An Interview with Indi Young, author of Mental Models
Use of mental models can stimulate innovative ideas, including ideas for new products and services for customers. This is a tool that can ensure our customers are front and centre in the design process. Author Indi Young describes design as organizing the way in which you enable “digital, physical, and environmental interactions that people carry out to accomplish something”.
It was a pleasure to attend a presentation Indi Young made in January 2008 about how to create and use mental models, and to recently interview her for IdeaConnection. Indi is the author of Mental Models: Aligning Design Strategy with Human Behaviour
Vern Burkhardt (VB):
You dedicated Mental Models
to your father and described him as having a “can do” approach that “laid the perspective for my entire life”. Could you talk a bit about a “can do” attitude and approach?
When I was little, if something broke around the house, we would say, “Oh, that’s a Daddy-fix.” My father would take apart whatever it was, the mixer, the TV, the car engine, and fix it. I thought everyone’s dad did this; I didn’t know that most people wouldn't dream of fixing an engine. He rebuilds engines constantly now. He has a collection of 1920’s “Nickel-Age” cars for which he mills some parts. He is fascinated by electric cars. He has built five airplanes. He installed Linux on his PC one week just to see how it worked when the weather was too cold to go out to the shop.
My mom is the same way. No whole wheat bread at the store? Buy some acreage, grow wheat, keep some of it, put it in a grinder, make flour, and bake bread. Seriously. She’s doing the same thing with wine these days, and her wine wins awards!
As a kid, I didn’t know the word “can’t.” I think this keeps me from feeling overwhelmed by challenges in the business world. Nothing is impossible.
You say the mental model method is a qualitative approach that is a hybrid of science and intuition. Could you please elaborate further on this idea?
Since the age of Taylor-ism when businesses learned to measure productivity, quantitative studies have been favored by businesses. So much so that if you say “this study is qualitative” people will immediately feel distrust for it. Emotion and humor are things that seem to have been implicitly banned from business considerations.
However, if you want to really wow the people you are serving, your customers or members, you need to understand the squishy aspects of their lives.
Organizations try to collect squishy data in the format of surveys all the time, but it doesn’t work. Surveys don’t allow you to really ask “why” and unpack the reasons behind a customer’s actions. Surveys communicate in a really narrow, limited way.
Mental models allow you to collect squishy data from a number of sources, dive deep into the reasons behind the behaviors, and present it as an authoritative snapshot of how you support your customer and where you have opportunities to do better. Mental models interpret squishy into actionable data that organizations feel they can trust.
Some of our readers may not be familiar with the term "squishy data". Could you explain what you mean?
Squishy data simply means information collected in the field that includes people expressing things colloquially and comments about their feelings and philosophies.
Up until now, it seems that businesses have focused on surveys and data collection that appears
to be more rigorous and academic. Organizations have been convinced that focus group data can have merit, but a focus group is a very controlled environment for collecting only the things you want to study, not surprises and new ideas.
Squishy data collection, like interviews and contextual inquiry, inevitably include things that seem outside of the expectations of your research. Often these things are the elements that cause businesses to shift direction based on a better understanding of the customers' world.
I gather that use of mental models increases the odds that an organization will be successful in its development of new products?
Yes. Mental models help you clearly see and prioritize opportunities.
Mental Models help you focus on people and their problems, instead of the product or service that you already provide. If you can just set all that aside for a week, forget what you do to support people entirely and focus just on people and their stories, then you will have a good picture to use. You can align your methods of supporting people with the various parts of this picture and get a sense of how well you are doing and where you can improve.
Would you describe the essence of a mental model, and what it can be used for?
A mental model is a diagram showing where your organization supports the behaviors, philosophies, and emotions of its people. Such as customers, members, employees, partners, students, and patients.
It is not limited to software organizations. Any organization with any product or service needs to study the whole
relationship a person has with your organization. It can be used to clarify strategic decisions regarding the direction of your organization
, to innovate and support people better, to spread stories and understanding internally, to guide development teams over long periods of time, to inspire team confidence, and to ensure continuity as teams change over time.
