Ethnography for Marketers -- A guide to consumer immersion

Ethnography for Marketers was recommended to me in 2007, I finally got round to reading it in 2010 and the other day I revisited the copious notes I took. This is a book about ethnography, research, projects and design. But why write a blog post that is a book review? Particular when the subject is essentially a text book?

As the title suggests this is a book targeted at marketers not designers but don’t let that put you off. This is a text book for user centred design that anyone who describes themselves in anyway “UX” should read. This is also a text that anyone managing UX projects should read.

Not only does the book provide a framework to conducting observational qualitative research – it goes into the detail of how to conduct that research:

• Recruiting for respondents
• Designing the research
• Project planning
• Project management – costs, risks and contingencies specific to design research
• Analysis
• And various ways to present findings

The book provides juicy nuggets for how to answer the perennial curly question of sample size with qualitative research and describes how to triangulate research with available data. It also talks about when not to employ observational research, what to use instead and what to use to as a shortcut to get similar results.

Product innovation that has sprung from observation abound in this book. For instance, consumers spraying room deodorant after cleaning the bathroom seeking olfactory cues for task completion leading to products with and marketed for their more appealing scent. Being a product design book you will have to imagine your own examples if you work purely in the digital realm.

Frameworks are provided to help designers turned researchers interpret observed behaviour, e.g. looking for normative consumer behaviours such as:

• Combining products
• Work-arounds
• Indifference – putting up with inadequate results
• Errors – customers and users blaming themselves
• Avoidance  of tasks
• Imagining perfection

Templates are provided for the various tools you will need when planning and conducting research such as an observation guide, a respondent information sheet and site report template.

Most useful is the description of the site visit and guidelines for interactions with respondents – timing of the visit, how to conduct yourself to build trust and rapport, what to look for, and participant turn offs that can jeopardise results.  Having conducted design research myself I know how tricky it is to ask questions sometimes. Mpolski provides descriptions of types of probes as well as specific examples to help you grasp the technique:

• “Describe how you…”
• “Tell me about your attitude toward…”
• “Let’s talk about…”
• “I’m curious about the ways you…”

Also discussed is what not to ask (avoid asking “why” specifically) with an explanation of the types of defensive response this provokes in participants.

Designing products and services based on user or customer behaviour starts with observing that behaviour. If you like me are a designer sans academic research or psychology background you might find this book a more than useful read. If you are thinking of beginning a design project with research but don’t know where to start, how to plan it, resource and budget it this is a must read – and don’t forget to encourage any project managers to read it too.

 Ethnography for Marketers  – A guide to consumer immersion, by Hy Marriampolski, Sage Publications, 2006

back row snap of the event

I first saw Sebastian Chan speak at Web Direction on 2007. He presented on social tagging (“folksonomy”) projects at the Powerhouse museum. The first of these projects was the digitisation of electronic fabric swatches. After that the entire collection was digitised and published available for public classification. Recently I saw him present and got an update on these projects.

When I first saw this case study presented in 2007 the stats amazed me:

• 95% of all available objects were visited at least once in the first 10 weeks of the collection being published online
• 86% of tags created by users were not found in museum (curatorial) descriptions
• 88% of tags were assessed as useful by museum staff

The result since has been an amazing amount of web traffic, content sharing and fresh perspectives on objects and audience interests. The technology has developed to include geo coding and search term generated tags or “frictionless” tagging as Chan called it. But in all that time I had one question, well make that two:

• Do the same items stay popular because of tagging? (i.e. does tagging perpetuate was is already popular?)
• Does the tag list get unwieldy over time?

I got the opportunity to ask Sebastian from the audience Q&A after he spoke at the Australian Infront Insight series of talks. Warning audio taken from iPhone and not that great.
Sebastian Chan on Tag Lists by erietta

What are the lesson to be learnt here? 

The museum found that audiences didn’t need incentives to describe objects. The audience tags unveiled a new lexicon of descriptions that in turn led to deeper and wider content exploration. If you have a large library of information, a catalogue, or a huge database of content open it up. Let your audience help you describe it and in turn, help other users find what they are looking for. I encourage you to check out the original 2007 presentation.

