Preston Smalley

I’ve been thinking recently about what engages people online to spend time on various web sites. People come to eBay for the thril of the auction, the high of winning. YouTube is famous for entertaining people.

I believe that as a designer, we must think beyond the utility of our experiences and consider the other sense and feelings we create in our customers. Two examples:

• Casino Design. Over the years the industry has honed the experience down to a science (see Matt of 37signals recent post on the casino experience). For example, there are good reasons that a slot machine doesn’t look like an ATM…

MONEY GAMBLING MACHINE

Sorry, you just lost $100.
[BET AGAIN] [QUIT]

• Car Interior Design. In persuit of building the ultimate driving machine, BMW considered every element of the experience inside the car. For example the sound made when lane changing, the way the knob feels when you adjust the radio station, to the thump made when the door closes. These are not accidents but done on purpose to reinforce the powerful nature of the car.

If your business model hinges not how people feel when using your product, then it is critical that you analyze the feelings people have (or want to have) when using your product.

Monday’s launch of eBay Express marks an important milestone not just for eBay but for the larger Search design community. As many of you know, I believe that the faceted metadata search system introduced in eBay Express is very much the direction that all large collections should head and am pleased with the progress we’ve made with this launch.

Marti Hearst (UC Berkeley’s Flemenco), Corey Chandler (eBay) and I spoke at CHI2006, also on Monday, about lessons we’ve learned while designing faceted metadata systems. A theme emerged in the Q&A around the importance of merging browsing and searching that was encouraging.

• Separate systems undermine confidence
  If different matches are found when you browse a category structure and another when you search using keywords–users aren’t ever sure they are seeing everything.
• Augmenting keyword queries with browsable facets works
  In the discussion, there was agreement that while many users start with a keyword query they appreciate adding additional browsable facets to their query.
• Augmenting browsing with keywords is powerful
  In a similar way enabling browsing from the homepage is a good starting point for people but then it can be helpful to add keywords to that query.
• Faceted Metadata creates a common language
  If the searcher and the classifier are different people, faceted metadata can be a good common ground which can assist recall.

Time will tell whether faceted metadata will prove to be a success, but now I am pretty confident there are at least another 100 believers out there following my workshop.

Here are the course notes posted by Marti.
Update: Jessyca Frederick reviewed our course presented at CHI. Thanks Jessyca!

Today Yahoo! made a big contribution to the larger design community. Not thru an innovative product offering or nifty new technology… but thru a site committed to sharing their design patterns with the rest of the world. LukeW (who I suspect played no small role in this) has more on this important launch as does Bill Scott and a newly launched Yahoo! UI Blog.

For those of us in large design departments we can appreciate what a huge asset well thought out design patterns are to our work. I hope other companies follow Yahoo’s lead and perhaps together we can build one common design pattern library some day.

I was discussing eBay search with a friend of mine that I hadn’t seen in a while and he brought up that he was impressed with a number of the changes we’d made over the past year. Flattered, I probed a bit further to find that it was how eBay handled bad queries that he found most helpful. This is now the third separate person that has mentioned this particular improvement to me.

Here’s how the feature works (launched in June 2005). When a buyer searches for something that returns low or no results we offer suggestions on how to improve the query. For example if a buyer searches for more than 4 keywords we suggest crossing out some of the more problematic ones along with the number of items they’d find with that query (patent pending). Here is an example of a Finding Nemo query:

 

In this way we enable buyers that would have perhaps left eBay to continue their shopping experience. In a similar way if a buyer pics too many options in a product finder on the left-nav we suggest removing some of the attributes. This is critical when a shopper goes into Tax-mode and fills out all the attributes as important to them–not realizing that the product they are describing actually does not exist. Here is an example of a Sony Laptop query: 

Please let me know what improvements you find helpful or areas that still need improvement so I can add it to my long list.

As I mentioned the other day, I’ve noticed anecdotally noticed that companies are having a difficult time filling their UI and UER positions. So I decided to see if there was any quantifiable evidence which would confirm what I already suspect (if the barrage of recruiting VMs I get is any clue).

I realized that the weekly BayCHI Job Bank email that I’ve received the past few years would be a great place to start. In graphing the jobs located in the Bay Area, you’ll notice an interesting trend (see graph below).

You’ll see that there is a sustained demand for UI Designers and User Experience Researchers that over the past year has gone unmet, creating scarcity.

Assuming this market is efficient we should see two things happen: increased wages offered by firms seeking to attract talent and over the long term an increase in supply (new employees enter the regional market).

Once again, I believe the firms over the next few years that can attract and retain top talent will be successful in achieving their business goals.

Update 4/25/07: I’m hiring, check my current open positions 

Technorati: job, bayarea, baychi, uijobs, user+experience

I helped drive forward a program of user experience changes to our search products during 2005. The first of which launched in May 2005 as the alpha test for a “New way to Shop on eBay”. In addition to general simplification of the Finding UI, it introduced the following improvements:

• Multi-faceted browsing (e.g. search by bed size OR brand OR product type)
• Matching buyer and seller language (e.g. CK = Calvin Klein)
• Multiple selection of concepts (e.g. see all the size 8.0 & 8.5 shoes)

Alpha Test (May – June 2005)

A limited test was run in five categories in May thru June of 2005 to a small percentage of users. Shown below is a screenshot from our bedding category. Press Release

New Way to Shop on eBay (May 2005)

Update:

• Finding experience launched within eBay Express (April 2006)