You use them at a macro and a micro level.
(Vern’s note: Visualize a mental model as composed of three levels resembling a city skyline. Mental spaces are like the city blocks. Towers are in the shape of the buildings, and tasks form the windows of the buildings.)
What is a mental space?
A mental space is how a group of people think about one part of what they are doing.
It is a collection of towers in the model representing things people do, why they do them, and how it makes them feel.
Usually mental spaces are separated from one another by an increment of time. I often describe them as different “head spaces.”
Within each mental space existing and planned product features are aligned into towers. Please explain this.
To complete the mental model, you take a list of the ways you support people such as services, forms, calculators, mentoring, coupons, ratings, monthly statements, awards banquets, free food, beeping noises … everything. And you look at each item. Each item will support one particular tower the best, and it will also support other towers “kind of”. You put a copy of this item below each of these towers, marking the one where it is the best match for the tower.
From this point, you are free to brainstorm new things to put below the towers, or fix up the things that are currently there, or combine them in new ways.
If you have leftover stuff that doesn’t support any tower, then you consider it. Is it really necessary? Or is it something orthogonal that you should really push to a different business line? Or is it something eye-opening that should be made a high priority?
(Vern’s note: “Orthogonal” means, in the case of geometry, perpendicular to or at right angles.)
And a content map?
The content map is just the list of items I previously described. Services, forms, calculators, mentoring, coupons, ratings, monthly statements, awards banquets, free food, beeping noises … everything that pertains to your relationship with those you serve.
I like to have a team other than those doing the mental model create the content map, and I encourage them to organize it how they see fit. When we slot the content to the towers in the mental model, we take everything out of that organized format, so it doesn’t really matter if it’s organized or if everyone agrees about the organization.
The purpose of developing a mental model is to identify gaps and therefore opportunities for new product development. Have you found your methodology to be one of the best ways to accomplish this?
Yes, that’s why I developed it.
Business stakeholders would be looking around the table with dogged expressions, pursuing a line of thought centered on the current product or service they offered.
I came from the perspective of the end user and wanted to represent them and make sure their world was understood. I tried a few approaches, and the combination of task analysis, people’s stories, and how the organization is currently supporting them really resonates with those stakeholders around the table. And it resonates with the development team at a smaller, more focused level, too.
I gather that mental models are very useful for developing software products. Can they successfully guide the development of products other than software?
And it can guide more than just product development. It can guide how customer service responds, how sales are done, internal workflow, how employees communicate with each other, and so on.
It even helps me understand my cat better. You can see the cat mental model on my book site. Just for fun and amusement, I talked with a number of cat owners, asking them about the underlying motivations of their cats’ behaviors.
You think the interview-based method for developing mental model diagrams is a good way to solve problems. Would you explain this?
Well, it’s one of several ways to collect data, and it’s a much better way of understanding people than surveys.
If I had three wishes from a genie, I think I’d spend one on wishing that organizations stop using surveys to understand people. Surveys are fine for understanding demographics and for testing awareness and for getting reactions to concepts, but not much else that businesses care about.
Interviews allow you to ask “why” and “why” and “why” to unpack the many layers of reasoning behind people’s actions. Field studies, following people around for a day or so, that’s the best. Diaries, customer service calls, email conversations, blogs, reviews … these are also good sources of multi-layered data.
I recommend a technique that makes sense for the situation and for the personalities on the team gathering the research. Just stop using surveys for this kind of stuff.
You are describing the benefits of dialogue with people, where one can really explore how people think, feel and what their priorities are. Why do you think some organizations do not realize that two-way conversation with customers and potential customers is preferable to conducting surveys when they are planning future business opportunities?
To put a point on it, organizations should listen better. Forget two-way conversations... try out one-sided conversations were the organization does very little talking and very much listening.
Organizations believe that since collecting data in surveys works so well for evaluating customers' reaction to things that it will work for planning future business opportunities. It's the classic problem: "I've got a hammer, so everything is a nail." They are familiar with surveys, so they use it for everything.