Related links:

• Seb Chan’s 2007 Web Direction’s presentation
• Fabric swatch project
• Some of Sebastian Chans’ other posts on the subject of tagging:

• http://www.powerhousemuseum.com/dmsblog/index.php/2007/06/27/a-reminder-about-user-incentives/
• http://www.powerhousemuseum.com/dmsblog/index.php/2006/09/01/taxonomies-of-tagging/

The Powerhouse Museum and the in-house digital agency Chan has been heading within it have liberated the collection and extended the museum experience beyond exhibitions and museum walls.  Sebastian Chan is head of Digital, Social and Emerging Technologies at the Powerhouse Museum, Sydney. I first saw him talk at Web Directions in 2007. Then he case studied social tagging projects and it was great to here how the initiatives have grown.

Amongst the strategies, tools and technologies discussed on the night were:

• Open APIs to enable sharing of collection data
• Geo coded images enabling data mash-ups of historical photos
• Social tagging enabling richer descriptions of collection items
• Keyword search tags enabling ‘frictionless tagging’ and driving improvements in how items are catalogued by examining how people describe them
• Online exhibitions (such as the Australian Dress Register) encouraging learning and cataloguing amongst smaller collectors
• Interactive games as part of exhibition displays encouraging learning through experience
• Mobile apps augmenting museum experiences (check out the Lovelace app in the App Store)
• Pre-exhibition blogs that help inform shows and even to crowd source artefacts to display

Chan calls this use of technology and the philosophy of an open collection the “post web accord”. He acknowledged the tensions that this raises:

• the family visitor versus the scholar
• the exhibit versus the collection
• the museum site versus new platforms
• formal education versus inspiration
• delivering experience versus displaying information

The benefits of embracing the post web accord speak for themselves. The exhibition “The 80s are back” began with a blog and 50% of the objects displayed were crowd sourced. After smaller initiatives the complete museum collection was digitised and published under a creative commons license with social and search tagging tools in 2006. Now links back to the site account for 50% of all web traffic. But it doesn’t end there. It’s always a curly question to consider where to publish content online, and whether content should be seeded elsewhere. Consider these results—the museum published images onto Flickr under the Flickr commons license. Images that had generated 31,000 total views up to 2007 took 4 weeks to generate 40,000 views. After 2 years the image collection had generated 2.5 million views and now over 1,800 photos have been added.

My main takeaways from the event?

• Push experimentation, pilot projects and grow good ideas.
• Embrace multiple viewpoints from the audience and from within your organisation (museum exhibitions are blogged from various viewpoints on their site).
• Encourage sharing by opening data and publishing on multiple sites–not only will it result in new information coming back to you, but communities will form around it.

This event was held on 21 August 2011 and was presented by Australian Infront at the Apple Store Sydney as part of the Insight series of talks. The talk will also be published as a podcast. The next event in the series will be a presentation by designer Vince Frost.

Related links:

A write up of the event by Justin Fox, Australian Infront founder: http://www.spamventdocument.com/infront-insight-06-sebastian-chan

Tram conductor - Illustration by Nam Nguyen

A reader of the article, Lisa Chow, cited an example from her own professional experience as a library consultant in the comments. A system to check out books replaced librarians doing the task but the self service model wasn’t necessarily efficient for users trying to achieve multiple tasks in the act of borrowing a book–like querying outstanding fines.

Customer efficiency is not about the streamlining of one or two tasks. It is about services anticipating the end goal of customers and providing resolution in a manner which does not increase effort for the customer. This must be a consideration in the design of systems, services and processes undertaken by customer facing staff. If you are curious to read more about the topic check out the article on UX Mag.

It was heartening to receive so many comments (after the hard slog of writing the article ;-) particularly one that provided another case study. The article is part of a series on “The 7 essentials of customer experience” by Different. The first was on predictability and the 3rd article on convenience will be published in coming weeks.

And thanks to my editors – Joe, Amanda and Mimi. It would have made no sense without you!

Having worked for a company behind open source software, I know how important community conduct is, on forums and other channels. In fact it was something that Geoff, as FarCry product evangelist had to (and I’m sure still does) moderate closely. This interaction between products and users is vital in fostering closer relationships between companies and customers, feature improvements and product innovation.

