Google Analytics Report: Q1+Q2 2011
David Frossard
1 July 2011
Data from the period 1 Jan 2011 – 30 June 2011

There were 850,559 visits to the site in the first six months of 2011 versus 600,228 for the same period
in 2010 – an increase of more than 1/3.

The site received 25,000 to 45,000 visits per week during the first six months of 2011, dropping by about
half during summer break and major holidays. Page views tend to peak at the beginning and end of a

Visits to peak each week (typical y on Monday, Tuesday, or Wednesday) at around
8,000 per day, typical y fol owed by a 2/3 or greater reduction on weekends.

Weekly minima are fairly consistent throughout the entire year to date. We seem to have a year-round
minimum – a constant audience – of at least 2,000 page views per day.

Visitors to access an average of 2.4 pages per visit.

Most pages typical y show bounce rates1 and exit rates2 of roughly 50 percent. Anything much higher or
lower is considered anomalous here.

The page /Campus-Email primarily contains just two links: to MyMail and Exchange Mail. That so many
users are coming here simply to access email indicates that they don't know how to access their mail
otherwise – either through a separate email client or by directly bookmarking the appropriate web page.
Could we add text here to educate users on more direct ways to access email? Bounce rate is 82
percent, exit rate is 79 percent.

The page /Blackboard_1 has very high bounce and exit rates, nearing 90 percent. Unlike the page
noted above (which has understandably high bounce rates), this one is quite complex. What do we
make of these high rates? Are people solving their problem and exiting, or giving up in frustration?

Visitors apparently find the top-level links to Academic and Administrative departments useful, and
access them often. (See below for more on where visitors go next.)

Campus Bul etins are very often accessed online. The school might consider phasing out paper copies
of these documents.
The percentage of visitors who leave the site (“bounce”) rather than clicking on links on the entry page.
The percentage of visitors who exit the site from a particular page, after perhaps having browsed the site already.
Page 1 of 10

What do people do after they click the Administrative Departments link at

After clicking on that link, 9.8 percent then go to Human Resources.

In second place, 9.3 percent immediately exit back to Would this be considered
relatively insignificant? If so, this page seems to be doing what it was meant to do.

Strangely, 6.3 percent immediately click one of the Administrative Departments links on that page, ended
up in exactly the same place.

Perhaps having picked the wrong original link, 3.4 percent immediately click Academic Departments

The most popular Administrative Department destinations after HR are Finance and Administration (4.5
percent), Academic Affairs (4.5 percent), and the Graduate School (3.6 percent).

Not surprisingly, the Student Life link is much more frequently accessed when students are in residence.
What do people do after clicking the Academic Departments link at

Inexplicably, 44 percent immediately click on the Academic Department link on the new page, returning
them to exactly the same page.

Twenty-six percent return to the home page. What are we doing wrong here?

Eleven percent click the Administrative Departments link. Is the difference between Administrative and
Academic departments unclear?

Smal er numbers then navigate to the Registrar's Office (1.6 percent), Military Science (1 percent),
Computing (0.9 percent), or Forms (0.9 percent).
Visitor Loyalty
While about 1/3 of al visitors are absolutely new to the site, and just over half have visited it
less than a dozen times, approximately 40 percent have visited the site more than 25 times. This implies a
captive audience making use of the site very frequently (or at least setting it as their home page), but a larger
audience of very new, or first-time users. Does this imply that “re-skinning” or otherwise redesigning should be a low priority? If we believe that captive users appreciate consistency and familiarity
more than new graphics, that would imply a low need for a redesign. But if, in fact, most users are new and
seeing the site for the first time, a redesign would seem to be even less needed.
Depth of Visit
How many pages are viewed during the typical visit to pages? Over half of visitors view just
one page during their visit – perhaps implying that they hit their target on the first try. If so, this would imply that
users of know where to go to find the information they need. More than 90 percent of visits end
– one way or another – after viewing five or fewer pages. Could it be that the site is more efficiently designed (or
Page 2 of 10

at least, better understood) than we sometimes believe it to be? Or are people just giving up in frustration?

More than 96 percent percent of page requests come from English speakers (or, at least, “browsers
configured in English”), the vast majority of those coming from within the United States, with a few more
from Great Britain and Commonwealth countries.

