I especially love this tidbit:

Happy shoppers are 73 percent more likely to make an online purchase and 38 percent more likely to make a purchase at a bricks-and-mortar outlet of the same store, according to the report. They are 75 percent more likely to recommend the store to others.

Here’s the article: http://bits.blogs.nytimes.com/2008/12/30/how-e-commerce-sites-stack-up/

Some points of interest in his conversation were about documentation “objectives”, and Concept Models.  With regard to the former, he says that each document you decide to create needs to have a role, like a person on your team.  It has a solid purpose, and as people’s roles can change throughout a project lifecycle, so can a document’s.

The latter, Concept Models, seems to be capturing interest with my big-picture (strategic-thinking, wide-eyed) IA friends.  I’ve heard others reference “mind mapping”, which I think is the same thing (just broader).  I like Dan’s examples of how he’s used them — typically for his own purpose (not usually as a client deliverable) to understand the client domain or relationship between client entities.  One new thing I learned from the podcast is that is can be good for mapping content types — sweet!  I can totally grasp the value of creating one to have as a reference while wireframing.

Please join us for our first MinneUX Book Club discussion. We’ll be meeting Wednesday, Feb. 11 between 6-8pm at Wilde Roast in NE Minneapolis, discussing Bill Buxton’s “Sketching User Experiences”, which I thought would be a good kickoff since it was the most popular choice for those who took the recent survey.

When reading the book, please write down your answers to these topics/questions and bring them with you — they will drive our discussion (yes, you can still join us if you haven’t read the book):

• 2 things in the book that really struck a chord
• 1 thing you either hated, disagreed with; or don’t understand
• Have you used or will you use some of the ideas in your practice/job/work?

For additional optional reading, check out these sites:

• http://www.billbuxton.com/
• Book review: http://www.uxmatters.com/MT/archives/000226.php
• Book review: http://www.experientia.com/blog/book-sketching-user-experiences/

Here are additional details about the venue and logistics:

• Time: 6:00-8:00 pm
• Date: Wednesday, February 11
• Location: Wilde Roast, 518 Hennepin Ave. E. (Corner of Hennepin and Central Avenues in Northeast Minneapolis). Web: http://www.wilderoastcafe.com/. Phone: 612-331-4544
• Ask the host for the book club (I’ll also put up a “MinneUX Book Club” table tent)

Can’t wait to see you there!

For example, it’s natural — when we hear our customer talk in-depth about how interesting a new product is — to want to know if they would want to buy it. Or, if a participant seems off-put by the color scheme, we naturally want to ask, “What would you prefer?”

Our clients don’t usually get to sit down “face-to-face” with their customers for an hour, so there’s a natural desire to get to know them better. As researchers, we need to finely understand the strengths and limitations of every method piled into our toolkit so that we can appropriately respond to their requests.

User research is primarily qualitative in nature, meaning that we only speak to a handful of people who represent our target audiences. The reason for this is because user research studies are intended to observe behaviors, first, then use interview techniques to elicit deeper detail around one’s mental model, desires, motives and meanings. What’s critical to understand is that we never outright ask questions about these things (like, “What would you like to see?”), but triangulate a series of questions that guide us towards inferences.

Market research, alternatively, is primarily quantitative in nature (via surveys), requiring a statistically-determined sampling of responses from our audiences in order to produce valid data. We ask customers direct questions and provide them a pre-determined set of answers to choose from. The analysis of large data sets reveal customer preferences, attitudes, demographcs, purchase trends, etc. When large data sets are tracked over time, then we can start analyzing for trends.

Referring back to the earlier example, if a client requests I ask a research participant, “Would you buy this?”, I respond that qualitative data is inappropriate and potentially risky (imagine that one overzealous paid participant could remain vivid in a client’s mind). Market research and competitor analyses may be two different ways to learn whether customers will purchase a product; quantitative data is critical.

Or, if a client would like to know if the participant likes the color scheme, I suggest that designing comps and asking “Which do you like better?” in a survey is much more insightful; again, quantitative data is critical.

User and market research aren’t interchangeable. It’s critical that we can clearly relay in plain language which methods are appropriate for what circumstances.

We were welcomed by a student soon to be looking for a job in the field of IA, so we spent a little time discussing “What is IA”? Here are some notes from that discussion:

“What is IA?”

