The following is a transcript of a live interview with Persona book authors John Pruitt and Tamara Adlin, conducted by Frank Spillers in October 2006.
This seminar is available with the slides and audio presentation here (requires brief registration, privacy protected).
Transcript of Personas LIVE!:
This interview follows the publication of Adlin and Pruitt's
personas book, "The Persona Lifecycle : Keeping People in Mind
Throughout Product Design"- published earlier this year...
Frank
Spillers Welcome...
For those of you who have joined us today for Personas Live – an interview with
Personas book authors John Pruitt Pruitt and Tamara Adlin Adlin who we have
live on the phone here. I’m Frank Spillers Spillers. This presentation, or
seminar, is aimed at understanding a little bit more about what went into the
publication and writing of the book. It’s a unique opportunity to hear from the
author’s themselves, John Pruitt and Tamara Adlin. We’ve got something exciting
stuff for you here, so let’s look at the agenda for today. So first up is the
Persona Lifecycle which is one of the major methodologies or themes that the
book wraps itself around. We’re going to look at a couple of case studies. Then
we are going to look at some Persona best practices and discuss the issue of
using personas in a way that maximizes the organizational return on investment.
And then we are going to open it up for some questions and answers with the
authors. Before we begin let’s do a little brief introduction. Tamara Adlin
Adlin is formerly the Customer Experience Manager at Amazon.com and John Pruitt
Pruitt is currently with a Senior User Experience Lead at Microsoft with Vista, Tablet and other projects. Tamara Adlin, would
you like to tell us a little bit more about yourself?
Tamara
Adlin Yes.
About a year ago I left Amazon after helping to set up their customer
experience group for their Platform services, allowing lots of other online
retailers to use the Amazon technology behind their site. I then started my own
consulting company called Adlin Inc. I do
a lot of customer experience consulting for online retailers, other online
businesses and also traditional product design.
Frank
Spillers Okay
thanks. John Pruitt, give us a little bio.
John
Pruitt Yes. I’ve been at Microsoft for a little over eight years
now, mostly with the Windows team but I’ve been about three years with MSN,
working on MSN Explore and a variety of other web products over there. I’ve been
with the tablet and, mobile PC division for a little over a year and a half,
closing in on two years now. We are working on getting the mobility side of Windows
Vista out the door as well as some new products like ultra mobile PC’s and some
new stuff that’s coming down the pipe.
Frank
Spillers And
I’m Frank Spillers Spillers and I help run the user experience and usability consultancy
firm Experience Dynamics. I’ve
been working with Personas for about 9 years now. It’s a pleasure to have you
both with us today and thank you so much for joining us. I just wanted to let
our audience know, that this is actually a re-run and we are recording this
session of the quick release we did of a similar presentation. And there is a
little bit more content in today’s session. That had an overwhelming response
so we decided to re-it again here today. Just a few weeks after the original
publication is where this session first came together. The reason that we
wanted to bring the authors on and talk about this book is because I think it
is a major landmark in the usability and user experience community. For the
first time we now have one place that we can refer to that is really
interesting and solid and useful information. If this book isn’t on everyone’s
bookshelf in the next few months then I really would wonder what they were
doing, because I think this is just a valuable asset. So, thank you so much for
putting this book together. Why don’t you start by telling us a little bit
about the structure of the book. How is
it laid out and why?
Tamara
Adlin Well, the first thing I want to say
is thanks for saying that. When we originally started off to do this book, our
idea was to put together a quick practical bag that was going to be a couple of
hundred pages and our intent was certainly to do that. The when we started diving
in we realized that there was so much to say and so many stories to get from so
many practitioners – that’s a big part of the book by the way – stories from practitioners
who are inserted all the way through as case studies. And the whole thing just
grew and grew till eventually it became the heavy book that it is today. But it
was also our purpose to make it a rich report, a book that you could thumb
through and find ideas, no matter where you are in the persona process.
John
Pruitt As Tamara Adlin said, our book really started with case
studies. We had organized a couple of workshops at the UPA (Usability
Professionals Association) back in 2001 and 2002. At those workshops we had
brought together persona practitioners from around the industry. And at the end
of those full day workshops the general consensus was that we need a book on
personas about the methodology, on how to do it well, and how to use them
right, with case studies and best practices. So Tamara Adlin and I and another
contributor Holly Jamieson decided we should do a book. But we thought it
should be a really small book, about a hundred pages.
Tamara
Adlin And we thought we could do it in
nine months…..
John Pruitt Yeah we thought it would take about nine months, but it just
grew and grew.
Frank
Spillers So
it turned out to have around 725 pages or so of content in here. Why should
someone buy this book, and put another 700 or so pages on their reading list?