In terms of visits by country (see below), the U.S. accounts for almost 92 percent, fol owed by China,
Canada, and India – each with less than 1 percent of total traffic. (The number of international accesses
to is approximately double, as noted below.)

Other requests came from browsers configured for mainland-Chinese (0.88 percent), Spanish (0.67
percent), French (0.27 percent), German (0.21 percent), and Japanese (0.15 percent). However, since
we don't have any alternate-language capabilities on, one would assume that those
accessing the site would general y know English to read it, even if their browsers were set for other
languages by default. In any case, there would seem to be little need for second-language features on at this time, unless there is an institutional desire to increase visitation by individuals of
particular countries or language groups.

Approximately 40 percent of page visits originate in Golden. About 72 percent are from Colorado in
general, with California (3 percent) and Texas (2 percent), with other states and almost 200 countries
and territories distant runners-up.
Page 3 of 10

Browser Statistics
What browsers are used to access In order of popularity:

Internet Explorer (al versions): 42 percent. Within that figure, 82 percent used version IE8, 12 percent
IE7, 4 percent IE9, and 2 percent IE6 (which we should no longer support).

Firefox (al version): 25 percent. Approximately 65 percent of those used a version of Firefox in the 3.6.x
family and 19 percent 4.x versions. Firefox 5 was released 21 June 2011, but did not figure in the tal y.

Safari (al versions): 16.92 percent.

Chrome (al versions): 15 percent.

Al other browsers: negligible. Even Opera barely figures, though it is slightly more popular in Asia and
OS Statistics
What operating systems are used to access

Windows (al versions): 76 percent.

Mac OS X (al version) 19 percent.

Linux (al versions): 2 percent.

iOS (al versions): 2 percent.

Android (al versions): 1 percent.
Note that while Windows is highly popular with our web users, Internet Explorer is proportional y less so. In the
past year, according to industry figures, Internet Explorer and Firefox use rates have stagnated while the big
winner has been Chrome.
If, as our Server Team has asserted through their own forensic exercises, more than half of incoming Mines
students use Mac OS on personal computers, this would imply that access to is predominantly
not by students. Do faculty and staff – though greatly outnumbered – actual y use far more
heavily than students do? (Or could the Server Team figure be wrong?)
Mobile Statistics
In the past six months, there were 24,165 visits from 13 mobile operating systems – a vastly smal er number
Page 4 of 10

than visited using a laptop or desktop computer (just 1-2 percent of desktop numbers, in fact). On peak days,
approximately 300 mobile requests from al devices were recorded. However, since we have no particular mobile
presence, that may reflect either lack of demand for a mobile website, or lack of a mobile website. Here are the
top 10:
What is the trend here? In the same period (1 Jan – 30 June) of 2010, there were 5,937 visits to from mobile devices. This year, the number grew to 24,165 – a 307-percent increase year-to-
year. By contrast, total visits to rose from 600,228 in the first six months of 2010 to 850,559 in
2011 – a 42-percent increase in that same period.
A 42-percent increase in web visits one year later is a significant number, one that should have us ponder our
future infrastructure needs. But a 307-percent increase may imply that urgent attention to the mobile arena is
needed. However, a look at the trend between November 2009, when mobile devices started accessing, and 30 June 2011 shows that part of the explanation for the huge increase is simply that
mobile devices started from such a low base. Nevertheless, the trend is strongly upward.
Over the past 18 months, visits by Android devices have almost doubled on, while Blackberry
access was halved. Access by iPhones has dropped by about 20 percentage points – almost exactly offset by a
20-percentage-point increase in iPad access. Over the medium term, it appears that iOS (whether on handheld
or tablet devices) and Android (on phones and tablets) will be the dominant mobile operating systems. If Mines
releases any mobile apps, they should be in this priority: (1) iOS for iPhone, (2) Android, (3) iOS for iPad. A
highly compatible website would serve al these and more and should be our first priority in a mobile presence.
Over the longer term, Hewlett-Packard's WebOS (acquired with HP's acquisition of Palm) may make inroads.
Windows Phone OS also shows slight upward growth. The big loser – particularly as iOS gains cloud access in
iOS 5 and shows that it is entirely sufficient for enterprise use – appears to be Blackberry.
Page 5 of 10