There is a good table in my book that outlines the three types of research (preference, evaluative, generative) and the different methods you should use to collect data for each. Surveys do not fall in the generative research row, where generative research is the type of study that an organization can use to innovate. [Vern’s Note: Indi’s proposed technique to collect generative data includes non-directed interviews, contextual inquiry, mental model, ethnography and diary. Interestingly she does not include customer feedback but does indicate it is a method of collecting data for preference (opinions, likes, desires) and evaluative (what is understood or accomplished with a tool)]
You’ve had a great deal of experience interviewing people you have not previously met. Has this improved your listening skills, and if so, in what ways?
It has fabulously improved my listening skills.
I have learned to notice when I make internal assumptions about what was said, to stop myself, and to ask “why”.
Even in conversations with friends, I have started to notice when I make an internal agenda for what I want to add to the conversation, so I try to drop it. Instead, I pay more attention to what my friend is saying and follow the topics that she is covering and ask her to explain what she means.
Try this yourself today. It’s really eye-opening. I have to confess, however, that I can’t-for-the-life-of-me do this with my family. Alas.
You compare the interview process to a toddler asking endless “why” questions, until the adult reaches the end of their field of knowledge—or perhaps their patience. Do you make a conscious effort to play the role of the inquisitive toddler when you are conducting interviews?
You talk about adopting our customers’ perspective, and one approach you recommend is using expressive verbs to describe actions from the customer’s point of view. Indeed, you advise that verbs are the most powerful way of getting one’s development team to shift towards a “customer first” approach for product design. Would you explain how this works?
If a team can really “walk in the shoes of their customer”, really understand what it means to live in that customer’s world and try to get stuff done with a wide variety of tools and methods available, in a hectic environment with many demands and distractions and stresses and decisions to make, then that team has a much greater chance of developing stuff that really makes a difference.
If a team can do this for a period of time, perhaps four weeks or six weeks, then it leaves an indelible mark on them. They can’t help but understand. Personas help preserve this understanding and communicate it.
Using the verbs for a few weeks, speaking in the actions and simple steps and feelings that a customer has, really transports a team inside the customer’s head. It gets team members out of their “I’m an employee here” mindset, out past the “customers would want this” exploration, and really into the customer’s shoes.
It seems that developing mental models is challenging, hard work and potentially very expensive. It this true?
Yes and no.
You can actually sit down and sketch a mental model by yourself in a matter of a few hours.
It’s a challenge to get yourself into the right mindset, though. Using customer’s verbs, representing their world, not your perspective of how your organization touches their world, helps you get into this mindset. Avoiding the personal pronoun “I” helps, even if you’re a potential customer yourself.
With practice it becomes easier. I sat down and sketched a mental model for people training for a marathon in less than an hour, although it was based on conversations I’d had with people over the previous year where I listened deeply and asked “why” a lot. Before the dot com bubble burst, I used to cloister a team with a whiteboard for a day and we’d come up with a mental model by the end of the session. So there are inexpensive ways to create mental models, though you would be missing the actual touch with the customer in this case, which might shoot holes in your confidence in the data.
It can be expensive time-wise. Paying for recruiting and transcribing and stipends and travel or telecommunications costs doesn’t add up to that much. If you were to hire consultants to create a mental model with your team, then it might cost anywhere from $50k-$100k. For the value you get from the mental model, that’s a bargain for some companies.
For small organizations, that’s prohibitive. In these cases we skip touching base with the customers, and I lead a day-long mental model sketch session with the team and help them use the model afterwards.
What are a few of the lessons you’ve learned about people collaborating to solve problems, to share knowledge or to do strategic future planning?
One thing I hear a lot is that the people who believe in user-centered design are not in positions of power or strategy in the organization. They often say, "Well, I'd like to do that, but I can't". So I give them a pep-talk, saying that anyone who has passion, persuasion, confidence, and charm can get decision makers to see the value of paying attention to customers and designing the whole user experience. The hardest part is believing in yourself enough to demonstrate to those in power the value of this approach.