Tamara
Adlin Well, that’s actually a big challenge
and I’m glad you brought that up. This isn’t a book that we intend for you to
sit down and read cover to cover. The Lifecycle, which you’ll hear a little bit
more about in a minute, came out of the original workshop which we already
talked about. We had a lot of practitioners there, and we looked at a lot of
their methodologies and we talked about it. And after the first workshop we
realized that personas seem to have a lifecycle just like regular people. And
they go through a fairly predictable set of phases, when you try to create them
and integrate them into a company. And when you look it at through a persona
lifecycle model, you notice that personas tend to run into snags or failures at
a fairly predictable time too. So we created the book around a lifecycle
approach so that could we could talk about how to integrate personas in with other
methodologies that are out there. We certainly don’t think they replace other methods
but rather they augment it. Also, so that we could talk about how this process integrates
with other development lifecycles. So, like this slide says, we give the intro
and overview and the background on where the personas idea came from and how our
lifecycle is structured. Chapters 3 through 7 are each on one of the phases of
the persona lifecycle. Chapters 8 through 12 are invited chapters that we
thought would introduce interesting ideas, research and also methods to the
body of the book. So it is designed to flip through, get a sense of what’s going
on, and maybe read a chapter at a time as you hit each stage in the lifecycle.
And as you hit any snags, as you most likely will, because everyone does, you
can flip through and get ideas, find out how other people have gotten through
it, get inspiration and get tools because we really wanted it to be practical.
And really the only way to practically handle a 700 page book is to use the
pieces as necessary and that’s hopefully what you’ll do.
Frank
Spillers And
you also have these sidebars which are really interesting. I had the privilege
of contributing three or four of these sidebars to your book, Why the
practitioner involvement? Was that just to pad it out? Why are there so many
stories? Practically every page you turn over there are stories from the field.
What’s the purpose of that?
John Pruitt I
think that came out of the fact that we started with workshops and with other
practitioners. The more we talked to other practitioners we realized that the
persona method isn’t and probably couldn’t be just a single approach. Like user
centered design generally, you really have got to fit the method to your
organization, your particular product space, or service space, like what are
you creating, building or selling. We personally found a lot of value in the
stories and best practices, and the case studies that people were bringing to
us, and we wanted to bring those out in the book as well and really highlight
them. I think that’s really the value of the book. To in and read these, in
some cases really detailed case studies about how they applied personas, how
they conceived of personas and what they did for their group. I mean they
literally are almost one per page. We have hundreds of little side bars through
the book.
Frank
Spillers So
the book almost reflects the state of the practitioner community in terms of
how everybody is trying to work together to try and figure out how best to
utilize personas and the user research in their user centered design and their usability
efforts. I need to say I think they are absolutely fantastic and if you haven’t
looked at these sidebars and looked at the book – it’s a really nice design and
a really nicely organization that compliments the state of the other
practitioner community. I just want to say for those of you who are listening
or are just joining us, we are speaking with John Pruitt Pruitt and Tamara
Adlin Adlin the authors of the “Persona
Lifecycle Handbook” just released a few months ago. If you are interested
in asking a question go ahead and “chat it” in the left-hand window. Those of
you who pick up the recording won’t be able to do that obviously. But for those
who are live on the call now, go ahead and chat that in there and I’ll pick up those
questions and field them as we go along. So let’s look at this lifecycle. You
know the more I think about this, this is more of a book that’s about helping
you and your organization use personas strategically and operationally. It’s
not really about writing a persona, although that is in there, the use of the
actual tactical technique. But the beauty of the book which I think was deliberate
on your part, was to have that organizational perspective and of how to use it
in your company of where your at.
Tamara
Adlin Absolutely.
We just found that there was so much information about persona projects hitting
these huge snags. We were in this lucky position of having an overview of so
many stories from other practitioners that we could start to see a pattern. And
that’s really how we started to organize the book. Everybody has become really
excited about personas, since Alan Cooper wrote “The
Inmates Are Running the Asylum” back in 1999. And the problem with that
book, in the sense that everybody got excited about it, but nobody knew how to
do this method. And that’s fine, no critique of Cooper on that. But people did
get excited about it, but nobody knew how to do it. And when they tried it they
would run into problems. And I think that is because personas are asking an
entire organization to change in some fairly significant ways. They are asking
organizations to look more closely and more carefully at user data way earlier
in the process than most organizations are used to. They are asking for
customer focus that’s really walking the walk and not just talking the talk.
They are asking them to think about users every single time we have a meeting
about this product. And if you don’t address the way that you are asking for
change, then the effort is going to fail because the inertia of old methods and
the typical lifecycles are just too strong.
Frank
Spillers Sure,
that makes a lot of sense. I actually brought a slide up here since you
mentioned the book “The Inmates are Running the Asylum”. And a lot of people
attribute personas to Cooper here, in terms of the starting point in the modern
use of personas in 1998 and 1999. So people were left hanging, and that was one
of the criticisms of the book. Your book fills the gap, which people have been
hungry for over the last four or five years it seems.
Tamara
Adlin We were certainly hoping so, yes.
Frank
Spillers Let’s
talk about that lifecycle then. What is the persona lifecycle?