Of note: 48 percent of mobile visitors are new visitors to, compared to 34 percent of desktop
and laptop visitors. New visitors more often use mobile devices to get their first taste of,
implying a trend toward mobile devices over time.
Apple offerings – iPhone, iPod, iPad – made up almost exactly 2/3 of mobile visits. Android devices accounted
for 30 percent, and Blackberry a (declining) 2.4 percent. SymbianOS (Nokia smart phones), Windows phone OS,
and other mobile operating systems were negligible. Since iOS devices do not support Adobe Flash, this
obviously implies that no Flash should be used on a future mobile website.
Note that iPhone, iPod, and Android access to this site is more than triple the access recorded for the iPad.
Since most Android devices in this time period were in handheld form factor, it would appear that handheld
mobile access is stil far more likely than tablet access. We'll see what the trends look like in coming months and
years. But, for now, a mobile web site optimized for handhelds would seem to have some value.
This is a wel -used site, visitation increasing by 1/3 in a year. We should monitor the trends to make sure our
hardware and infrastructure remain adequate to the task.
Page 6 of 10

Data from the period 1 Jan 2011 – 30 June 2011

Though it is a more special-purpose, limited, seasonal site, received more than half a
mil ion visits in the first half of 2011, compared to 850,000 for That seems to me to be
a remarkably robust showing for

The general visitor trend is downward during the first six months of the year, from around 30,000 per
week in early January to around 15,000 per week at graduation. No doubt that will pick up in the second
half of the year as it did in 2010, where it hit a yearly high in September, fol owed by gradual drops in
October, and November.

Visitors to look at an average of 3.2 pages per visit, compared to 2.4 for, and spend slightly more time on the external site as wel .
Where do visitors go after accessing
These are the 10 most popular pages:
These particular pages may be of significance as we develop our mobile website. While the mobile site is also –
or perhaps primarily – for current Mines students and employees, some attention might be paid to recruitment.
Current mobile plans do not specifical y address any of these target pages.
Page 7 of 10

Visitor Loyalty
When compared to, almost half again more visitors to are visiting for the first
time. At the other end of the scale, only half as many (as a percentage) have 200+ repeat visits – perhaps for
news and events, or just as a home page? On, we get “more chances to make a good first
impression” (to paraphrase the advertising axiom). If we believe that we can do better recruiting with a redesign,
it should be pursued. However, there is a lot of new traffic on (from people to whome the site is
stil “new”), somewhat undercutting the urgency of a redesign.
Depth of Visit
The number of pages viewed on a typical trip to is only slightly fewer than the number of pages
viewed by a typical visitor to On, 84.3 percent visit five or fewer pages on a
visit. On, the figure is 91.9 percent.
While 85.5 percent visit are stil from within the United States, the percent of visits from other
countries is almost double that of Some extra language features – such as welcome
messages in Chinese or Arabic, say – might aid slightly in recruiting.
Page 8 of 10

Browser Statistics
Do visitors to (many of them, presumably, prospective students) and
(presumably mostly current students and employees) use different browsers? For the most part, no. Browsers
used to access do so in approximately the same proportions as browsers used for Safari is used slightly more often on and Chrome slightly less. We need to
support IE, in al its strangeness, but, in fact, we need to support Firefox, Chrome, and Safari as wel . There is no
“insignificant” browser in this group.
OS Statistics
The same operating systems, in mostly the same proportions, are used to view and
Interestingly, iPhone and Android access outpace Linux on this site (not the case for Thus, we
might postulate a slight tendency toward increased Linux use associated with increased exposure to Mines. That
would be an example of an interesting, but irrelevant, hypothesis generated through the use of Google Analytics.
Mobile Statistics
In the first six months of 2011 there were 25,313 mobile visitors to – a slightly larger number
than seen on, despite the fact that received 1/3 fewer visitors overal , from al
sources. If in fact visitors to are younger visitors than to (that is, if they are
typical y teenagers – prospective students), this could foreshadow an increasing demand for mobile access to
Mines web information in the future.
Mobile Devices
Visitors accessed with the same mobile devices, used at approximately the same rates, as
those who accessed (see, above). The trend is also strongly upward since November 2009,
having quadrupled in that period.
Page 9 of 10

Though it is a much smal er and more specialized website, of interest primarily to the outside world, rather than
to current students, it is stil very successful and much visited. However, year-to-year visitation has increased
only around 8-10 percent, far less than the 1/3 gain of its sister site. Current infrastructure for is
likely to work longer than for the other site.
Page 10 of 10