Menu Search
Jump to the content X X
SmashingConf London Avatar

We use ad-blockers as well, you know. We gotta keep those servers running though. Did you know that we publish useful books and run friendly conferences — crafted for pros like yourself? E.g. our upcoming SmashingConf London, dedicated to all things web performance.

Collaging: Getting Answers To The Questions You Don’t Know To Ask

When conducting user research, we all know that asking the right questions is just as important as how you ask them, but how do you know exactly what questions to ask? What if the discussion topic is very personal? How do you get a complete stranger to open up? There is a better way to conduct an in-depth interview, and it doesn’t involve a clipboard. Just imagine what you could discover if the participant’s answers weren’t limited to a predetermined set of questions. This is where collaging can help.

Collaging is a projective technique by which participants select images that represent how they feel about a particular topic. The participants then explain to the moderator the reason they chose each image. The collage becomes an instrument through which participants are able to express needs and feelings that they might not otherwise have been able to articulate. This information enables us to better understand the user’s world and how to design for it.

Further Reading on SmashingMag: Link

A Picture Is Worth A Thousand Words Link

So, you might be asking yourself, “Why should I have people make collages, rather than just ask them point-blank questions about their needs and feelings?” It’s a great question, and the answer is, sometimes the most valuable answer is not in response to a direct question, but one that’s elicited. An image can be a powerful stimulus that evokes a strong response, triggers a memory and draws out feelings that exist below a person’s own level of awareness.

Gerald Zaltman, author of How Customers Think, states that “95% of our thoughts and feelings are unconscious.” There is just so much that we carry around in thought but never share until something triggers it. When we ask a participant a set of pre-defined questions, we are predetermining the scope of the interview. Instead, by presenting a visual stimulus, we are letting the participant start the conversation and bring up topics that are meaningful to them.

There are also times when you don’t know the right question to ask. Sometimes you do, but you don’t know how to ask it. Depending on what you’re researching, participants might have difficulty opening up to you. The research topic might be too personal, controversial or sensitive for the participant to just open up and start discussing with a complete stranger. The collage becomes a catalyst for discussion, an ice-breaker.

What You Can Learn Link

Collaging is a method of building empathy with your users. You gain an emotional understanding of the user’s feelings, problems, state of mind and so on, which is imperative to know when designing for them. Collaging can also help you better understand the user’s needs, in turn helping you to address them in your product.

The wonderful thing about this method is that participants might reveal stories that prompt a line of questioning about a topic that you never expected to explore (as we’ll see in the examples below). You honestly don’t know what you will learn from each participant’s collage.

A Brief History Of Projective Techniques Link

Collaging is not new. The method has been well used to conduct qualitative marketing research for at least 40 years. Its use in marketing has mainly been to assess feelings towards brands and products. Other projective techniques — tests such as the Rorschach, word and sentence completion, draw-a-person, and thematic apperception — date back to the early 19th century. All of these methods are rooted in psychology, but their application has expanded to other fields such as advertising, management, sociology, anthropology and, more recently, user experience (UX), to name a few. Collaging as a user research method has yet to be widely adopted in our industry, but I’ve seen a steady increase in its use and popularity over the past few years.

Conducting A Collaging Exercise Link

Listed below are all the steps necessary to conduct your own collaging study. Let’s walk through them together.

1. Choose Your Topic of Interest Link

The collaging exercise should focus on a specific topic. You will be asking participants to choose pictures that reflect how they feel about this topic. For example, if I were redesigning a website, I might ask the participant,

“Select pictures that reflect how you would and would not want the new website to greet you.”

You could word this in a lot of different ways, such as,

“Create a story about how you would want the website to communicate with you. What qualities should it have? What qualities should it not have?”

Or, to learn more about a participant’s day-to-day struggle with a problem, you could simply say,

“Select pictures that reflect your experience with using [x].”

2. Create a Collage Board and Get Pictures Link

You will need a board or a large piece of paper to which the participant can tape their pictures. It doesn’t need to be fancy. In the past, I have just used 11 × 17-inch ledger paper. If I were asking the participant to create two separate collages, I would divide the paper by drawing a line down the middle. I’ve seen other people put a target on a collage board and ask participants to stick pictures on the board according to how closely they “hit home” for them. Feel free to be creative here, and find what works best for you.

