Marketing & SEO Discussion List - LED Digest

Home arrow Full Issues arrow 2007 archives arrow LED Digest 2502: Styling Visited Links
LED Digest 2502: Styling Visited Links Print E-mail
==================================================
                 The LED Digest
             Moderated Discussion List
     "Effective Online Advertising, Since 1997"

      Data > Information > Knowledge > Wisdom

www.GetWebContent.com/LED : the LED's Key Sponsor
 The Web's Most Experienced SEO Content Providers.

www.SEOToolSet.com/training/ : the LED's Premier Sponsor
Bruce Clay's Search Engine Optimization Training & Certification

==================================================
List Moderator:                       Published by:
Adam Audette                          LED Digest
adam, led-digest.com     http://www.led-digest.com
..............................................
September 28, 2007                   Issue no. 2502
..............................................


            .....IN THIS DIGEST.....


====== NEW ======================

        --== Google Selling "JumpStart" ==--

                ~ Eran Adams
"Google has been contacting me and trying
to sell their new 'JumpStart,'..."

                <Moderator Comment>


==== CONTINUING =================

        --== Web Stats ==--

                ~ Michael Martinez
"The worst services to use are the ones
that rely on Javascript..."

                ~ Mark Bishop
"I've been consolidating my sites all
to Google Analytics."

                ~ Terry Riley
"StatCounter.com has everything you
are looking for."

                ~ Al Toman
"Google Analytics is cool, however, may
be confusing."

        --== Site Usability Mistakes ==--

                ~ Greg Thibodeaux
"...the average web user just doesn't know
about the visited link convention..."

                ~ Alicia Lane
"I'm interested in how other designers deal
with this common dilemma."


=========== NEW ==================================

From: Eran Adams
Subject: Getting Sold "JumpStart" by Google?

Hi all,

Google has been contacting me and trying to sell their new
"JumpStart," which is about designing and launching a AdWords
keyword campaign for my website. Anybody have a take on this one?

Eran Adams
http://www.magnetic-products-store.com

<Moderator Comment>

JumpStart was launched sometime in 2004, so it's been around awhile.
I don't know how successful it's been... but I checked the link for
Google's JumpStart here:
https://adwords.google.com/select/jumpstartwelcome and saw that it's
only targeted to people in China now.

Eran or other LEDers in China -- can you visit that page and either
get us a screenshot or tell us what the info is?

I tried using the anonymous proxy client Tor http://tor.eff.org/ but
didn't have any luck either (Google knows me too well I guess).

-Adam


======== CONTINUING ===============================

From: Michael Martinez
Subject: Stats

> I've heard about Google's web analytics, John Smart
> just mentioned WebTrends, my old techie suggested
> WebStat... I'd like to hear what you think the best
> options are, and why.
        - Tom Anson, LED Digest 2501
        - http://www.led-digest.com/content/view/1913/190/

Every third-party tool that requires you to place code on your pages
will under-report your traffic for a variety of reasons.  The worst
services to use are the ones that rely on Javascript (which includes
nearly all of them now).  Javascript is problematic for several
reasons.

First, not everyone uses Javascript (although recent statistics
indicate that more than 90% of all users may now be surfing with
Javascript active).

http://marketshare.hitslink.com/report.aspx?qprid=3

http://www.thecounter.com/stats/2007/September/javas.php

http://www.webhits.de/webhits/browser.htm

Secondly, where the Javascript is placed on the page has a direct
correlation to how many missed hits the service overlooks.  The
lower down your page, the more likely the service won't count a
visitor who clicks in and out quickly.

Server connection times (user servers, your servers, and third-party
stat servers) can also kill the counts.

There are also occasionally server down times on the third-party
services.

If your server collects raw log data you can use one of the popular
Web statistics packages but the amount of hotlinking to images,
scraping by screen readers and scavenger bots, and experimental
crawling by new search engines and research tools can greatly
exaggerate your visitor counts.  You need to manually edit your
configuration files.

But that's not all.  Your site may serve page types that are not
correctly defined in your Web stat package configuration files, so
you may have to subtract some page types and add some page types.

And then the default report formats usually don't provide very good
information (usually just top 10 or top 30 listings). If you cannot
get to everything in a category, the package does you no good.

Most if not all of the free packages don't seem to do totals.  And
the Perl-based analytics packages can choke on high traffic sites
because they corrupt their data files or run out of disk space.

You won't see much if any data normalization, either.  Two query
referral strings, for example, will be treated as totall separate
and distinct queries even if they only differ by a space.

And the statistics packages may aggregate referrer URLs and IP
addresses.  You may or may not get reverse domain-name lookups.

In short, there is nothing out there which does everything for free
and not much out there that I would trust farther than I could throw
it.  Where necessary I work with free Web stat packages knowing that
I have to live with inadequate and usually wrong data.

Through the years people have sworn by their favorite packages.
Every one I have tested has come up short.  When I can go into the
raw server log and see data that is not present in the reports, I
know the Webstats packages are not doing their jobs.

I have not tried every statistics package but I have used a lot of
them.

They all have short-comings and deficiencies.  They are best used
for trend analysis but you need to collect at least a year's worth
of data to start looking at trends (for most sites -- some verticals
and sites can see trends in smaller time frames).

Good luck.

