Write Articles, Not Blog Postings

[email protected] (Jakob Nielsen) Mon, 09 Jul 2007 06:10:00 -0700
Newsgroups gmane.comp.web.alertbox
Message-ID <SPARKLIST-7313542-6538402-2007.07.09-06.10.01--gcwa-alertbox#[email protected]>
Jakob Nielsen's Alertbox for July 9 is now online at:
  > http://www.useit.com/alertbox/articles-not-blogs.html
		
Summary:
To demonstrate world-class expertise, avoid quickly written, shallow
postings. Instead, invest your time in thorough, value-added content that
attracts paying customers.
		
----------------------------------

User Experience 2007 conference

  > Barcelona, November 4-9
  > Las Vegas, December 2-7

31 full-day tutorials
2 keynote presentations: e-commerce, the state of usability

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


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

INTRANET DESIGN ANNUAL – CALL FOR ENTRIES

Think you have a good intranet? Enter it for the annual award competition
to identify the year's 10 best intranets.

Deadline TODAY

Info on what we need and where to send it:
  > http://www.nngroup.com/reports/intranet/2008

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

URL as UI

Edward Cutrell and Zhiwei Guan from Microsoft Research have conducted an
eyetracking study of search engine use that found that people spend 24% of
their gaze time looking at the URLs in the search results.

MSR used Microsoft's own search engine (fair enough), but their results
match what we found in our eyetracking which included the current market
leader as well as the #2 search engine in addition to Microsoft.

Users have evolved a firm model of search behavior which they apply across
search engines, which is why it's probably a lost cause to make a
non-standard search user interface.

Searchers are particularly interested in the URL when they are assessing
the credibility of a destination.

This finding confirms the importance of human-readable URLs, which I
discussed 8 years ago:

  > http://www.useit.com/alertbox/990321.html

Another interesting finding from the MSR paper is that longer summaries
(or snippets, as they call them) are useful for informational queries,
whereas short summaries are best for navigational queries where users are
looking for a pre-determined site or page.

Web-wide search engines may have trouble distinguishing between the two
types of queries, though of course they should try to use the best style
of summary for each.

You should be able to do better with the search engine for your own
website or intranet. You should know enough about your own keyword space
to determine what words are used to look for a specific product
(navigational query) as opposed to looking for answers to a question or
researching an issue (informational query).

Full academic paper (warning: PDF):

  > ftp://ftp.research.microsoft.com/pub/tr/TR-2007-01.pdf

Analyzing the search logs from your own site:

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

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

DON'T BUY AN iPHONE

Buried toward the end of the New York Times review of Apple's new mobile
phone are the following damning statistics. Time to download homepages of
some websites:

  * NY Times: 55 seconds
  * Amazon: 100 seconds
  * Yahoo: two minutes (!)

The iPhone reviewer also experienced several dropped phone calls.

(Readers outside the United States: you may remember the concept of
"dropped calls" from the mobile services you used in the late 1980s.
Sadly, they are still a fact of life on the lower-quality networks in the
U.S. And the iPhone's network is reportedly the lowest-quality of them
all.)

Advice for consumers:

Don't buy an iPhone - or any Internet phone - that doesn't run on a much
better network. It doesn't matter how good the on-screen user interface
is. What's important is the *total* user experience which is ruined by
minute-plus response times and dropped calls. The best mobile browser is
useless when pages arrive at a snail's pace.

Advice for businesses:

It's still not time to launch a mobile Internet strategy in the U.S. Wait
until we catch up with Europe and Asia in mobile connectivity.

(Disclosure: we did work with one client on making their service mobile,
but that was a special case where mobile access made immense sense. Thus
it was correct for this company to pioneer a mobile design and to do the
extra work to make sure it was as usable as possible within the current
technology limitations.)

Eventually, it will be time to add mobile components to websites and
intranets, even in the United States. By then, you'll need a separate
design for mobile devices.

The iPhone has 20% of the pixels on a small computer monitor. Yes, this is
twice the pixel count of the previous generation of smartphones, but still
too little to support the interface-rich design style that dominates
current websites.

When I analyzed leading companies' homepages, I found that only 20% of the
screen real estate was allocated to content of interest to users:

  > http://www.useit.com/alertbox/20030210.html

Interestingly, 20% is exactly the amount that will fit on an iPhone. So a
different design that stripped away all the overhead would deliver the
same info for mobile as for the desktop.

The impact of the network on the iPhone's total user experience is a good
example of a general lesson: usability is a chain that's no stronger than
its weakest link. Take e-commerce: navigation, search, product photos,
product descriptions, the shopping cart, checkout, and site credibility
all need great design. If any one of these fails, then no sale.

My colleague Tog likes the iPhone, but he is commenting on its on-screen
UI, which I like as well, so we are not really in disagreement. See:

  > http://www.asktog.com/columns/072iPhoneFirstTouch.html

Remember:
 
user experience = UI + everything else touching the user

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

DIAL-UP STILL 1/4 OF USERS

The Pew foundation has released a survey of how Americans connect to the
Internet. Usually, surveys should be ignored, because they deal in
self-reported responses as opposed to empirical observations of actual
behavior. But in this case I think we can view the respondents' answers as
reasonably accurate, because all they were asked is how they connect.

  47% have broadband access at home
  15% have a dial-up connection

The remaining respondents didn't use the Internet, didn't know what
connection they have, or only accessed the net from outside the house.

In summary, broadband is now about 3 times as common as dial-up, but that
still means that 1/4 of home users are restricted to dial-up speeds.
Ignore download speed at the peril of losing a quarter of your customers.

  > http://www.pewinternet.org/report_display.asp?r=217

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