jared M. Spool
2003

Role playing in the development of the personas is less necessary if you have met enough people behind the personas. Role playing is a technique for developing perspective and empathy, but I don't feel it's the only way to get there.

Once the personas are developed, role playing is a frequent technique for matching any proposed designs or treatments with the scenarios. "So, how would Mary achieve her objective of sharing her son's soccer game with her parents?"

As for first-person vs. third-person, I don't think it matters really. In my experience, the team members are intelligent enough to handle the dislocation. I can't imagine being with a team who says, "I didn't understand how to design for this user until we changed all the pronouns."

Frankly, I don't think you'll notice a difference in any measurable way between first- and third-person descriptions.

19 Jun 2011 - 12:03pm
Byron Stewart
2008

Hi Jared,

Can you please elaborate on the role-playing process?

  • Who should be the actors?
  • How do you cast roles? 
  • Is there and audience?
  • How do you prep for role work?
  • Could professional actors be used?
  • How do you know when you’re done and what does success look like?
  • How is role-play helpful in matching designs with scenarios?

 

 

I hear what you’re saying regarding the 1st person vs 3rd person perspectives, thanks.

Byron

 

19 Jun 2011 - 1:45pm
Jared M. Spool
2003

Role playing with personas isn't a formal activity. It's done in the moment, as you're working through a design solution.

The point of role play is to put the designer/developer into the shoes and context of the user, to see if the design they're imagining works and what the boundaries and failure points are.

You're done when you have a design that satisfies the needs of the users, as far as you can tell. (To ensure you did it correctly, you probably want to prototype it up and try it with the real users.)

I don't think using professional actors would be valuable because then the designer/developer isn't getting first-hand experience.

Jared

19 Jun 2011 - 4:06pm
Eva Miller
2009

I think personas are naturally third-person because you are trying to talk about clusters of behavior that are not your own. That's the point. Mostly, we are not the user, which is why you set out to make personas in the first place.

The challenge after that is (as others have said) to continue driving design decisions back to these "people." I actually find it useful to be able to say, "But this isn't about what you or I think is cool. It's about her. It's about him."

Feels natural to me. Third-person has never been an issue. Far more difficult is keeping them alive in the design process, and just changing the voice you write them in won't do that.

Eva

21 Jun 2011 - 11:06pm
ambroselittle
2008

I've always felt that first person persona docs sound corny and fake. I see the docs as a kind of storytelling, the persons who did the research are telling you about this "person." When you tell someone about someone else, you don't put yourself in the first person.

-ambrose

22 Jun 2011 - 9:05am
Joel Eden
2006

I agree. One subtle, yet important piece to this is that through observational methods we notice behaviors, actions, etc that the person being observed would never notice themselves, and could never be aware of. We all know this...that self-reported behavior is very different than observed/studied behavior due to many factors.

Therefore, I think it might feel strange to have a first-person persona description of behavior that that person wouldn't be aware of, due to subconscious behavior, etc.

As Ambrose said, it's a third party telling of the results of observations, and other methods. We want to use these methods to build empathy, and a third-party telling may allow for more natural telling of observations that will lead to empathy more than just the person's perspective would.

Joel

On Wed, Jun 22, 2011 at 12:44 AM, ambroselittle wrote: > I've always felt that first person persona docs sound corny and fake. I see > the docs as a kind of storytelling, the persons who did the research are > telling you about this "person." When you tell someone about someone else, > you don't put yourself in the first person. > > -ambrose > >

29 Jun 2011 - 11:53am
Don Dunbar
2009

"the person being observed would never notice themselves, and could never be aware of. We all know this...that self-reported behavior is very different than observed/studied behavior due to many factors."

That's is very true, but often overlooked is that the observed behavior can be just as inaccurate as the self-observed behavior. Because we notice something in another person's behavior, that they don't seem to notice, does not mean we are correct and they incorrect. Our observation also has the bias of our own experience. 

29 Jun 2011 - 12:58pm
Jared M. Spool
2003

With all due respect, I think this argument of bias is silly.

I will take a thoughtful, well-research, highly biased persona over a made-up, yet somehow un-biased persona any day.

I'm going to ignore the issue of bias for a long time, because that's not what is preventing good designs from being built today.