John
Pruitt Well, the persona lifecycle is five stages that we related
to the human procreation process. The initial stage, which we call “Family planning”,
is really about organizational introspection. It is doing what we do as
user-centered design people, kind of turned on our own organization. What do we
do as a process? How do we push users into the center of our design effort? Do
we already do that? Do we do user testing? What other kind of user research do
we do? How do we design our products and decide on features and that sort of
thing? So, it’s looking at ourselves and then doing some analysis of “Does this
persona methodology fit, and how can I best make it fit” So, it’s a planning
process. The second part of it is also going out, and collecting the data that
we sometimes already have, or generating a new data collection effort to inform
what personas we need to create. That is to really provide the meat of persona
creation. The second stage is conception and gestation and that is exactly what
it sounds like. It is the phase where you create your personas out of data, and
in some cases you create them out of educated guesses and assumptions. And then
you go through a process of maturing those initial personas into something
that’s really substantial and substantiated. The third stage is called birth
and maturation and that’s really about getting your personas out and into your
team, so that they become alive and your team recognizes who they are. It’s
also introducing the persona methodology as well as kicking off a communication
campaign that focuses on your personas as a progressive disclosure of
information and regular updated basis so that personas don’t just get
introduced and then die on the vine. The fourth stage is called adulthood and
that’s really when the personas themselves are mature. They are beginning to be
known on the team, and they have got to have a job to do. They are adults now,
and they need to contribute to your organizational society. So it’s about
getting personas to help you to design activities and product decision making,
and through the development stage. So it’s in the planning part at the
beginning and the part in the middle as well as at the release part. So that
when your product is going out to market, your personas have something to do
there as well for you.
Frank
Spillers So are we talking at a project level that all this is
happening? Or are we talking at an organizational level over a period of years.
So, is this months or years, or is it both?
John
Pruitt It’s actually both.
Initially you need to think about the project that you are doing personas for.
Is it a product or service, a software product, a hard product or a goods
product. In some cases as in web products, that may be a six month process. For
Windows Vista that was a five year development cycle. So adulthood is about
applying personas through that entire development process. If you are fortunate
enough to do it, and have the bandwidth and or support from your upper
management to think about personas and user-centered design across your entire
organization, so not just a specific project or a single product but groups,
suites or families of products. This applies to that as well. The final stage
is called lifetime achievement and retirement and that it really about when
your product has gone to market and you want to recognize what the personas
have done for you, if anything. It’s about ROI – return on investment. It’s
about really being accountable. We invested time and effort into personas. Did
they deliver something? And if so, what is that? So that chapter of the book
really talks about different kinds of ROI and how do you measure what personas
have done for you. And then the retirement phase is about starting the cycle
all over. So if you were going to do the next version of your product, or if
you were going to go into a related project area or starting something entirely
new, do your personas live on, do you retire them, do you reincarnate them, or
do you reuse some of the data. If you look at the persona lifecycle diagram it
is kind of circular, but it’s actually spiral in shape. That’s because the
persona lifecycle doesn’t always turn a full circle. It actually continues to
grow and change and your personas will evolve over time. That’s how you think about
your target customers.
Frank
Spillers I wanted to ask Tamara Adlin what is the deal with these
fears and these circles – can you clarify that for us?
Tamara
Adlin Sure.
I want to add one quick comment. I think that two of the places that
practitioners will find the most initial value especially those who have tried
personas already, are in the conception and gestation phase and the adulthood
phase. There are lots of new ideas in all of these phases but the thing we’ve
gotten the most positive feedback on is conception and gestation which shows
really talks in practical ways, step by step, how to get from raw data to personas. And adulthood,
where I would say 90% of the personas efforts fail, people put posters up on
the wall, and then expect them to magically work. But in adulthood, again, it’s
very practical and it tells you how to actually use them in an everyday
environments. Going back to the spiral, we thought long and hard about this
graphic. We wanted each of the circles to give you a sense of how much time and
effort, relatively speaking, you are going to spend in each phase and a sense
of how they morph into each other through the transparency. And the cyclical
nature of it. We wanted something that was very eye-catching and sort of
captured the idea in an unusual way. And that’s how we came up with this. And
you should have seen all the drawings we went through to get here. So, that’s
where that came from.
Frank
Spillers Okay. I want to ask you a question about retirement and
lifetime achievement. That’s the one that when I first saw it I wasn’t sure
what that meant. I thought of the beanie baby, where it gets retired or thrown
away. Do you take your user profiles and put them up on the shelf when you are
done with them? Or pull the poster down, or whatever the case may be. How does
that play out with a product that has a product for say one target customer that
they are working with over a number of years, say five or ten years. Do you
just keep refreshing the persona data or do you literally throw them away and
start all over again and then its time for another product to be launched at
that target audience?
Tamara
Adlin I
love the analogy to beanie babies. No one has ever said that before and I think
it’s great. Let’s pull that analogy a little further. Maybe you grow out of
your little lion beanie baby. Maybe your company has launched a particular
product and now they are moving on to something new, which is related to the
old product, but it’s not exactly the same. Or maybe they have been working on
the same products for many years. Or maybe they are working on B2 of a product.
