State of the Mines Web 2012
David Frossard
Released 24 January 2013
In 2012, the main internal campus website,, continued to grow significantly in
both visitors and pages viewed.1 Approximately 800,000 unique visitors spent time on the site in
2012.2 They returned to the site multiple times, resulting in about 2 mil ion total visits and 5 mil ion
pageviews. Visits peaked on Tuesday, August 21 (the start of the fal semester) at more than 15,000
(up from about 12,000 the previous year). About 61 percent of visitors in 2012 were repeat customers,
while 39 percent visited for the first time – to be expected in a site tailored to the internal Mines
community. Bounce rates remained at about 50 percent. These details are summarized below:
General overview of web traffic (2012)
Compared to 2011 (after compensating for missing data), visits in 2012 were up about 15 percent;
unique visitors increased about 20 percent; pageviews increased 20 percent. The number of pages
viewed per visit, average duration of visits, bounce rate, and percentage of new visits did not change
Though any steadily increasing system presents questions of sustainability in the long haul, our server
administrators believe we have sufficient hardware to sustain this rate of increase for the foreseeable
Due to a period in December 2011 and January 2012 in which no analytics data were col ected, these
numbers are certainly slight underestimates – by approximately 4 percent in 2012 and 8 percent in 2011.
A “unique visitor” is defined as a visit from a particular IP address. A “visit” is a separate accessing of the
website, irrespective of IP address. A “pageview” is a single page accessed by a visitor. “Bounce rate” refers
to the percentage of visitors who enter the site then leave (or “bounce”) rather than viewing more pages. As
a rule of thumb, 50 percent is a typical bounce rate and to be expected.

Other metrics of interest:

Preferred language of visitors (as determined by their browser language preferences) in 2012:
English (all variants), 95.3 percent; Chinese, 1.5 percent; Spanish (European and South
American variants), 0.7 percent; French, 0.25 percent; Korean 0.25 percent; Brazilian
Portuguese 0.21 percent; German, 0.15 percent. Compared to 2011, this represents a slight
(less than 1 percent) decrease in access by English speakers, but a greater than 60 percent
increase in visits by Chinese speakers, albeit from a low initial base. Other language
preferences remained more or less unchanged.

Visits by country of origin: United States, 90.1 percent; India, 0.8 percent; China, 0.8 percent;
Canada, 0.6 percent; United Kingdom, 0.5 percent. Other countries with significant visits
included Australia, Indonesia, Japan, and South Korea. These figures represent a 2 percent
decrease for US-originated visits compared to 2011. India and China increased their share of
visits by about 15 percent this year. Figures for other countries remained fairly stable.

Preferred web browser: Internet Explorer (al variants) continued its downward trend as 30
percent of users accessed with IE, compared to 39 percent in 2011. Firefox
use dropped from 25 percent to 21 percent. Safari grew its share of users from 17.5 percent to
20 percent. And Chrome was the big winner, increasing from 17.3 percent to 26 percent
year-to-year. Mobile browsers also continued their longstanding and dramatic upward trend
(see more on mobile below).

Preferred operating system: Windows (al variants) decreased slightly as the operating system
of choice for accessing our web services, dropping from 74 percent to 69 percent in 2012 year.
Mac OS X increased its share slightly, from 20 to 21 percent. Linux remained relatively
constant, dropping from 2.1 percent to 2.0 percent. The biggest increase in visits came among
mobile operating systems, primarily Android and iOS, up 75 percent.

Use of mobile devices: Mobile-device access to again increased dramatical y,
as it has for each of the past few years, from 1.5 percent of visits in 2010, to 3.5 percent in
2011, to 7.5 percent in 2012. At this rate of increase, half of visits to will be by
mobile device in just a few years. Of note: Approximately half of mobile visits to are made on phones (almost entirely Android or Apple models) and half on
tablets (again, primarily Android or Apple models). Apple phones and tablets remain the single
most popular mobile platform of our customers.
In conclusion, visits to continued to increase significantly in 2012. In part, this is a
structural increase, largely because more departments (with more web pages, and more students to
view those pages) joined the CMS in that year. There were simply more reasons for Mines students,
staff, and faculty to visit the site.
Also, visitors using mobile devices continue to increase significantly, as they have for several years,
indicating that we should re-evaluate our mobile strategy frequently. Providing content for
smal -format, mobile devices is likely to be the largest challenge to Mines web services in the next