Hell, just to be ornery, I'm going to write all of my personas descriptions in the SECOND person for the next few weeks.

Jared

29 Jun 2011 - 1:59pm
Don Dunbar
2009

 

Jared, that is fine. But, being ornery doesn't prove your point. It ignores what in science is considered an important aspect of finding evidence for an argument.  

What I am saying, Jared, is that you might have a point. But, being ornery neither proves it or puts you in a good light. It suggests that you have no actual proof to back up your point and instead resort to ridicule to draw attention away from the discussion. 

But I will say that ignoring bias just makes you more prone to it. I am not saying make a big deal of it. Just to be aware that it colors our interpretation of users behavior. 

30 Jun 2011 - 7:42am
Jared M. Spool
2003

Don,

I'm not sure what point you think I'm trying to make, especially since you think science is required to back it up.

If you tell me what point you'd like me to back up with science, I'll try to find some to satisfy your needs.

In the meantime, orneriness has a big place in design and UX. Most UX people I know are driven by a stubborness that pushes them to find better solutions than the status quo. I'd suggest that the reason we're still discussing this point is due to a certain ornery quality that you're putting on the table. So, don't knock the ornery. It pays the bills. :)

As for being in a good light, I don't really pay attention to my luminence. I've done substantial research on how teams use personas in the last 10 years. I've published quite a bit on the subject, but have a lot more data beyond what we've released. (One day, we'll come up with a way to publish data as quickly as we collect and analyze it. In the meantime, we just stockpile all this awesome knowledge we struggle to get into the world.)

I let our work do the work of providing my light. Earlier in this thread, I published what I thought to be the relevant findings we've had on the subject. Your persistence on this bias thing feels to ignore it. Thus, my unlit ornery response. I could provide more data, if we have anything relevant in our stocks, but I don't really know what you're trying to prove at this point.

What I think my point is is quite simple: Most persona projects fail. Most fail because of a misunderstanding of the process and the deliverables. Very few fail because of the way the persona description is created. (By "very few", I mean none, but our data is limited, so I'll leave in a small margin of error.) That means that bias in the description is not an issue that will swing an unsuccessful project to become a successful project.

I don't disagree that our biases color our interpretation of our users' behaviors. In fact, I know that to be true. I have a lot of evidence from our research to back it up. I'm saying that said bias isn't an important factor in project success.

If your goal is to build better products/services by having better tools, like personas, then I wouldn't focus on bias in the descriptions. I'd focus on the research component and the method with which the team gets the personas into the working design and development process. Making perfectly unbiased personas won't move the needle one bit on the success scale, in my opinion.

That's my heavily informed ornery opinion. I'm sure it won't convince you, but until you tell me what data will, I'm going to stick with just saying that much.

Jared

30 Jun 2011 - 10:50am
Don Dunbar
2009

Hi Jared, what I was getting at is what you addressed in your response. It wasn't your point, but the delivery of it. From your ornery post, I could not actual tell what your premises for dismissal were. From your reply I can understand your point and see what backs it up.  I use science as an example and place of reference because we often use the findings from science to guide and inform our work in design. Design isn't just the science, but it does inform our work heavily. That's the one reason I use it. And another is because that's how my research is conducted ( I do academic research in interaction/interface design for information visualization systems).

Your point about my persistence on bringing up bias. I wasn't so much re-bringing it up, but addressing other's direct comments.  My real concern was just to address the problem of getting so caught up in making a persona seem real, that we forget that they are interpretations based on our own experience.  I don't think that is a problem with the document, but I do see the potential for a problem when designing for a user who we may not know as well as we think. I don't mean to suggest we aim for perfectly unbiased personas ( we couldn't get those anyway). I just mean to keep in mind that there is bias. I'm sure that most if not all problems are, as you say, in the development process rather than the persona development. 

I do know what you mean about  bias in the document not making or breaking the success. I think that is true and it is fine. It is just that, for those of us in the academic research side rather than the business research side, we do have to take, at least some, account of bias. Or at least we need to acknowledge that it may affect our findings.  We don't have the choice to just ignore it. By the way, I learned to do research from Cognitive Science and HCI classes, where acknowledging bias was a big deal.   We would spend quite a long time discussing which biases might be influencing our experiment. 