Example of a completed collage.
Example of a completed collage.

Participants will need to be able to choose from a pool of about 150 to 200 pictures. The pool of pictures must be a mixture of random pictures. They should not have a running theme (i.e. no pictures of just animals or people or medical scenes or nature, etc). The pool should be a good mixture of all sorts of pictures. You can use stock photography or even pictures clipped from magazines. Here are some online sources of free images:

Print the pictures small enough (approximately 3 × 3 inches) for participants to have plenty of room to put as many as they want in their collage. You’ll also want multiple copies of pictures to replace the ones used by participants. In the past, I’ve printed pictures on stickers, which worked well but was a little more expensive.

3. Moderate the Study Link

  1. Set up the room.
    Lay out all of the pictures on a big long table. Make sure they do not overlap so that the participant can see them all. Put the collage board, some tape and a pen on another table.
  2. Give the topic and instructions.
    Instruct the participant to pick out at least four or five pictures that reflect how they feel about the given topic. Then ask them to tape those pictures to the collage board, and add a caption to each one explaining why they chose it.
  3. Leave the room.
    I prefer to leave the room for five to ten minutes to give the participant time to peruse the pictures without feeling any pressure. When I reenter the room, I tell them to take as long as they need and to let me know when they have completed the collage.
  4. Discuss the collage.
    The collage is finished. Now comes the fun part! Have the participant explain to you why they chose each picture. This is your opportunity to learn as much as you can about how the participant feels about the topic. Let the collage and the participant guide the interview. Be sure to follow up with questions and to probe deeper when needed and appropriate. Keep in mind that the experience can be very personal and revealing for some participants, depending on the sensitivity of the topic. The collage might make it easy for a participant to open a door that they don’t necessarily want to walk through. Be mindful of the participant’s comfort level when probing deeper into something personal.

4. Conduct Analysis Link

When conducting your analysis, keep in mind that what’s really important is not the pictures they chose, but why they chose them. The analysis and report should focus on what the collage reveals about the participant, not the collage itself. It would be interesting if multiple participants chose the same pictures, but even more interesting if they chose them for the same reasons.

When to Conduct a Collaging Exercise Link

Consider collaging during the early stages of product development, when user requirements are being gathered. The method is also helpful at any time in the product’s development if you feel the design team is having trouble understanding and identifying with the users. Sometimes designers need to step back and remember exactly who they are designing for.

As mentioned, this method can be most useful if the topic is sensitive, but it’s great for impassive topics, too. Collaging can be used if you just need a fun activity to put the participant at ease and break the ice before a formal interview. For example, I have conducted collaging exercises with cancer patients, with people dealing with chronic pain and even with women about their menstrual cycles and feelings about birth control. On the other hand, I’ve conducted collages to learn more about people’s daily commutes and how they feel about public transportation — much lighter topics.

Collage Examples Link

The examples below are from collage exercises that I’ve moderated. Each one shows how a picture can change the line of questioning in an interview. The topics, which were discussed because of these images, might never have been brought up in a traditional interview.

Picture from a collage that was done for research on people suffering from chronic pain.
Picture from a collage done for research on people suffering from chronic pain.

The participant wrote the caption, “That’s my daughter consoling me when I’m in pain.” When discussing this picture, I was able to probe deeper into how the participant’s pain affects their family and how they deal with it. We were then able to discuss what role family plays in how they manage their pain.

Picture from a collage that was done for research on how people feel about their commute.
Picture from a collage done for research on how people feel about their commute.

The participant wrote the caption, “Wish my commute was this enjoyable.” I was able to follow up with questions about what their ideal commute to work would be like and what they wished they could change about their current commute.

Picture from a collage that was done for research on cancer patients.
Picture from a collage done for research on cancer patients.

The participant wrote the caption, “Vomiting!” This led to a line of questioning about the side effects of the participant’s cancer treatments and their coping mechanisms.

