Transcript MS Powerpoint format

http://www.ukoln.ac.uk/web-focus/events/conferences/museums-web-2006/
Impact Analysis For Web Sites
Brian Kelly
UKOLN
University of Bath
Bath
Acceptable Use Policy
Recording/broadcasting of this talk,
taking photographs, discussing the
content using email, instant
messaging, Blogs, SMS, etc. is
Email
[email protected] permitted providing distractions to
others is minimised.
museums-web-kelly-2006-06 tag used in del.icio.us
UKOLN is supported by:
A centre of expertise in digital information management
This work is licensed under a AttributionNonCommercial-ShareAlike 2.0 licence
(but note caveat) www.ukoln.ac.uk
Contents
Usage Statistics
• Strengths and weaknesses
Impact Analysis
• What is it?
• Impact Analysis tools and techniques
Web 2.0 and Impact Analysis
• Using Web 2.0 tools for impact analysis
• Web 2.0 is impact
Sector Statistics
• What is it?
• Would it be useful?
A centre of expertise in digital information management
2
www.ukoln.ac.uk
Usage Statistics
Usage Statistics
Exploit Interactive example:
• EU-funded e-journal
funded from 1999-2000
• Steady growth in usage
measured by Sitemeter
& usage logs
Deeper analysis:
• Popular article on
"Are You Linking To
A Porn Site?", April
1999
Issues: Is popularity of Web site based on "wrong hits" and
growth reflects growth in Internet usage?
How can
weofaggregate
in a meaningful www.ukoln.ac.uk
way?
A centre
expertise in digitalusage
informationdata
management
3
Usage Statistics
Usage Statistics
Usage statistics:
• Can provide useful and valuable data …
• … but can also be flawed
Implications:
• Collect and use as they can be useful for
formative statistics e.g. has a new marketing
campaign worked; what keywords do users use to
find our Web site; …
But also:
• Be honest about limitations and don't over-hype
statistics
• Use as a part of a portfolio of metrics
A centre of expertise in digital information management
4
www.ukoln.ac.uk
Impact Analysis
Impact Analysis
Usage statistics:
• Data on who is using Web site and how they're
using it e.g. how many visitors are there?
Impact analysis:
• Attempting to measure the impact of the Web site
or the contents of the Web site e.g. how has the
Web site visit affected the visitors
Challenges:
• More difficult to measure than simpler 'visits'
• Requires thought as to aims of Web site
Requires
• Objective and subjective metrics e.g. 'possible'
indicators of impact
• Automated and manual measurements
A centre of expertise in digital information management
5
www.ukoln.ac.uk
Impact Analysis
Types Of Metrics
Automated:
• Usage statistics
• Link analysis
• Search analysis
Manual:
• Focus groups, surveys, etc.
Hybrid:
• Blogs, RSS feeds, social bookmarks, etc.
Let's use a case study for the Institutional Web Management
Workshop (IWMW) 2006 Web site – an annual 3 day event for
University
Webinmanagers
A centre of expertise
digital information management
www.ukoln.ac.uk
6
Link Popularity
http://www.linkpopularity.com/
Why link popularity?
• A link to my Web site is an
indication that A We b
author felt strongly about
the resource
• Links help drive traffic (and
so can boost usage
statistics)
• Links enhance Google
rating (and so can boost
usage statistics)
ItMany
Note
maythat
be desirable
you can receive
to remove
a monthly
links from
email
your
report,
ownThis
Web
site
analysis
tools
now
provide
RSS
feeds.
site,
which
otherwise
can
provide
growth
information
maymanagement
just
reflect
trends
growth
in
your site
can
help
Web
theiron
monitoring
activities
A centre
of expertise
inmanagers
digital informationin
www.ukoln.ac.uk
7
Impact Analysis
8
Search Analysis
Why search engine
analysis?
• Search engines help
drive traffic (and so boost
usage statistics)
Technorati:
• There's more to search
engines than Google
• Searches Blogs, RSS
news feeds, etc.
• Use of tags can help
searching social
networking services such
as Blogs, Flickr,
etc
A centre of expertise in digital information management
www.ukoln.ac.uk
Impact Analysis – Manual Techniques
Talking To Users
Impact analysis can be carried out by talking to /
listening to the user community:
• Focus groups
• Visitor books
• Evaluation forms
• Anecdotes
• Media watch
• …
Such techniques are well-known – but how can
technologies be used to support such activities?
A centre of expertise in digital information management
9
www.ukoln.ac.uk
Impact Analysis
Using Blogs
http://www.meanboyfriend.com/overdue_ideas/
http://j4.livejournal.com/2006/06/19/
2006/06/10_years_of_the.html
Blogs can make it
easier to gather quotes
& other impact
measures:
• Finding Blogs (e.g.
tags, Technorati)
• Blogs associated
with Web site
Blogs
provides
indication
of user
satisfaction
and of
Official
Bloggers
hasgood
proved
useful: set
a standard
for others;
impact
analysis
of event
"I've come
back
with ideas
claimed
a tag
in tag space;
shown
benefits
of trust;
… to …"
A centre of expertise in digital information management
www.ukoln.ac.uk
10
Impact Analysis
Being Open
http://www.ukoln.ac.uk/web-focus/events/
workshops/webmaster-2006/publicity/
11
The impact of ideas can
be maximising by
allowing the ideas to be
used freely:
• Open source
software
• Open standards
• Creative Commons
licence for content
CC licences for IWMW
2006 resources allows:
• Info about event to
be used by others
• Blog articles, news,
etc. to bewww.ukoln.ac.uk
syndicated
A centre of expertise in digital information management
Impact Analysis
Providing Mashups
http://upcoming.org/event/69469/
A centre of expertise in digital information management
12
You can maximise the
impact of your Web
site by allowing others
to make use of your
content e.g.