   "That's my heavily informed ornery opinion. I'm sure it won't convince you, but until you tell me what data will, I'm going to stick with just saying that much."

    Actually, your heavily informed ornery opinion is much more convincing than your flippantly ornery opinion. Since I don't actually know you outside of your work and this forum, I read your response quite differently than you intended it. It came off very high minded and dismissive, but with out any substance. 


 

30 Jun 2011 - 1:05pm
William Hudson
2009

Apologies if this has already been mentioned - it's been a long thread and I've been on holiday for part of it.

I agree that it probably doesn't really matter whether the personas are first, second or third person, BUT one of the key issues we have in interaction design is that people mistakenly believe that they are the user. Writing in first person does not help in this respect. Choose third (or second if you like). BTW, there is a recent, interesting paper on a study apparently demonstrating that people are more creative when thinking in the third person (as it were). See http://psp.sagepub.com/content/37/4/492

Regards,

William Hudson Syntagm Ltd Design for Usability UK 01235-522859 World +44-1235-522859 US Toll Free 1-866-SYNTAGM mailto:william.hudson@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.

30 Jun 2011 - 1:24pm
Jared M. Spool
2003

I originally joked about writing them in the second person.

However, if the goal of writing them in the first person was to get the reader to relate to the persona, then it seems to me the second person would have even a better chance of achieving the goal, since it would be direction for role playing:

"You're name is Mary. You run a small accounting business with 400 customers, primarily focusing on individual and joint tax returns and estate planning. Every year, the bulk of your business happens between December and April, as most people file before the the IRS deadline..."

Yet, as I stated before, I don't think the voice is important to the success of personas in today's design world. I think most teams have members who are smart enough to take the descriptions, no matter the voice, and do what they need to do. If they are not succeeding today, changing the voice from third to first (or second), won't make them any more successful.

Jared

2 Jul 2011 - 9:59am
Byron Stewart
2008

Hi Jared,

Do you use role-play/improv  in your persona work ,if so how?

Do you use role-play in any way as you develop personas?

Or, is it used after personas have been written?

 

Great discussion thanks for all your feedback!

Thanks

Byron

 

2 Jul 2011 - 11:34am
Jared M. Spool
2003

Hi Byron,

Do you want a different answer than the one I gave back on June 19, when you asked the same question? 

http://www.ixda.org/node/30277#comment-82313

Jared

2 Jul 2011 - 12:25pm
Byron Stewart
2008

Hi Jared,

I wasn't sure from your answer if you used role-play in your practice.

And. if you've used role-play or imrov as a tool while writing personas?

Stepping "into the shoes"  to understand user motivations let's say.

Do you use Cooper, Grudin & Pruitt, or Ad Hoc  based personas?

Thanks

Byron

 

4 Jul 2011 - 3:24pm
Jared M. Spool
2003

Hi Byron,

We use role playing during the design process. We don't use it during the persona creation process.

If you can tell me the difference between Cooper and Grudin & Pruitt personas, I can tell you which one we do. I'm very familiar with those folks work, but didn't know there were differences in the persona approach.

Ad Hoc Personas aren't really personas, per se. (They are unfortunately named.) We do something like them, but not when we're developing personas. We use the technique to identify what the current team knows and where further research might be warranted.

Jared

4 Jul 2011 - 5:42pm
Byron Stewart
2008

Hi Jared,

Here is an excerpt from Pruitt and Grudin's, "Personas: Practice and Theory" that begins to describe the difference between the two approaches.

"Cooper's approach can be effective, but our use of Personas diverges in several ways. He emphasizes an "initial investigating phase" and downplays ongoing data collection and usability engineering...Cooper emphasizes communicating the design and its rational among designers and their clients...We have extended this, using Personas to communicate a broader range of information to more people: designers, developers, testers, writers…and others."

Pruitt and Adlin suggest using professional actors to stage role-playing activities in which the actors play the personas. The actors would be used for persona kickoff meetings, spec or mock-up reviews. What are your thoughts on this approach? 

I’m also exploring the use of theatre techniques in the development of personas. Theatre encourages the personas to ‘come to life’ in the designer’s mind. Through bodily and Method acting exercises, designers achieve a greater empathy with users.