So it is great to see Get Satisfaction create a campaign around this. Get Satisfaction is a service that allows customers to send feedback, bugs and feature requests to companies. They have created a campaign called the Company Customer Pact. An accord, or code of conduct if you will. As interactions between companies and customers get closer through social media it will become more and more important that people are on the same page. Check it out, and get on board.

http://www.ccpact.com/

Elsewhere in the organisation other stakeholders held competing and contrasting views of what needed to be designed in the schema and what labels needed to be used. User testing the IA was seen as a means to streamline and manage the internal decision making process by bringing everyone together on the same page and letting users themselves determine the outcome.

The project was a simple engagement but an important one in an organisation that is seeking to embed UCD and customer centric thinking. The client team invited their colleagues to witness proceedings. Luckily during the testing itself I completely forgot that up to 17 people were watching me! time was scheduled in between the user testing sessions to debrief. I collected the thoughts of the observers; their impressions, implications and design ideas. The debriefing sessions were important. Not only did I benefit from the input of subject matter experts, but members of the organisation got to see the process itself. I was able to field questions about the method, diffuse doubts on the spot and collectively we arrived towards the findings. When it came to the presentation of the recommendations there were few surprises and high engagement.

The testing was not about flogging a dead horse to validate how much something did or didn’t work. Sufficient time was left between sessions for important changes to be made and tested. Stakeholders learnt by observing the process that user testing is not a research exercise. It is a design process. One that is more effective, more pertinent and faster than design by committee.

The project … had proceeded too fast – apparently because of contractual obligations – for clinicians’ feedback to influence it, Dr McCarthy said.

…

FirstNet, commissioned in 2008 from the US health computing giant Cerner in Missouri, insisted on redundant information, such as confirmation that male patients were not pregnant, Professor Patrick found. But it obscured essential information, forcing users to click through six screens to find the phone number of a patient’s GP. Excessive system downtime led one department to revert to a whiteboard to ensure basic patient information was accessible.

– Julie Robotham, 7 March 2011, smh.com.au

Worryingly this project is “part of a 10-year electronic medical records plan”. Hopefully lessons have been learned that will be carried across the programme.

In other news MediaWatch outed itself and other television stations for badly designed closed captioning. I occasionally put on teletext when the house mates are too noisy but the footage illustrates just how frustrating it must be to be hearing impaired and watch television.

MediaWatch do close the story with an example of closed captioning done well on Channel 7. Closed captions, like IT software design projects can be done well. All that is required it seems is a will and a budget to deliver a satisfactory experience.

Facebook wall populated/polluted with Branchout status updates

See all those horrible, horrible application status messages. I like to think of myself as pretty web savvy but I lost some serious web cred today when I inadvertantly spammed countless friends.

How I became a web idiot

After returning from a workshop I was feeling pretty tired and looking for any semi legitimate opportunity to avoid actual work. Then I got a Facebook message from Ahmed my workmate. OK it was a Facebook message – but he’s a colleague, so its kinda work related, right? I could be missing an office meme here! OK. Click.

Then I am faced with a question. “Who would you rather work with? Ahmed or Ari?” Now here’s the thing. I work with both these guys, and a whole other bunch of adorable folk. We joke around a lot, and Ahmed and Ari in particular. I was thinking this was a deliberate cute game for Ahmed to rack up some teasing ammunition if he could get more people to “vote” for him than vote for Ari. Thats what it looked like anyway. Faced with a choice, I noticed an option to choose both. Joy. Diplomacy wins. I have no favourites.

Branchout being douche bags and the self selecting check box

After this I was faced with 2 more choices of friends I would rather work with. Both. Then 2 more. Both. Then 2 more. Both. Then 2 more – actually I would rather work with you. Then 2 more – oh, there’s Angela from high school, she’s really smart, sure I’d love to work with her! Then I noticed the check-box. Shit. Had I been spamming my friends all this time? I unselected the checkbox not wanting to promote BranchOut to anyone else without knowing what it was about. In my tired stupor and like an hypnotized pokie addict I played on. And on. Oh and a few more selections. Until I noticed that the check-box I had unselected defaulted back to select state with the presentation of each new pair.

Its evil mechanism is explained in this message between my friend James, who got spammed by me being gamed by Branchout.

Me explaining myself to receiver of spam-like branch wall post.