Pitfalls To Avoid Link

Sometimes a stimulus can be too strong and can disrupt a participant’s train of thought and be a distraction. The example below resulted from two participants choosing the same image for the same study:

Pitfalls to avoid

The participants wrote the captions “Disgusting” and “Gross.” Their captions and their reasons for choosing the image were similar, but in no way did they relate to the topic of interest, which was pain management. The participants couldn’t explain how the image related to what they felt about the topic, but they still chose it because they were drawn to it and it provoked a strong emotion. In this case, we decided to remove the image from the pool because it was obviously a distraction. When conducting a collaging exercise, remove any pictures that you find derail the participants.

Conclusion Link

Collaging is a great method for learning more about your end users. Depending on when the collaging study is conducted during the product’s development cycle, your findings could do any or all of the following:

  • Aid in persona development,
  • Be used in early ideation for creating new products,
  • Reveal how people feel about the experience of using an existing product,
  • Help to define new requirements or enhancements for features.

The method might not be right for every user research initiative, but try it if you feel something is lacking from your traditional interviews. You will be amazed at what you can learn when you throw away the clipboard and let participants direct the interview.

Resources Link

  • How Customers Think: Essential Insights Into the Mind of the Market, Gerald Zaltman
  • Customer Intimacy: Pick Your Partners, Shape Your Culture, Win Together, Fred Wiersema

(al) (il)

Footnotes Link

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9

↑ Back to top Tweet itShare on Facebook

Kyle Soucy is the founding principal of Usable Interface, an independent consulting company specializing in product usability and user-centered design. Her clients have ranged in industries from pharmaceutical giants like Pfizer to publishing powerhouses like McGraw-Hill. She has created intuitive interfaces for a variety of different products, everything from web sites to touch screen devices. You can connect with her on twitter.

  1. 1

    “Reveal how people feel about the experience of using an existing product.” Definitely! It’s possible to take this a step further, too. I’ve had great success using before & after collages in the context of usability testing. It’s an easy way to probe for potential disconnects between product and brand experience. Here’s how:

    1. Before a usability test, have participants create a collage to illustrate what the company or service provider (aka brand) means to them. Ideally, do this enough in advance that there’s no recency effect on the day of testing.

    2. After a usability test, have participants create a collage to describe the experience they just had with a specific product or service.

    3. Compare the collages. The differences can be revealing!

  2. 2

    This is a great idea, we have several collage books in our office of our previous work.

  3. 3

    Patrick Samphire

    February 6, 2012 10:56 am

    I’ve always found collaging a useful technique in any creative project (web or otherwise), but the distraction issue is a major one, and not just on extreme images. Images of attractive people, for example, get picked even when not relevant, so the analysis has to be very careful.

    Collaging also allows you to see how ideas intersect in people’s heads, but to get that more complex interaction you need to give them the freedom to arrange the images in a freeform manner.

  4. 4

    Thanks Urban, Patrick, and Kadee. I’m happy you enjoyed the article and found it useful.

    Robert, I’ve never conducted before & after collages. I bet the results would be interesting. Thanks!

    Nick, I’m happy you learned a lot from this article. I would suggest interviewing 10-15 people per user group — as many as you would to create personas. For example, if you were conducting research for a university website the user groups may be undergrads, grads, prospective students, parents, and alumni – potentially 50-75 people to interview. This would be the ideal range, but you may have to make some compromises given the tight timelines and budgets that most projects have. As Kim Godwin says though, the most expensive research studies are the ones you don’t do – or, don’t do well.



  5. 5

    Hi Kyle, I really enjoyed this article. I just began working with a UX design agency not too long ago and am still working on understanding the business and everything that goes into it. I found your article very refreshing and new – I haven’t seen anything quite like this in the blogs I’ve read on UX design. This helped me understand their research process a little bit more. Thank you for bringing something different to the table.


  6. 6

    Nick de Kleijn

    February 6, 2012 4:10 pm

    Really liked this article! Teachers on our school really try to guide us to do these kinds of interviews instead of the normale quiz-like interviews. Learned a lot from this article, and can’t wait to take this to practice!

    I do have one question; how many people do you recommend to interview to get the right amount of information to work with?

  7. 7

    hi again.

    please tell me how this article has any connection to reality.

    everyone knows clients always have a very specific idea of what they want and are not likely to want to express this through a dumb colladge.

    please find more normal things to write about that relate to techologies such as CSS2 and html’s


    • 8

      Hi dipshit, I mean diptits.