• Syndication
• News feeds
• Third party
applications
• …
Note CC licence for
event details permits
such reuse
www.ukoln.ac.uk
Building A Community
http://www.frappr.com/iwmw2006
Building a community for
your Web site can:
• Maximise impact by
allowing interested
parties to discuss
their shared
interests
• Provide you with
feedback & ideas
• Allow you to provide
targetted information
Web 2.0 services such as Frappr, Blogger, MySpace, etc. allow
Web communities to be easily set up (and may be particularly
centre
of expertise
in digital information management
www.ukoln.ac.uk
valuableA to
the
'Net Generation')
13
Sharing Resources
http://del.icio.us/lisbk/iwmw2006-parallel-kelly-phipps
A centre of expertise in digital information management
14
Social
bookmarking tools
like del.icio.us:
• Provide an
indication of
impact (others
have felt Web
page worth
bookmarking
• Support
community
building (finding
others with
similar
interests)
www.ukoln.ac.uk
Impact Analysis and Web 2.0
Maximising Impact & Web 2.0
We have seen how Web 2.0 technologies
(Blogs, RSS feeds, syndication technologies,
third party services, etc) can:
• Maximise impact by providing additional
access mechanisms for users
• Minimise resource effort needed by
making use of 3rd party services
• Be used to measure the impact of our
Web services
A centre of expertise in digital information management
15
www.ukoln.ac.uk
Bigger Picture
The Bigger Picture
Usage analysis and impact analysis aren't selfcontained, but form part of a bigger picture
User testing
Usage analysis
• Who's
accessing Web
site?
• How are they
finding site?
• What's popular?
16
• How do users
navigate site?
• What's difficult
to find?
• Have they
achieved their
aims?
Impact Analysis
• How has Web site
influenced the users?
• What do the users
find interesting?
• What changes do
users make as a
consequence?
These areas are complementary and should be addressed as
part of a strategy for ensuring user needs are being satisfied.
The strategy should be hybrid, covering software and
A centre of expertise
information
management
www.ukoln.ac.uk
processes,
realin digital
world
interaction,
etc.
Bigger Picture
Sector Statistics
Another aspect of the 'bigger picture' is
statistics for the sector:
• What are the characteristics of the sector?
• What legitimate assumptions can we
make?
• What are the benefits to be gained by
sector statistics?
UKOLN has worked with Nedstats who have provided sector
statistics based on 22 higher educational institutions who were
willing to host their (invisible) counter.
Summary based on > 25 million page views and > 4 million visitors
See <http://www.ukoln.ac.uk/web-focus/events/workshops/
webmaster-2006/talks/sidhu/>
A centre of expertise in digital information management
www.ukoln.ac.uk
17
Bigger Picture
Visitors Accepting Cookies
Overall:
Accept 95.93%
Decline 4.07%
External:
Accept 94.64%
Decline 5.36%
Internal:
Accept 97.90%
Decline 2.10%
2006 Average
Accept
96.8%
Decline
3.2%
Cookies needed to
monitor user trails
Taken from <http://www.ukoln.ac.uk/web-focus/events/workshops/
webmaster-2006/talks/sidhu/>
A centre of expertise in digital information management
www.ukoln.ac.uk
18
Bigger Picture
How do people find my site?
Three basic ways
Overall
External
Direct Entry
56.29%
39.98
External Referrer
20.93
21.62
Search Engines
28.66
38.38%
Taken from <http://www.ukoln.ac.uk/web-focus/events/workshops/
webmaster-2006/talks/sidhu/>
A centre of expertise in digital information management
www.ukoln.ac.uk
19
Percentage of visitors from search
engines
Visitors from search engines – Number of
visitors
Sheffield
Cass
UCL
Aberdeen
TVU
CCCU
Number of visitors per day
University
A centre
of expertise in digital information management
20
General
www.ukoln.ac.uk
Conclusions
To conclude:
• Usage statistics should form a part of an impact
analysis strategy
• A more comprehensive impact analysis strategy is
needed
• This should be part of a user-satisfaction strategy
• Web 2.0 technologies can help
• Embedding Web 2.0 culture (openness, trust,
etc.) can help the impact of Web sites which aim
to maximise engagement with its user community
• "Impact Analysis For Web Sites", QA Focus
briefing document no. 99, provides further advice
– see <http://www.ukoln.ac.uk/qa-focus/
documents/briefings/briefing-99/>
• Would sector statistics be useful for the museum
sector?
A centre of expertise in digital information management
21
www.ukoln.ac.uk