And so that’s analogous to maybe you grow out of loving your beanie baby by
just putting it on a shelf. Or maybe you just stay in love with it like the
Velveteen Rabbit so it becomes real. So anyway, to get more concrete about this
– its retirement, reuse or reincarnation. Either you are moving on to a product
that’s not related to the first product that you created or you if you have
made the personas come alive in the minds of your team you have to figure out
how to undo that to be ready for the next set of personas. On the other hand,
maybe you are going to do V2 or V3 or Version X of your product and you find
that the persona has evolved a little bit. You’ve released the first version
and now Mary didn’t know anything about online banking, and now maybe she knows
a little bit about online banking and so now you are going to have to look a
little bit at the data and how the world had changed and maybe she might have
evolved a little bit . Reincarnate means well maybe it’s not Mary anymore, but
maybe the data you used to create Mary might go into Meredith. So, every single
time, what we suggest is to figure out what you need out of personas for the
next round, how much you are going to have to introduce, new or revived
personas. But at the heart of the effort you have to go through and look back at
your raw data again and determine if it’s still relevant and how you are going
to use that in the persona effort
Frank
Spillers Okay. Let’s talk about the idea of personas. If you are not
using the word personas you might be using user profiles or something else. Is
that okay? There’s been a lot of buzz and what seems like a lot of hype over
the last few years, less so now – but around 2002-2004 it seemed like everybody
had personas, and everybody had a profile of their customers. Tell us a little more
about the terminology of using personas.
John
Pruitt Sure. The concept of
personas has been around for a long time. Longer than Cooper’s book which came
out in 1999. Cooper re-conceptualized what personas were and also coined the
term “personas” which popularized the idea. But abstractions of user’s
representations of individuals that are related to groups of people but are
caricatures of them have been around for a number of years. But really it began
way back in industrial design with a book by Henry Dreyfuss (I believe). And
this has gone on and been captured by various marketing folks in terms of
representations of user segments and market profiles, user profiles, customer
profiles in that regard. The flavor of what’s included in those representations
differs. So for marketing purposes, not all of these include photos and try to
make the customer feel extremely real, or realistic and have a personality. But
they all capture essential information about who your target audience is, so
that you can do a better job of decision-making with those users in mind.
Frank
Spillers So how about data-sources then. This is something in my
side-bar that I felt was a very important issue in terms of differentiating
where your data comes from when you fill the persona or user profile. Is it
enough to take your marketing demographics for example this is Charlie, he
makes $70-75K, he is 30 years old and likes to play with gadgets. Is that
enough? Or does that have to come from ethnographic studies, such as a day in the
life, two weeks, five week, five month studies with the target audience and getting
deep into their behavior.
Tamara
Adlin We
had really rich conversations about this over the years that we worked together
on this project. In fact, one of the first debates that John Pruitt and I had
was what data you actually needed for personas. So the way I would answer this
is that it’s always best to start to create your personas out of richly varied
data. In the best case scenario you have ethnographic research, you have market
research. You have research from the US
government on the behaviors of people similar to the people that you are
targeting. You have a huge range of information. However, what I think we found
is that the focus that personas can give you is so valuable in an organization that
even if you base your personas on raw assumptions, or create what Don Norman
effectively called “ad-hoc personas”. Which at the very least get everybody
aligned in their assumptions about the users. If you asked around in your
organizations right now, for people to describe the user that you are
targeting, you would be surprised at how much variety you would get. So even if
you just align the assumptions in your organization, personas can be incredibly
effective because the focus outweighs any of the risks of them not being
“correct”. However, the more data you can use, certainly the better. Certainly
for buy in, and certainly for comfort in the entire organization in focusing on
the personas.
John
Pruitt I want to add something,
going back to your original question there as well. Part of what Alan Cooper
really added to the notion of customer representations, is making them real and
giving them personality and substance. And in Alan’s case in particular,
there’s a big focus on making sure the personas have goals, aspirations and
motivations behind why they would be using your product, and what do they do in
life, and what matters to them. And I think it’s that human-ness that personas
have, that take them way far beyond other representations like market
segmentation and user classes of those kind of thing. There become very
powerful proxies for our customer targets.
Frank
Spillers So for example, the agency “Organic” has what they call a “Persona
Room” and they have developed a scenario of the person. They had a stage
designer come in and build a living room. I don’t know what they are trying to
test or trying to create, but the idea is to bring the design team in to
empathize and build out. What Levis (the jeans company) does
is conducts studies and recreates the closets of people. From their
ethnographic fields they will come back and rebuild the teenagers’ closet in
their offices, and use that as a tool. The criticism that some bloggers have
made about Organics personas room is that they never mention that their persona
recreations are built on real data. They are not gathered from any type of user
intercepts or interviews. Is it enough to emphasis like that, to just say we
are going to emphasize and we know our customers. Or do people actually need to
budget for field studies and spend the money to go out there and touch base and
make contact with their customers and really get to know them and their goals
and so forth?