 

Thanks for all the great feedback

Byron

 

4 Jul 2011 - 5:58pm
Jared M. Spool
2003

In our research of teams using personas, I haven't seen anybody who claims they either use Cooper or Pruitt/Adlin personas. Neither is a rigorous approach, so anyone following their lead would likely introduce lots of their own variations.

Keep in mind that the hardest part of making personas be successful is cultural, so no textbook approach will have all the answers.

As for role playing and theatre techniques, it sounds like what you're talking about is Gamestorming (http://www.amazon.com/exec/obidos/ASIN/0596804172/?tag=userinterface-20). While personas are part of that, there's a lot that goes beyond it.

I haven't seen anyone who used professional actors for personas development or communication. Feels to me to be a bit excessive and probably not very effective, since it separates the designers from the personas.

Jared

4 Jul 2011 - 6:42pm
Byron Stewart
2008

Yes, there are many variations out there...

How are they not rigorous approaches?

I'm very familiar with the Gamestorming book.  Bodystorming is covered in the book and I use it as a tool often,

The use of actors as personas is meant to be a communication tool to achieve buy-in for the use of personas.  It's a way of introducing the newly developed personas into the company/culture and to key stakeholders. Not to replace the designers in the connection process.  I think that it makes for a memorable and impactful experience,

Do you think using actors as personas to aid in the buy-in process would also be excessive?  As you know, getting people to use personas is the big challenge.

Thanks,

Byron

 

1 Jul 2011 - 10:05am
William Hudson
2009

Apologies if this has already been mentioned - it's been a long thread and I've been on holiday for part of it.

I agree that it probably doesn't really matter whether the personas are first, second or third person, BUT one of the key issues we have in interaction design is that people mistakenly believe that they are the user. Writing in first person does not help in this respect. Choose third (or second if you like). BTW, there is a recent, interesting paper on a study apparently demonstrating that people are more creative when thinking in the third person (as it were). See

1 Jul 2011 - 8:05pm
whitneyq
2010

Heck, I use them all, at various times.

I usually write the description in third person.

I include quotes "from" the person in first person.

When I'm prepping people for an exercise where they are assigned the persona, I might address them in 2nd person to help them get into the role. then, I want them to speak in first person as the persona.

If I use them in a presentation or report showing differences in how the personas might interact with or react to something, I do that in first person as well, usually.

I guess the point is that it depends how you are using the persona. And that the key is to use them, not just "write" them.

W

2 Jul 2011 - 10:07am
Byron Stewart
2008

Hi W.

Can you please explain how the persona exercise you mentioned works - roles are assigned?

What's the process, what are your goals?

Does it work and, why?

Have you ever assigned roles during the development of personas?

Thanks for your feedback.

Byron

 

29 Jun 2011 - 3:05pm
Scott McDaniel
2007

That's just the sort of thing your persona would do.

Scott

30 Jun 2011 - 9:05am
Louise Hewitt
2010

I'm going to write them in Latin.

And hey! I had to look up 'ornery'. Whatever happened to 'plain language'.

There is a little point here that's been bugging me for a while - and it has nothing to do with personas.

As the field is getting more bloated, and more practioners are competing for attention/supremecy/contracts - we do seem (as a field) to be becoming a little, how can I say, wanky. (for any non-english listers, 'wanky: pretentious, self-important, stylised, all mouth and no trousers kinda thing).

If I remember rightly, and to be fair, I'm a bit old now, one of the things UX was all about was cutting through all the pseudo-science-techspeak-marketingflannel-bullshit and making things simple, transparent and accessible to all.

Not what I'm seeing on this thread.

Persona. definition: an imaginary person based on research from which you can hypothesise (sorry - guess) the likely reasons why and ways which a individual will use a system. purpose: To help people building systems to make them easier to use by / more attractive to / more relevant to etc. the desired or likely audience.

If it does this, then it's good. If it doesn't then it's not. Lets not get all 'wanky' about it ;D

Lou.