      I think you fail to see Smashing Magazine is for creatives too, with ideas for creativity, idea generation, and concept processes, etc. And not just geeks sat in a dark room self-pleasuring over CSS and HTML code.


      … great article by way

      • 9


        many thanks for your reply.

        It seems like you are not processing your concepts properly.

        SMashing Magazine is for people with IDEAS and SKILLS.

        If you would like any help with your website career feel free to email me.

        Regards and thanks.

        • 10

          Hi again diptits,

          You are absolutely right, Smashing Magazine IS for people with ideas and skills, i think you may have stumbled across it by mistake whilst waiting for Warcraft to load on your gaming PC.

          Thank you for the kind offer of help with my ‘website career’ too. I’m not sure how anybody can have a career as an actual website. Do you mean as a web designer? If you could share your email address with us then I think your Zen-like knowledge would help us all.

          Kindest regards, and thank you in advance

    • 11

      Patrick Samphire

      February 8, 2012 3:05 am

      I feel very sorry for anyone who hires you to build their website. I guess you just churn out the same thing over and over again without thinking about the users at all, right? Don’t worry. You’re not alone. There are plenty of hacks and charlatans out there, and you’ll fit right in.

      Anyone else who is actually interested in making the most *effective* websites they can will always be interested in new ways of figuring out what site users need and want from a site.

      • 12

        I think diptits should take the blinkers off and be open to different ways of getting information out of clients, and achieving the client’s goals in the best possible way to the user.

        Perhaps the blinkers help with Warcraft?

  8. 13

    Great piece. I use collaging and inspiration walls a lot for developing my own designs, but I hadn’t thought about using it so proactively with others. Seems like it could be a really effective way of understanding how other people are thinking creatively.

    • 14

      Thanks, Robin. I’m glad you liked the article and it gave you another way to think about your own collages and inspiration walls.

  9. 15

    This is a great exercise, I carried out one very similar for the re-design of the Scottish Parliaments website, I carried out some visual research user sessions, where I created sets of mood-boards and presented them to the user group to spark the conversation about the visual styling & imagery to be incorporated into the site.
    The users were very responsive, I got some very direct feedback which influenced the look & feel of the website.
    When designing for the web we always consider the user in terms of functionality but not so much for the appearance, & at times I think we should!

    • 16

      Thanks, Frantastic. I agree that “conversation starters” like this method are just a wonderful way to break the ice with participants.

  10. 17

    Great article Kyle, indeed this is a very interesting technique. I was wondering though that if you show to the participants a set of collages, won’t these collages start to form the context of the response? Is it possible that you may miss some types of responses (or even guide them to a different direction), depending on the collages that you have chosen to show?

    • 18

      Thanks, Panagiotis. You raise an interesting concern about biasing the participants response. During this exercise, I don’t show the participants completed collages. Instead, I have them create their own collage by picking pictures out of a random set of 150-200 images. Now, bias can still enter into this exercise if you choose images that you expect to elicit a certain response. That’s why I suggest using images that are as random and varied as possible. This way, the participant gets to make a collage that truly represents how they feel about the topic with minimal influence . Does this help address your concern?

      • 19

        Indeed, randomness seems to minimize the bias problem, mixed with a lot of trial and error (as you have clearly shown with the “disgusting” and “gross” pictures). Thanks a lot Kyle!

  11. 20

    great article. Love to see how other people think about things differently.

    I have few collages at home but never had anything on mind before making them but gotta take a look at it now.

  12. 22

    Thanks for such an informative article! I know this one is a little out of the norm for Smashing, however I am always seeking new and better ways of connecting with clients and discovering their “true” needs, as opposed to “what they think they want”. I can’t tell you how many times I’ve been asked (demanded) to add a feature, only to find that that once added, it is never used.

    This technique is interesting, and while it may not suit every situation, I can already see applications where I would love to apply this or something similar.

    • 23

      Thanks, Todd. I’m glad you like the article and I hope you get a chance to put the method into practice.

  13. 24

    Love this article! I was wondering if there’s any way to do colleges just like these online or by a phone app?


↑ Back to top