Tamara
Adlin It
depends on how advanced the organization is. You’d be surprised how many
organizations out there are still doing very self-centered design. They are
just thinking about what they would want. If that’s the case, then anything
that could bounce you out of that self-centered design, at least to have a
little bit of empathy with your target user group is better than what you were
doing before. If a company really does say that they are doing user centered
design, they may or may not actually be doing what we call tree user centered
design where you really go out there and gather a lot of data. The purpose of personas
with respect to data: one way to start thinking about that is to make the data
useable itself within the organization. The job of developers is not to analyze
customer data. The job of developers is to develop. Our job as customer
centered design folks or user researchers is to do that research, and get it to
those developers in a way that will actually enhance their daily work and
inform it with data. That’s one way to look at personas. As a communication
technique. And of course, to your point: in some cases in some organizations
you really have to be able to say, hey, here is the raw data. You want to go
take a look at it, take a look at it, you’re more than welcome to. And we suggest
that you make the raw data available within your organization, say on an
intranet site. Once people do that once or twice, they tend to start trusting
the personas a lot more. They don’t really want to read all of the raw data.
They just want to know that you’ve read all the raw data and interpreted it
wisely in the persona.
Frank
Spillers Is there a credibility issue there John Pruitt, with regard
to where the data comes from say for example in a Microsoft type environment?
John
Pruitt Absolutely. So, when
you create assumption personas, which we think are extremely useful and
powerful, we also think you need to make sure it’s clear that these are
assumption personas. So, you always want to maintain a sense of credibility, in
terms of how far can we push these things and have our assumptions been validated
in any way. So, credibility for the personas is something that does come up and
we have a slide later on in this deck where we talk about some of the issues
with personas and maybe we’ll come back to that in a minute. One thing I do
want to add related to going out and visiting customers. A really powerful
thing that we have seen happen with personas is that once you have the personas
in place and the team starts to become familiar with them, then part of the
communication plan we think, should be going and finding out real
representatives of those personas. So if you have the Abby persona or the
Patrick persona then you recruit some of those for either field studies or lab
studies. You bring them in or you go out and visit them and you bring your team
along with you tell them “Okay today we are going to go and see two different
Abby’s. And one of the Abby’s is a man, surprisingly enough”. And they see
these instances of the persona in real life. Every time I have seen this happen
your team mates go away thinking “Man, that guy was such an Abby, I can’t
believe it”. It brings the persona all that much more into focus and really
helps people understand who the persona is, and what the essence of this
persona is.
Frank
Spillers So, it’s a little bit like a usability test where if you
bring in your team or your executives and have them see, hear and feel the user
struggle and succeed its much more powerful than a report or a graph.
John
Pruitt Absolutely. And in
fact what I’m suggesting here is that on top of that you call your participants
by the persona name that they represent. So we do that as we bring people into
lab studies. We say “Today we’ve got two Patricks’ and a Sandra and an Abby”.
And when we report our findings back we say things like “The Sandra’s had these
kinds of issues or reacted to the prototype in this way”. And it really is a
powerful thing.
Frank
Spillers So let’s talk about why personas work. With reference to
the naming of them, how effective have you found that in your own experience. I
want to share an anecdote before I do. On a recent
blog posting the former Microsoft blogger Robert Scoble on his Scobelizer blog was talking
about the ditching of the name whatever it was. The blog post was called
“Ethically Bankrupt Personas”. If you follow the interaction of the developer
and Scoble, basically they are getting hung up partly on the name and the
naming of it. They say “Let’s get rid of Eric or Bobby” or whoever it was. Do
you find that people get attached to names, faces and races? There was an issue
with Cooper that came up a few years ago and someone was asking “Should we
include different races?” What’s your personal experience around the
effectiveness or the confusion around that?
Tamara
Adlin I
think that’s actually really funny. Because all of the people on that blogpost
I looked at are using the name which means that the persona is actually
working. But the fact that they are saying “Let’s get rid of Mike” or whoever,
means that they all realize there’s a Mike and they know who he is. If they
want to get rid of him that’s one thing. And they know about Mike versus
Saundra versus Abby or whatever, but the alternative would then be saying “the
administrator” or the user.
John
Pruitt Or not talking about
users at all.
Tamara
Adlin That’s
right. So I love that. So if they don’t like the name, if they don’t resonate
with the details of the persona that’s one thing. But that’s a very different
level of conversation, much more detailed in terms of users and goals, than it
would have been if they had said “I don’t want to talk about the user” which
never would have happened. That kind of debate with developers is delicious.
Frank
Spillers Is that one of the outcomes of persona process is to get
that dialogue going about the users
Tamara
Adlin Right.
All of a sudden you are having that discussion about the users. I just think
that blog is one of the best arguments for personas that I have ever seen. The
fact that developers are saying lets get rid of this one. And there is a big
debate about stereotypes and about racial equity or gender equity in personas.