F*!k you BranchOut. You are winning no friends, fans, connections, branches, leaves or whatever stupid metaphor you’re going for here.  Clearly the clever designers know how to employ game strategies to lull people into joining their application. That’s OK but only to a certain extent. A line was crossed! Yes, both James and I were “intrigued” and clearly, I am an idiot. But that check-box is a small call to action, and not only that – it re-sets itself to be selected when any normal sane person would assume that once they had interacted with and changed this setting, that the setting would apply permanently. Not so. To all my friends, sincere apologies for polluting your wall with rubbish.

Senova's challenge: can we design our industry together?

Melis Senova began her presentation with this premise: if you interpret every choice as a design decision, you can look at your life as a designed experience. And concluded: if we all design our lives, can we design our industry?

Her presentation was a challenge to apply the principles and characteristics of user experience  design – collaboration, transparency, sharing, iterative, open, designerly, innovative, empathetic, cross disciplinary – to design our experience of working in our industry.

Melis Senova: attributes of user experience

Melis Senova: Business as usual

Zafer Bilda opened the night with a service design case study that followed a familiar working approach (1) define the business objective (2) survey the landscape with a comparative analysis (3) conduct observational research (4) create customer and staff personas (5) study service tools in use to design a new experience.

Perhaps we can take this working model and apply it to Senova’s challenge. Is a bar camp in order?

Service Design Drinks 6 was a full house.  Service Design Drinks 7 will be on 23 November and will be reviewing highlights from the upcoming Service Design conference in Berlin.

Related links

http://servicedesigning.com.au/2010/09/21/photos-from-sydney-service-design-drinks-thinks-6/

Erietta: Tell me about your current role and your areas of expertise?

Philipp: Theoretically I’m an Experience Architect and a junior version of that I guess and I think what my areas of expertise … mostly comprise of are user research … the gathering of the data and how to structure this gathering cause there’s actually quite a lot of theory involved in how to gather info for which kind of results you look for. For example a very experimental approach compared to reviewing a product that already exists; you have to adjust your techniques each time and of course you also have to adjust the guys you are actually interviewing.

Of course [there's] the whole process of actually getting something out of this collected data—the task modelling as a more interaction focussed method of getting information out of it or the affinity diagramming which is more of a qualitative basis that gets you more into these patterns that really help you to image people or the average people actually [using] this, and the expectations they have and how they perceive the product.

The next step which I’m familiar with but not as practised in is actually transferring all these collected data into a reconstruction of a product or the entire concepting of a new structure which basically is wireframing and interaction design and all that sort of stuff.

I’ve learned about, in school actually, quite a few different approaches than now I’ve been getting exposed to at Different. I learned a design approach from Rosson and Carroll which is called scenario based design. They have a really interesting approach of slowly iterating a scenario. Just one story and you stick with that story and slowly expand on it …

The actual concepting I haven’t had much exposure to that, at least in the internships. Paper prototyping … to get your first research findings confirmed, whether they actually work, whether you got the idea of the product right. I did a video prototype actually for a project in school that was a head up display for a car and I designed a video simulation for that which was a lot of fun and the testing for that went incredibly well. The participants were really able to imagine they were actually sitting in the car and that was very interesting. That’s as far as I get cause I’ve never been included in the implementation. That was the border so to speak … I don’t know what it comes down to when people start coding it, the problems that can arise there.

eri: You’ve talked about a couple of internships, is this your start with your professional UX career?

Philipp: I guess you could say that. The first intership was a mandatory part of my studies. In the fifth semester … you have to do a 6 month internship in some kind of company that can make use of this particular set of expertise and abilities … I reckon this would be my first professional experience apart from the fact that I didn’t get paid.

Both: laugh

Philipp: I don’t know whether it counts professionally, but whatever. This was my first real exposure because interestingly my first internship was on the client side … at Bosch. They were all internal projects, and it was just a totally different atmosphere because they had time, and money and room to discuss. It was also much harder because within a company you always have the marketing department as being kinda the king … This makes it incredibly hard to actually get something through or get something passed them … Its also much more stiff and conservative atmosphere as compared to a consultancy which my second internship now was in which was a lot more flexible, more time pressure with less money and budget to do things. Always working against constraints and with what the client wants and all this stuff so its really interesting to see both sides.

eri: So tell me about the course that you did study and the qualifications gained in that course?

