Transactional Email and Confirmation Messages

Jakob Nielsen's Alertbox for October 20 is now online at:
  > http://www.useit.com/alertbox/confirmation-email.html

Summary:
Automated email can improve customer service, strengthen relationships,
and help websites bypass search engines. But most messages fared poorly in
user testing and didn't fulfill this potential.
		
----------------------------------

User Experience 2008 conference

  > Chicago, November 2-7
  > Amsterdam, November 16-21

In-depth training:
  > 31 full-day tutorials

Keynote presentations:
  > Don Norman: Sociable Design
  > Chris Nodder: Usability for Evil

Full program:
  > http://www.nngroup.com/events

DEADLINE to save $200 on Chicago registration: NEXT FRIDAY, Oct. 31

----------------------------------

RESEARCH UPDATE 2008 TOPICS ANNOUNCED

We made our way through more than 1,000 research papers published during
the last year to select the 40 most relevant ones for presentation at the
November conference.

About half the day will be spent on Web 2.0 and social uses of the Web,
because there are now finally actual empirical research findings to tell
us how users behave with these designs. At long last we can move past the
hype and speculation that characterized discussions of Web 2.0 in previous
years. The facts are in.

Topics for 2008:

  * Web 2.0:
    - Participation incentives for Web 2.0 systems
    - Social Network Sites: user-centric and topic-centric;
      why people use these sites and what they do there
    - Social Networks in the enterprise
    - How to exploit social networks to provide more
      relevant information to your users
    - How to use rewards and status to boost participation
    - Reviews and ratings: why people post reviews;
      the rise of star reviewers; rating user comments
    - Question and answer systems: is it better to pay
      for answers or to get them for free?
    - What questions get more answers and how expert are
      "experts" on community q&a sites?
    - How early technology adopters use Web 2.0 systems
      to make sense of information
    - Blog design: the effect of post length

  * Security:
    - The problem with passwords
    - Are graphical passwords better than alphanumeric ones?
    - Security questions: when are they safe?
    - How can we use security questions to make access
      control more user-friendly?

  * New usability methods and techniques: 
    - How to use Mechanical Turk to collect user data
      reliably and cheaply
    - Is remote usability testing better or worse than lab testing?
    - What's the most efficient web layout of questionnaires?
    - How the combination of task coverage, number of users,
      and participant recruitment impacts the number of
      usability problem discovered in a study

  * Search:
    - What do people look at on the search results page?
    - Are longer description snippets better than short ones?
    - What kinds of user behaviors do users most often
      engage in on websites?
    - What sites get revisited and how often?
    - How do users get to those sites? Problems with the back button
    - Why do users monitor websites?
    - What kinds of monitoring tools are most effective?

  * Mobile:
    - What are the top non-voice activities?
    - What is mobile Web used for?
    - What are the information needs of mobile users?
    - How people search on the mobile Web
    - When and why people watch video on mobile devices
    - How to do mobile usability testing

More on this seminar:

  > http://www.nngroup.com/events/tutorials/research_update.html

----------------------------------

CLUENESS ABOUNDS

Sometimes I wonder why I keep doing it. For example, repeating the advice
to "watch what users do; don't just listen to what they say." I discovered
this insight 25 years ago, so doesn't everybody know it by now?

But then I get out among real people and rediscover that most Internet
professionals still don't have the faintest clue about usability.

Last week I was at a conference where one speaker presented an overly
fancy website built completely in heavy Flash that constructed a virtual
environment for users. The design looked great in the demo, but one could
wonder how well it would work with real-world users.

I didn't even have to ask the question whether this design had been
subjected to user testing: another member of the audience did it for me,
which gives me hope for the future of the industry.

The answer was no - no user testing of this flashy design.

I could have guessed as much, but the speaker's arguments for skipping
testing shows that we still have far to go in explaining the facts of
usability.

(a) There was no time for testing because of the compressed development
schedule of the virtual world. By the time everything had been programmed,
it was time to launch.

(b) In any case, focus groups don't have much credibility. Even if they
had run one, participants in that group might have given them one set of
feedback and then after launch another focus group could have said
something completely different. Better to just rely on the design team's
own experience than to listen to customers who may change their mind.

Both arguments are fallacious:

Re. (a): Apparently this Flash guy has never heard of paper prototyping.
It's easily possible to mock up major parts of the interaction flow and
get user feedback in a few hours without writing a line of code.
  > http://www.nngroup.com/reports/prototyping

While paper prototypes can be used to test the benefits of the big
picture, they are less useful for the intricacies of low-level GUI
interactions, such as exactly what happens as you move your mouse in a
certain way. But these smaller issues can be prototyped in smaller pieces
of code and tested separately, long before the full application is done.
(Just as long as you also test the big picture as a paper flow.)

Re. (b): Any time somebody uses the term "focus group" when discussing
usability, you *know* they are clueless. User testing is conducted on a
1-on-1 basis, one user at a time. It's not a matter of opinion, but of
user experience. What happens *during* the interaction, as the user is
engaging with the design. It's not a matter of discussing the design based
on a demo (where opinions can indeed change), but a matter of observing
use, interaction, and engagement.

----------------------------------

SCANS OF BRAIN ACTIVITY WHILE USING THE WEB

(I would have used the headline "This is Your Brain on Websites", except
that it's too cute and not direct enough for our newsletter writing
guidelines.)

Collecting MRI scans of older users while they search the Web shows large
amounts of activity in multiple areas of the brain that control decision
making and complex reasoning.

  > http://news.bbc.co.uk/2/hi/health/7667610.stm
	
In press reports, this finding is getting the positive spin of keeping our
brains fit and ship-shape. No doubt there's something to this, because the
"use it or lose it" effect is well documented in terms of what happens to
the brain as we get older.

But there's the opposite spin as well: the brain scans also confirm what I
have been saying for a long time - that it's extremely taxing for our
cognitive resources to attempt to use the Web at its current level of
abysmal usability.

As also shown by the previous finding that user performance degrades by
0.8% for every year you get older:

  > http://www.useit.com/alertbox/middle-aged-users.html