But I think the key is to remember that you’re trying to capture real human
goals, needs and behaviors. And if those goals, needs and behaviors vary
according to gender or race, I mean when you are talking about a system for
people to enter their password into a US immigration server, then yes. I think ethnicity, background and language
totally matter. If you are talking about some other type of software or some
other website, maybe they don’t matter at all, right? So, online banking, a US
online banking system, does gender really matter? I don’t know. I would doubt
it.
Frank
Spillers So since personas are grounded in behavior, they typically
tend to generalize based on behavioral attributes. And in the book there is
that separation. In fact early on in the book, one of the sidebars that I
contributed to was the distinction between a marketing persona and a design
persona.
Tamara
Adlin That’s
right. Marketing personas are for targeting. They are asking things like what
magazines does this person read? Where should we advertise? How should we let
them know about this new product? Which is very different than saying what
goals do people have when they actually use the product? And if you are doing
marketing then you need to know the age range, the gender, in many cases
perhaps ethnic background, depending on the product. That’s why marketing and
product design are two different animals. It’s a different set of needs.
Frank
Spillers When I get asked that question in persona trainings the way
I usually answer it around should we use races and different genders is that
generally personas are descriptions of behaviors that are extrapolated hopefully
out of real and concrete user observation in their settings. It’s the general
practice to have diversity and to have awareness of different races and genders
and abilities and to include those. I’ll even do that just as a way to remember
who I’m serving. One of my users is an international user for example. So I’ll
include someone who is not North American just as a reminder so that when I am
presenting the data, so that I can say “Oh yes and by the way the user in this
country, in China or in India, struggles particularly with these localization issues for example".
Tamara
Adlin Yes.
And in that case it doesn’t matter.
Frank
Spillers I think we have a question. Let’s take our question now.
Then we are going to move into case studies. I have just been putting up a few slides
there with a quote from Alan Cooper about designing for one person or audience.
There have been some really interesting questions on the Internet related to
“BaseCamp” the project management tool. Should you design your user needs or
should you just design to a general audience. Too much to get into in this
discussion though. So here’s a question: “Have personas been proven to be the
best design and/or marketing strategy tool versus more standard design
techniques which view the customer base as being more diverse than a single
persona and thus segmented accordingly?”. This is segmentation versus
behavioral slicing issue: “Are there concrete measurements of the success ROI for
implementing personas versus traditional design marketing approaches?” This is a
great question.
John
Pruitt Yes, great question.
So the answer is – are personas the best method? There is no data out there
that says they are the best method. In a certain way I wouldn’t ask a question
that way. The reason is because what personas are is a way of getting your team
focused on end users. I really think the best approach is to think of them in a
supplemental way to other user centered design work and other proven market
research efforts. We really believe that the best personas, start with good
market research and segmentation data. You then build personas from a variety
of additional data sources including qualitative field research and the graphic
type research to really get a good rounded view of what your end user is. So, describing
personas in that way is supplemental. They are not just a single method.
Because they can also be applied to a lot of different aspects of your
development process including early on decision making, some aspects of product
design including visual design and styling. They can influence and be a part of
the marketing and messaging. They can be part of documentation. So, because of
their multiple applications, it means their ROI can be measured in different
ways as a part of the product. So can they affect your bottom line in improving
sales of your product, or decrease support costs for your product and calls to
the product support line and those kinds of thing? Yes. And there are case
studies that have shown personas influencing bottom line dollar, support cost,
ease of use, and those kinds of things. They also have broader effects which
are harder to measure. Personas can act as a lever to get our catalysts to get
more user centered design activities into your organization overall. So it’s
like a foot in the door technique. You get personas in, and people resonate
with them, and they start discussing customers. They kind of like what they
bring. And then this gives you the leverage to say “Hey, let’s do some further
research or do some user testing. We’ve never done that before. Or let’s do a
focus group, or let’s do some user survey stuff”. We’ve seen lots of cases where personas
actually influence your entire organization towards a more user centered design
approach.
Frank
Spillers Yes. I suppose you have had a lot of people writing in as
well with all the contributions you’ve got in the book. There are literally
dozens of examples of that shared where you can see people are using them. I
think that point was really important and thanks for the clarification and
expanding on that John Pruitt. I get asked that question a lot myself. People
confuse my enthusiasm for personas for a trumping of market research. Its not
that personas are better. What you are seeing in the marketing world in the
last five years or so, is marketers exhausting their traditional market
research design techniques because we are trying to achieve standards of
usability, ease of use and user centered design that require usability and
behavioral research methods such as personas. And that’s really the point. If
you want usability, then don’t use marketing techniques like focus groups and
surveys alone. Add to those and expand on those. That’s where a lot of
marketers are getting very interested in persona methodologies and behavioral
research. This is a good point to talk about case studies, so let’s talk about
case studies for a minute. Can you talk about some of the more interesting ones?
