User Experience Results Tree

What is the difference between “user experience” (UE) and “usability?” Below we'll try to show the difference by building an outline (or tree) that will show how all of the pieces fall into place and propose definitions for some of these evolving terms.

Our conclusion, as you too may have presumed, is that UE is broader than just usability... but in making the tree it became clear that the relationship is one of timing and values. UE includes items that are chronologically before and after actual use, and it includes objectives that the supplier of the website is interested in—vendor values—as opposed to interests solely of the visitor. We’ll use the term “publisher values” to encompass non-profit site authors as well. Keep in mind as you read that I am a usability person, not a UE person... I value usability.

User Experience:

A. Before Use:

1. Usefulness (although usefulness can only be confirmed by the “acid test” of actual use, the potential usefulness is established prior to anyone using a tool, when the creators determine the specific functionality that they intend to provide)
2. Integration of brick-and-mortar organization

B. During Use:

1. Sells (makes the visitor take away more than what they came for; a publisher value):
   1. Encourages (...the visitor that the site has what the visitor wants):
      1. Tagline
   2. Pleases the eye:
      1. Sans-serif fonts
      2. Appropriate background color
      3. Conservative quantity of colors
   3. Engenders trust:
      1. Matches non-Web corporate identity
      2. Anti-aliased graphics (a subtle mark of professionalism)
      3. Appropriate color hues (a subtle mark of professionalism)
2. Satisfies (gives the visitor what they came for, a customer value; I propose that only this level truly comprises usability, although the usability expert's influence and value can span the entire UE tree):
   1. It’s there (the sought information or service is on the site)
      1. Welcome blurb
      2. Questions are answered
      3. Contact information easily accessible
   2. They can find it…
      1. …by searching
         1. Search results get the job done
         2. Search on all pages, with box and button
      2. …by clicking (browsing)
         1. Links are clear
         2. "Utilities" are easy to find
         3. Effective 'click tree'
         4. Logo in top left, linked to home
         5. Visual representation of the information hierarchy
         6. Conceptual flow from upper left to lower right
         7. Simple, outline-like site map
         8. Primary navigation is obvious
         9. Secondary navigation is obvious
         10. Breadcrumbs as links
         11. "You Are Here" indicator
         12. Plain wording (no jargon)
         13. Visited pages are distinguished by link color-coding
      3. …even with accessibility constraints
         1. Alt tags used well
         2. Links don't just say "Click Here"
         3. A style sheet (CSS) is used.
         4. Text sizes are "relative"
      4. ...quickly
         1. Graphics file size doesn't slow navigation
         2. Intro panel or animation not excessive
         3. Concise wording (no extra wording)
         4. No 'happy talk' (interfering sales talk)
         5. Graphics used only for core message

C. After Use:

1. Effectiveness (did the visitor or the publisher accomplish their goal?)
2. Delight, pleasure, a sense of fulfillment (the overall result makes the visitor want more)
3. Branding (customer’s recognition of the organization is reinforced)

Usefulness and effectiveness are very special items as regards usability, and related to each other. If someone asks me to craft an interface for a product that calculates how many old-fashioned typewriter ribbons are needed to type a document of X words, who am I, a usability person, to comment on its usefulness? Perhaps the requestor has an awesome need for it. Of course my design would support both cloth ribbons, which can be reused, and mylar Selectric® ribbons which can only pass once through the typewriter. If I failed to include such functionality, the program might be ineffective(!). Notice I've chosen to say it's the program, not the interface that is ineffective.

If functionality is undesired  (not useful) by anyone, or its results, irrespective of how hard they were to achieve, do not answer the intended need (not effective), the issue is not one of usability. Note that I do consider usefulness and effectiveness to be our responsibility, but only in a paternal way. For instance, if I write a lengthy usability report for a useless, ineffective product without pointing it out, I have been irresponsible, but to a higher goal that is subtly outside the scope of usability; I would be hypocritical. As you see in the tree, I suggest that they have a timing relationship to usability; however critical, they are not technically the deliverables of a usability person.

Branding and credibility are valid but they are publisher values, not customer values! They are things the vendor wants to do to your mind. If I create a virus that brings up my company logo every time you touch your keyboard, I've accomplished branding, like it or not, but I assert that there's no tie-in with usability. Similarly, integration with offline channels is a crucial component of total usability and an element of effectiveness but is similar to effectiveness in that usability only delivers the slightest portion of if.

So why do we build such a tree, and now that we have it, what do we do with it? I don't know why I built it. I think it's an irrational compulsion to look for rational relationships... find patterns, a higher sense of order, to solve a puzzle. I often think of the 20-page checklist from Deniese Pierotti. It has an exhaustive list of every atom of usability, grouped to one level. It's a wonderful tool, but seems to need something more. I suppose that in identifying the superstructure (the molecular structure?) the path to the goal is more clear. For instance, we can more clearly judge the impact of omitting particular objectives: if we omit every item in a node, it is clear what we are really giving up.

And notice the node, B.2.2.2, "browsing": this is where almost all Web dogma occurs, and no wonder. Look how many items it takes to offer good browsing. Should we stop spending so much time and effort on browsing, then? No... we should recognize and accept that the richness of those items represents the manifold power of a great website and how the Web has elevated research so far, so fast. A great site builds on the 1500-year old affordances of the printed world, exemplified by items such as an encyclopedia or The Yellow Pages, and adds the database of the entire world. It's no coincidence it needs a lot of nuances to afford all of the power... it just takes a lot of vigilance—and a little bit of new expertise—to put them all on every page.

jb August 2, 2003