Michael Martinez
http://www.michael-martinez.com/


-------- new post - same topic ---------

From: Mark Bishop
Subject: Stats

> What I'm looking for is something that is free
> or very inexpensive. I'd like it to distinguish between
> unique visitors and returning visitors.  If it could
> tell me referring page and keyword choice, that
> would also be good.  If it tracked how long a visitor
> stayed on the page and where he/she went, that
> might also be good.
        - Tom Anson, LED Digest 2501

I've used probably 10 different stats programs over the past 5
years. I also prefer free, simple, and easy to use. While many have
strengths and all have weaknesses, I've been consolidating my sites
all to Google Analytics. For me it gives me:

1) Easy to implement - just include the GA code into bottom of the
every page on your site (Or at least every page you want to track).
If you have an external footer this is very simple.

2) Easy to understand - the summary pages alone offer rich and easy
to read information with graphs and charts and are highly visual.

3) All the details you can wish for. In a simple way, you can view
keywords, unique visitors, sources, length of stay.

4) I track five websites and they are all on one place. Very easy to
switch back and forth between the sites.

5) When you have time, you can dig deep into the data and break the
data down in almost any way you'd like. When I do this, I always
find new features and details that I didn't know I could get.

I'm not sure how it tracks repeat visitors to tell you the truth.

Good luck

Mark Bishop
http://www.workforcelang.com
http://www.wlstranslations.com


-------- new post - same topic ---------

From: Terry Riley
Subject: Stats

Tom,

StatCounter.com has everything you are looking for.

Terry Riley
travelfox.com


-------- new post - same topic ---------

From: Al Toman
Subject: Stats

Tom,

Statcounter is a well recognized stat service.  Their offerings
range from free to paid. They are referenced in BusinessWeek:
http://www.businessweek.com/globalbiz/content/jul2007/gb200...

http://www.statcounter.com/

Also, Google Analytics is cool, however, may be confusing.
http://www.google.com/analytics/

Al Toman
studio9 web design


========= Begin Sponsor Message =========

Why pay more for less?

Some content providers charge extra to search-engine
optimize copy. And some are so clueless, they don't
offer SEO at all. A lot of good that'll do your site.

At GetWebContent.com, we SEO everything we write
without any additional charge whatsoever.

Get it all, www.GetWebContent.com/LED

========== End Sponsor Message ==========


-------- new post - new topic --------

From: Greg Thibodeaux
Subject: Usability

[On the topic of making visited hyperlinks a different color than
unvisited hyperlinks...]

> Good, user-friendly, accessible Web sites tell
> their users where they are, where they've been
> and where's left to go. They don't keep them
> in an endless cycle of 'Did I click on this yet?'"
        - Lisa Barone

> This is a great point -- yet many sites
> don't follow the rule.
        - Adam Audette

> Perhaps there are so many who will just be confused
> by the changing colours that it is actually better to rely
> on viewers to know what they've already read. (Radical
> idea!) I would be very interested to know if anyone has
> hard numbers on how many Web users actually
> know this convention, today.
        - Michael Linehan

I have to say that I completely agree with Michael on this issue and
have come across this issue many times. I would assert that the
average web user just doesn't know about the visited link convention
and is more likely to be confused by the changing of link colors,
maybe even more so  if they don't remember where on the site they'd
already been.

I would definitely like to see some recent usability testing figures
on this....

Greg Thibodeaux


-------- new post - same topic ---------

From: Alicia Lane
Subject: Usability

> I am constantly staggered at the numbers of people who:
> 1. Put the url that they already know in the search box - go
> to Google - and then click through to the site - IF it has
> come up in the results!!!
        - Michael Linehan

I have noticed this too myself and have wondered if it has more to
do with the visitor's browser setup. Many people have their default
browser home page set to Google or some other search engine. I have
to wonder if many (especially aging baby boomers with decreased
visual acuity and fine motor skills) just find it easier to click on
that nice big search box in the middle of their screen, rather than
navigate ALL the way to the top to that wisp of a browser address
field. Sure, it's an extra step, but both steps present them with
nice fat objects (search box then a nice big link) to click.

Or, if they are using their browser's search box on the right as you
suggested, perhaps it's out of force of habit. That, and the mouse
cursor tends to be closer to the right side of the screen since most
people are right-handed.

> I've had numbers of clients say, "Just one last thing
> to fix before we put it up. Some of the links are a different
> colour."  And these are educated, capable businesspeople,
> who I certainly expected to be fairly knowledgeable.

This scenario is actually the first thing I thought of when this
visited / unvisited topic came up. In principle, the idea is good.
But in practice, we do run into resistance from clients. Why? My
theory is that clients surf their own site with mental goals very
different from that of a typical visitor. Clients are already
familiar with their own business, so their goals usually involve
clicking through the site as quickly as possible looking for errors.
They're not paying attention to what they've already seen, because
they've already seen it all 100 times. So the feedback from the
visited / unvisited links doesn't even register with them,
until...hey, what's this? The links are a different color? !@#$%
designer!

I suspect that this happens even if they subconsciously benefit from
this link color feedback on other sites. It's all about context.

So, I'm interested in how other designers deal with this common
dilemma. Do you explain up front, or do you have a favorite method
of quelling client concerns over your "incompetence"?

Alicia Lane


-------------------------------------------------------
The LED Digest is sponsored by:

GetWebContent.com
The Web's Most Experienced SEO Content Providers.
Free no-obligation proposal: http://GetWebContent.com/LED

SEOToolSet.com
Bruce Clay's Search Engine Optimization Training & Certification
Join the certified SEO directory: www.SEOToolSet.com/training/

The Archives: http://www.led-digest.com/content/view/126/189/

Subscribe: http://www.led-digest.com/content/view/52/187/

Unsubscribe, Change Email, or Hold / Resume Delivery:
http://www.led-digest.com/content/category/4/17/201/

(c) Copyright 1995-2007 Orange Wheel, LLC. All Rights Reserved.
-----------------------------------------------------------------

"A will finds a way." - Orison Swett Marden