Tamara
Adlin One other quick point before we
move onto that. Part of that last question should also be about are they
(personas) better for design. And I would say this real quickly. It’s almost impossible
to design a product for 35 – 55 year old women. I mean what design decisions
are you going to make that really have to do with navigation and clarity and consistency
across the user interface for a range. But for a specific person, if everybody
in the organizations with the thousands of tiny decisions that get made every
day are all focusing on the same person, then you are more likely to come out
with a product that makes sense from end to end and is internally consistent.
The value of that is enormous and it can not be achieved with ranges. So
traditional market data does not help with that kind of consistency.
Frank
Spillers Let’s talk
about bottom line stuff then and persona lifetime achievement or ROI. Can you
lead us into a couple of the case studies that are in the book that are most
interesting – the Medco and the Best Buy studies. As you
put this book put together, what have you heard and seen in terms of companies
profiting from and the evidence of ROI coming straight from the efforts of
personas? Tell us a little more about that……
Tamara
Adlin Well, like the slide says, the ROI
should be measured in terms of improvement to your bottom line. More people
should buy the product and use it and like it should cost less to support. But there are also improvements that are
worth an awful lot of money. That can happen in terms of your team and your
company and you internal process. So what we recommend is that during family
planning you identify what the problems are that you want to solve with your
persona effort. These could vary from organization to organization. And then
during the ROI phase you actually measure those and see what kinds of results
you have gotten. And now I can turn it over to John Pruitt to talk about the Medco
Health case study.
John Pruitt So with Medco
it was actually with a website that they had applied personas to www.medcohealth.com What they did with personas was redesign their
main site to navigation. What they found was a 33% increase in transactions,
they had less abandonment. They also had more prescriptions being ordered
online which is pretty incredible. Those are really solid numbers. Moving on to
the next slide. I like this example here as well because in this case it wasn’t
about building a software product, which is what I think about most of the
time. It wasn’t about websites but it was really about the physical layout of a
store. The fact that these guys did some really good market research and then
did some good ethnographic research and then built personas and then took them
really seriously and went out and redesigned a fairly good number of their
stores. And then they looked across these test stores compared to the rest of
their stores and compared what was happening with sales. The test stores were
showing an average of 7% more in sales. Those are some great numbers. It’s not
a 100% clean experimental design, but both of these case studies are good
examples of showing improvement in the bottom line of your product in places
where it impacts your dollars. We’ve got
other cases outlined in the book and I’ll just name a few that are interesting:
The dish maker “Pfaltzgraff” did some really great work with personas, not only
redesigning their website but also rethinking about their actual plate ware and
what they were doing as products. And also how they are presenting the website,
how they are presenting the catalogue and how they go about selling those
products. So that’s a really great case study.
There’s a
good one in the book about a game maker company in
that went from redesigning their website using personas from which they
basically went from having zero sales of their game products through their own
website. When they did the redesign, they discovered through their customer
data, that who they thought their real customers were, was in fact way-off.
They discovered that their real customers were mostly women who were playing
these casual games. So they redesigned
with those personas in mind. They created two personas with the redesign in
mind and saw an incredible rise in the magnitude of downloads of their product
and eventual sales of their games to these customers. This is a really powerful example of putting
your user first and really thinking about why would they come here, what are
they going to do, what do they get out of it, and positioning and designing your
product in that way.
Frank
Spillers We eluded to some of the best practices throughout about
how personas can be used and evangelized and so forth. And whether they are
hype or helpful, used or misused in an organization. And we talked about
assumption personas. Let’s turn to best practices here. Do you document cases
in the book about efforts that didn’t work?
John
Pruitt To talk about that,
could we move to the slide. It’s the one where there’s a clown on it and it
says “Personas are Powerful”. We found over the years, particularly early on
when things were just catching on, though it still happens today, that persona
efforts are not always met with success. You go out and create them and they
just sort of die. There were four main themes that came out of us looking across
a lot of case studies and working with companies that had done them, and
talking to practitioners. Those four things are: firstly, lots of personas are
done as a grass roots effort. So a few people on the team get fired up about
the process. They have read Cooper’s book, or they might have read our book and
they then start to do it. And unless you get good support, not just from upper
management, VP’s and the like, but really I mean support from the influential
people on your team. And a lot of time the influential people are just your
peers. And so, picking out the peers are really key to having a really
successful user centered design effort and getting them bought into personas is
a key thing to overcome.
The second thing is that in a lot of cases the persona
characters themselves were just not believable. In some cases they are designed
by assumptions, by groups of people, who create them. And in other cases, they
were based on data but the links to the data is not clear. So, we outline ways
to combat this in our book. The essence of this is that you do have to care
about credibility. And you do have to make sure that your personas, when they
are based on data, that it is clear. And you should always go back to data.