On 29 Jun 2011, at 20:04, Jared M. Spool wrote:

> With all due respect, I think this argument of bias is silly. > > I will take a thoughtful, well-research, highly biased persona over a made-up, yet somehow un-biased persona any day. > > I'm going to ignore the issue of bias for a long time, because that's not what is preventing good designs from being built today. > > Hell, just to be ornery, I'm going to write all of my personas descriptions in the SECOND person for the next few weeks. > > Jared > >

30 Jun 2011 - 11:05am
monkeyshine
2010

I think the message here is for us all not to be wankers. ;) I'll put my neck out there and agree...lots of verbosity that sounds a lot like loving to hear ourselves sound smart. I thought Jared's point was consistently clear...if your research isn't solid, bias,first or third person...these things are the least of your worries. Period. ...but what about ornery? Language doesn't get much plainer than that. :)-

1 Jul 2011 - 6:05am
Louise Hewitt
2010

Guess I'm still a blonde after all these years eh?

[reaches for dictionary...] On 30 Jun 2011, at 18:20, monkeyshine wrote:

> I think the message here is for us all not to be wankers. ;) > I'll put my neck out there and agree...lots of verbosity that sounds a lot like loving to hear ourselves sound smart. I thought Jared's point was consistently clear...if your research isn't solid, bias,first or third person...these things are the least of your worries. Period. > ...but what about ornery? Language doesn't get much plainer than that. :)- > > On Thu, Jun 30, 2011 at 10:48 AM, Louise Hewitt wrote: > >> I'm going to write them in Latin. >> >> And hey! I had to look up 'ornery'. Whatever happened to 'plain language'. >> >> There is a little point here that's been bugging me for a while - and it has nothing to do with personas. >> >> As the field is getting more bloated, and more practioners are competing for attention/supremecy/contracts - we do seem (as a field) to be becoming a little, how can I say, wanky. (for any non-english listers, 'wanky: pretentious, self-important, stylised, all mouth and no trousers kinda thing). >> >> If I remember rightly, and to be fair, I'm a bit old now, one of the things UX was all about was cutting through all the pseudo-science-techspeak-marketingflannel-bullshit and making things simple, transparent and accessible to all. >> >> Not what I'm seeing on this thread. >> >> Persona. definition: an imaginary person based on research from which you can hypothesise (sorry - guess) the likely reasons why and ways which a individual will use a system. purpose: To help people building systems to make them easier to use by / more attractive to / more relevant to etc. the desired or likely audience. >> >> If it does this, then it's good. If it doesn't then it's not. Lets not get all 'wanky' about it ;D >> >> Lou. >> >> On 29 Jun 2011, at 20:04, Jared M. Spool wrote: >> >> > With all due respect, I think this argument of bias is silly. >> > >> > I will take a thoughtful, well-research, highly biased persona over a made-up, yet somehow un-biased persona any day. >> > >> > I'm going to ignore the issue of bias for a long time, because that's not what is preventing good designs from being built today. >> > >> > Hell, just to be ornery, I'm going to write all of my personas descriptions in the SECOND person for the next few weeks. >> > >> > Jared >> > >> > >> >

30 Jun 2011 - 11:47am
cambeck
2010

 It is impossible to remove bias in personas. They are inherently biased. They have to be. Otherwise we're choosing them by random selection, which doesn't help our clients or our bosses.

This has nothing to do with if they were written in the first, second, or third person, but rather because we had to choose, rightly or wrongly, a certain person to portray from a very diverse, complicated audience. No one is an everyman. Neither should our personas be.

We may have had good justification for portraying one person instead of another (and we'd better), but the very fact that we've conducted research (assuming we have, which isn't always a given) colors us to the point that it's impossible to forget knowing what we now believe to be true -- that choosing to write about THIS person is a better decision than choosing to write about THAT person (See Knowledge, Curse of, "Made to Stick" by Chip and Dan Heath).

More important questions pertain to the presence or absense of research in building the personas, and relative bias of that research. It is impossible to remove all bias from all research -- but as practitioners, it's important to recognize the bias that's important to remove from our studies.

I may be only reiterating what others have said. But this has been a good discussion.

4 Jul 2011 - 7:04pm
Jared M. Spool
2003

Byron,

Lack of rigor: It's hard to find two teams that have used the same approach to produce their personas. Even amongst those teams with successful persona use, the methods of creation vary wildly.

Actors for selling the personas: The fact is, if after you've created the personas, people aren't convinced they are valuable, no amount of selling (or paid actors) will change their mind. The best way to convince people to use personas is to have them involved in the creation process directly.

Jared

Syndicate content Get the feed