Secil Watson, Senior Vice President of Internet Channel Strategy, recently wrote a wonderful article in Interactions magazine, the January/February 2008 issue, about how she got the attention of these people at Wells Fargo by orchestrating a small win as her first project. They saw the immense value her approach provided and began to trust her. Now her department is in a strategic position. (Vern’s note: Secil Watson’s article was titled “The Business of Customer Experience: Lessons Learned at Wells Fargo”, and it can be read online at the Interactions magazine website.)
Something similar is happening at Qualcomm, where Srinivas Raghavan is shifting the focus of the executives to the value of user-centered design.
Two appendices to Mental Models
and a number of additional resources, including an early version you have written of a Python script that can be used to automatically generate a mental model diagram are available on your publisher's website. Is this an innovative new model toward publishing a book? One in which the publisher enables the author to share, on an ongoing basis, new resources and ideas, and to communicate and collaborate with readers?
Lou Rosenfeld's new model of publishing incorporates user centered design as its core. He reached out to his target audience to understand the world they inhabit and decided how to support them better with online resources and active communities. He ran usability tests on both the printed form of the book and the PDF form and used the results to fine tune the functionality. For example, there are little "timelines" that appear at the beginning of every chapter in the PDF version showing you where you are in the book, and how long the upcoming chapter is in relation to the rest of the chapters.
I have been writing a blog for two years now which has helped spread the word about the concept and the book, so much so that we sold books in 30 countries outside the US within the first week. I expect the blog to become a center of discussion among people practicing this work, as well as a place to upload improvements to the mental model tools and little tweaks that people make to the method so that it works within their organization with their resources and their time limits.
This is an innovative thing for a publisher to be involved in.
Are you currently working on some other interesting articles or projects?
I have been giving a lot of presentations at conferences and at companies and universities about mental models. Your readers can see a list of upcoming appearances on my blog.
I just finished some work with Hewlett Packard capturing a very complicated production workflow into a mental model. I am working with the online money/spending/tracking tool for Mint.com. They have some creative people there who have a lot of passion for making their users' finances controllable in an easy fashion. I am mentoring various folks through the process with a weekly or bi-weekly review of their work. I recently taught a remote workshop to a forward-thinking agency, Taten, in South Melbourne. That agency sees a lot of application for mental models in the superannuation industry in Australia.
And you are available to assist firms wishing to include the mental model approach in their strategic planning?
Yes, I am available to assist any organization. I work with all sorts of groups, ranging from under-funded non-profit organizations, start-up companies, and universities, to the Fortune 500 corporations.
I tailor each engagement to meet the needs and budget of each group. I do anything from the actual research to a couple hours of guidance to private workshops. I have several associates who help me collect and analyze the research data, so we're capable of helping teams who don't have the internal resources to produce mental models.
Adopting the customer’s perspective is not a new concept; for several decades large and small organizations have strived to “put their customers first” in their service and product support. Indi Young does not just talk about the need to put customers first in developing new products, writing computer code for a commercial program, developing a service offering, or otherwise meeting the needs of existing or potential customers. She has provided a great deal of “how to” information, templates and specific examples to assist with collection of generative data through interviews with customers and potential customers, and to use this information to produce mental models. All this should support of the design process and to increase creativity and innovation.
I am intrigued with the publishing model that Rosenfeld Media has followed since 2005. Its focus is on publishing “user experience” books providing practical advise that “that tell us how” to design complex products and systems. Mental Models fits this strategic publishing direction and therefore was a good choice.
Readers please note: Lou Rosenfeld, the publisher, has offered IdeaConnection readers a 10% discount when they purchase Indi Young’s Mental Models: Aligning Design Strategy
. When placing your order through the ordering system
you can enter the discount code IDEACO10. With this purchase customers receive the digital edition that is optimized for on-screen use. Also, the book's images
are available by way of Flickr.