Once you create them, we believe that you are not done. You have to validate
them and continue to update your understanding of them. The third reason why
personas fail is simply that they are not communicated well. We have seen lots
of cases where personas are created and a little poster is made. It’s put on
the wall, or a few of them are scattered around the walls of your building. And
then you’re done. We found that your really have to treat personas as a
communication campaign that you launched and then you launch them with the
notion of progressive
disclosure of information that can’t learn about every aspect of your
personas immediately. It’s just like getting to know a real person. When you
first meet someone you are lucky if you remember their real name. You hopefully remember their face, and maybe
what they do a little bit. But you get to know them over time and then you
start to develop a relationship, trust them and ultimately good friendships
have lots of influence on you when you really know somebody. And that’s really
the way personas need to work. When you communicate them in very explicit ways
over a period of time during your development cycle.
Frank
Spillers We are getting really close to the end of our time here.
Just to say, the book, that we are discussing here, the issues and themes of
the 725 page “Bible” as I would call it, can be picked up at www.amazon.com . It is a Morgan-Kaufman
publication. It runs around $55-$60 and you can search Google for “Persona
Lifecycle Book” and you should get a number of links directly to the book.
Its definitely worth the money, and it probably pays for itself through results
over and over again in terms of keeping you on track. Even if you are as
seasoned practitioner or if you are new to personas this book will come as a
breath of fresh air. I have definitely been recommending it on my trainings as
well. As we wrap up here in the theme of the lifecycle tell us a few things to
consider regarding the lifecycle framework. What are some of the things that
people should keep in mind?
Tamara
Adlin Well
I think its everything that it says on here. It’s just another method. Our book
is long partially because we want to talk about how you integrate this with
other methods that you are doing. And a big take away from all this is – doctor
heal thyself. Think a lot about yourself as a user centered designer and think
about how you are making your own work usable within your organization. It’s
not a one size fits all process. There is benefit even if you just do a little
bit. And you should be systematic and strategic in your approach. That’s really
what it is all about: making yourself and your own work usable to the people
who are consuming the customer data that you are bringing in.
Frank
Spillers Okay, great. And just here as we are wrapping up a few
questions. We have already taken a few as we have gone through, but here’s one.
“How many personas are needed” one person is asking?
Tamara
Adlin Short
answer is it depends…..which is always the short answer. Basically, we talk
about looking either at major roles, or major goals or segments in your product.
And typically a lot of companies have a
way that they are thinking about this. Maybe they are thinking about the
audience or the presenter. Or maybe they are talking about the person who plays
it economically safe, versus the risk-taker. You are really looking for the key
differences that make a difference. So you are looking at prioritizing anywhere
between one, two, three or four personas in typical projects. The rest of it
really depends on the data you have on the product that you are creating.
Frank
Spillers Okay,
another quick question here. Have you ever seen a design team use a real
persona to represent, act and think like the persona.
Tamara
Adlin Hire
an actor I guess!
John
Pruitt Or we have also seen
some cases where companies have identified a target user and said “This is a
great example of our key customer”. In that case they keep going back to that
same person. It’s a dangerous thing to do. There was also one company that we
worked with where there was an individual in the company that served as the
persona. That person did a lot of user research and their job was to “be Bill”
or whoever that persona was.
Tamara
Adlin Part
of our validation process is going back to the users at the end of the project
and seeing if they really start acting like the personas. That’s what you hope
will happen, that you were right, and that these are the goals and this is how
the product will satisfy those goals.
Frank Spillers Great.
Well we are actually up on our time here. I want to thank you both very much.
This has been a most interesting and valuable session here. Thank you John
Pruitt Pruitt and Tamara Adlin Adlin, authors of “The Persona Lifecycle
Handbook” that is definitely worth the $50-$60 that I think it’s going for out
there. This seminar will be available for archive playback and in addition to
that John Pruitt and Tamara Adlin will be presenting a workshop on personas at
the Neilson-Norman event in October in the US and London in November.
Tamara
Adlin People
can get in touch with us anytime. Just email us and we’ll be happy to talk to
you.
Frank
Spillers Email
is up there on the screen right now. John Pruitt’s email is jpruitt <a> microsoft <.> com and Tamara Adlin’s
is tamara <a> adlininc <.> com. Finally,
I want to thank you guys so much for your time and you insights. This has been
really great and we look forward to hearing more from the stories that you
gather as you interact with more practitioners. And hopefully there will be a
follow-up book in a couple of years. For those of you who may be interested, we
may have John Pruitt and Tamara Adlin on our regular web seminar courses coming
up so stay tuned to www.experiencedynamics.com
or hop on the email list there and we’ll
keep you posted. Thanks a lot and enjoy the rest of your day.
End of interview.
This seminar is available with the slides and audio presentation here (requires brief registration, privacy protected).
Best Regards,
Frank Spillers, MS
RSS
Email
Twitter
A new study (July 2008) entitled 'Factors that Improve online experience' by Sathish Menon and Michael Douma from the Institute for Dynamic Educational Advancement offers key insights on the current state of online user experience. 
Design Like You Give a Damn is the name of a 