• Where things go and what they’re called
• IA’s use research to make recommendations
• IA’s also advocate for standard web conventions (there are only a few conventions – always have a home button, clickable logo, back button goes back)
• IA started with librarians
• Librarians will teach you the system to the library (like the Dewey Decimal System) – on the web, there is no one to help – thus our job is to design systems where people won’t need help
• People have goals, but the goal is never to use the site – it has to do with the use of the product (we need to get out of the way as much as possible)
• Business value of UX design: goal is to get people to give you money (it’s a competitive advantage)
• Outstanding question: Thermostats – are they usable?

Axure

• “Thou shalt not change production code” this is a tool for designers to play around and practice
• Axure is the designers sandbox
• Prototype is a communication tool to the developers
• DO NOT USE PROTOTYPE FOR PRODUCTION CODE
• Design in code slows everybody down, because everyone wants to be a designer
• Easier to be quicker on your feet, more flexible to have a flexible design
• Changing flash is a big deal, changing a prototype is not a big deal
• Axure creates “Wireframes that move”
• We are not developers, Axure makes me do things I couldn’t do and without making code
• Use Axure for a POC (proof of concept) test – we learned that this isn’t the way everyone wants to search
• Really important to do the test plan first before building the prototype
• Axure prototypes help build confidence that you’ll release with usable functionality
• Axure style editor allows you to change font, background color, headers across the prototype, like CSS to a website
• When adding functionality for a current system, good to use a high visual and high interactive fidelity prototype so that users can still visualize the current system with the new functionality
• Good for user testing, getting feedback
• Also great for communicating with developers
• “Designed by developers for designers”
• Wireframing, prototyping and documentation in one tool
• $589 for a license, includes 1 year of free updates ($150 after that to keep up with updates – 2 releases per year)
• Keep things black and white to prevent people from using the prototype as the final product
• Testing before the design and after the design to make sure things are interpreted correctly when you handed off the design
• Slap and map – slapping up a design and a sitemap to use to display functionality — adds higher visual fidelity
• In-line frame would import html file that has the swf embed file – for clicking around, you’d need to code the swf file to have that functionality
• Don’t add a ton of flash

1. A great interviewer will constantly drive the user’s dialog from the general to the specific.  I recommend doing this by asking for specific examples (“Tell me the last time you…”) and grow your probing from there.  As you gather them, keep the user’s stories neatly tucked in your back pocket, ready to bring out anytime again for other areas of deep probing.  As much as they can role model and visualize those concrete experiences, the deeper reflection you’re able to elicit.

2. When a user is sharing a specific example, creatively embody that experience by triangulating your questions by exploring it from all different angles.  I like to do this by asking the “what”,
“how”, “why” and, after peeling off that layer, repeating the series of questions.  It’s a constant unraveling of experience, behavior and emotion.

These techniques are the most reliable way of capturing self-report behaviors, which is far superior data to future-predicting behaviors (the poorest kind of user data).  And, generalized questioning — which attempts to get users to vocalize the specific answers being sought — is ungrounded in experience and doesn’t fare well for generating rich detail.  Go for the rich detail — somewhere in there are the answers.

Of course, so much depends on what kind of research already exists, and what the purpose of doing qual would provide.  Looking at it in general, here’s my response:

“In general, quant before qual is best to help you shape your recruitment in finding the right people to research.  If you have basic market research, this likely won’t be necessary.  Your boss wants to create personas from a specific behavior set.  You could easily find them by popping up a survey on the home page with a short screener to get the right behaviors — then you can survey them after the interviews to validate your findings and determine just how large of your market they are.”

Thoughts?

“My recommendation would be to start a blog and regularly (minimum – once every couple of days) write/post about what you think is interesting in the field/line of work that interests you. Add the url to your email sig.  I say blog – because it’s the easiest to set up, publish, and maintain.

Simultaneously (say on alternate days) add insightful, thoughtful comments to blog posts you think are interesting – pointing to your blog. 

It’s all about getting your name out there – associated with the themes and topics you’re interested in doing work on.

There’s lots there.

Lots.

If I had 1 recommendation, it would be: Start.

Doesn’t have look good out of the gate. Posting is most important.

How’s that sound?”

Sounds pretty dang good!  Good advice for the newbies.