Philipp: The course was officially called Information Design and I think the priorities of this course were psychology: the psychology of information acquisition, the psychology of memory, the psychology of how humans try to understand things. Mental modelling is a big topic there for example. The other part was design. I’d say psychology was 40%, design was 40% and 20% was technical stuff like databases and coding and HTML and PHP and all that stuff. But that really got neglected which many were not really satisfied with because they were [planning] to get web design [skills] so they really needed PHP and HTML expertise. They were a bit disappointed but I wasn’t since I didn’t like it anyway.

The design stuff, we had real drawing courses as well as interaction design courses, interface design. We had web design courses … just those very specific things you need to know when designing an interface, like the fold, the order of items a viewer looks at first and how do you guide them with these tricks for navigation and all that sorta stuff … we also had tiny courses on business regulation and business law which was very cool to know so you have a basic knowledge of what you can call your own without anyone copying it. Just so you can have a basic understanding of … copyright … they also had library management and lots of economy information design.

eri: What were your favourite subjects in the course?

Philipp: Psychology. All of the different psychology courses.

eri: How much hands on experience did you get with research? How theoretical was it? How practical was it?

Philipp: Something that was really cool with my particular course of study was that it was heavily practical. From the third semester onward all our courses were solely project based so we always had company partners like Neckermann (a big e-commerce company), like Bosch, Porsche. We worked with all kinds of companies … so all our projects were maybe a watered down version of real projects but they actually were real projects. We were for example, analysing the web shop of Neckermann and were doing usability testing with them, with real participants. As far as I can remember each of those psychology courses always had a practical part were we actually did a survey or a usability test or online survey so we always had this exposure to practical stuff. I can’t remember a single course that was theory only.

eri: What subjects or skills do you think are most relevant to a user experience role and why?

Philipp: First and foremost I think you need to be really strong in your analytic skill. Without ever having heard about a thing before you need to be able to approach something, view it from all possible angles, see through it, see what’s behind the first layer, what it is comprised of , how it works, innately how it is related to other things …. having like X-ray glasses for everything. That not only helps you to understand the products you’re designing or redesigning, it also helps you to understand the people that are talking about it. That’s probably the second important thing to be really good at analysing the people side of things and be able to drill that down to their initial motivation to say something without leading them to say something that they wouldn’t have if you hadn’t asked that in a particular way.

eri: What do you wish you had been taught?

Philipp: All the crazy client management stuff. Because of course when you are learning everything in theory you always think of the best case, and if everything goes well, if you had enough time and money. But the practicality of the daily user experience professional is so much harder cause you constantly need to justify yourself, your methods and explain to everybody what you are and who you are and just defend your methods really. Of course if you’re dealing with clients they usually wouldn’t know anything about user research or usability engineering so you have to explain it to them as well and make them feel like it is something desirable so they’re actually buying [into] it and you need to guide them along the process and show them ‘these are the steps and you will get this deliverable and you will take part in this procedure to contribute to it’.

eri: What UX skills cannot be taught?

Philipp: … being analytical. What’s connected to that is making things easier. As Leonardo da Vinci said simplicity is the ultimate sophistication. Its not easy to make something simple. Just because you understand something doesn’t mean that you’re able to make it simple for anyone. There’s actually a lot of effort, a lot of thinking, a lot of creativity involved in making something very complex so easy that everyone can understand. Its both something that you have to have an innate talent for, being good at explaining as a basic skill, and you need a lot of practice and knowledge.

eri: So I’m going to throw in a bonus question, you’ve just completed this internship, what sort of job are you going to look for now?

Philipp: This particular set of activities that the Experience Architect here [at Different] has, everything from the user research, up to the very point where it goes to implementation, so its research and concepting … I’m not sure whether this is a common set of skills that a job role would include in Germany … they either have only researchers … or a designer that really does everything; that is able to do the user research, the concepting and the implementation as well. I’ll probably end up in a research company … The usability testing, the research, coming up with personas and all that stuff is actually something that I enjoy most. I’m not enjoying sitting in front of the coding software and doing the actual implementation. Its just boring and tedious. I feel like a machine when I do that.

eri: well good luck in your next steps, and thank you very much for the interview.

Philipp: No worries, my pleasure.

Related Posts

UX Studies: Interview Series Introduction