The outward-facing marketing site also saw increased visits and pageviews in
2012, though not to the extent experienced by its sister website. Over 620,000 unique visitors
navigated to the site, with almost 1.2 million total visitors and almost 4 million pageviews.3 Visitors to
this site looked at more pages, for a longer time, and with a lower “bounce” rate than did visitors to The 45.5 percent bounce rate here is comparatively low and indicates a site that is
“sticky” – a site that has multiple pages of interest to a typical visitor. Half of all visitors were new (as
would be expected in a marketing website). The largest numbers of visitors were found on Monday,
January 9, just before the start of the spring semester, and Monday, August 20, just before the start of
the fal semester.
General overview of web traffic (2012)
However, in 2012, visits were just 2 percent more than in 2011. Unique visitors increased 5 percent
while pageviews increased 3.9 percent. Visitors viewed about the same number of pages per visit, but
for a longer time. The bounce rate and percentage of new visits decreased slightly compared to 2011.
We seemed to have reached more or less a steady state with the server, whereas is still seeing substantial increases in visits year-to-year.
Other metrics of interest:

Preferred language of visitors (as determined by their browser language preferences) in 2012:
A slightly higher percentage of international visitors is found on as compared
to English (all variants) accounts for 93.4 percent of visits; Chinese, 2.3
percent; Spanish (European and South American variants), 1 percent; French, 0.3 percent;
Brazilian Portuguese 0.3 percent; Russian, 0.3 percent; Turkish, 0.2 percent. These figures are
substantial y unchanged since 2011.

Visits by country of origin: Again, visitors to tend to be somewhat more likely to
reside outside the United States than visitors to The United States
These data are not affected by the missing-data issue experienced by, so these numbers
can be considered complete and accurate.

accounted for 84.3 percent of 2012 visitors (versus more than 90 percent on
China supplied 2.2 percent of visitors; India, 1.3 percent; Canada, 0.9 percent; the United
Kingdom, 0.7 percent; Saudi Arabia, 0.6 percent; Iran, 0.55 percent; Nigeria, 0.4 percent;
Australia, 0.4 percent. These numbers represent a slight drop of US visitors and slight
increases from various other countries, but the proportions remain substantially similar to those
seen in 2011.

Preferred operating system: Windows (al variants) was used by 67 percent of our visitors,
down from 73 percent the year before. Mac OS X was the choice of 21 percent (up from 20
percent). Linux captured 1 percent, down slightly from 2011. Most of the remaining traffic came
from mobile operating systems like iOS and Android, the fastest-growing segment for our
customers. (See more about mobile visits, below.)

Preferred web browser: Customers of our marketing site (i.e., our potential students and their
parents) use a variety of browsers. Internet Explorer (al variants) continued its downward
trend, being used for 40 percent of visits in 2011 but only 32 percent in 2012; Safari rose from
20 percent to a 23 percent share; Chrome showed the biggest increase, from 15 to 23 percent;
Firefox dropped from 22 percent to 18 percent. (Mobile browsers continued their long upward
trend and accounted for most of the rest.)

Use of mobile devices: Mobile-device access to again increased dramatical y,
as it has for the past several years. In 2012, 10.9 percent of page visits were made with a
mobile device, typically a phone or a tablet computer (up from 5.7 percent in 2011 and 2.3
percent in 2010). Apple devices continue to represent about 70 percent of all mobile visits to Android devices represent 23 percent, Blackberry about 2 percent, with a
number of mobile platforms accounting for the remaining 5 percentage points. In 2012, we saw
a shift by our visitors toward tablets and away from phones. Tablets accounted for about half of
all mobile visits.
In conclusion, the marketing site – used primarily by potential and future members of
the Mines community – may be seen as a preview of our future web needs. The trends seen here –
particularly concerning the rise of mobile web access – show us where the web at Mines needs to go.
Clearly, our mobile-device strategy should be revisited frequently and with some urgency. On the other
hand, as more people use tablets to access Mines pages (rather than using smal -screen phones), this
may tend to lessen our need for mobile-device-specific pages. Since we specifical y refrained from
using Flash in our main websites, tablets (including the iPad, which does not support Flash) can
access our ful web sites with ease. But if access via mobile phone also continues to increase
dramatically, perhaps new mobile-specific apps or web pages wil be in our future.
Note that this document is a general summary of web trends. For more details and specific numbers
over several years, take a look at our new document (available at
called Mines Web Statistics at Your Fingertips. It details specific numbers from 2009, when we began
to col ect detailed analytics of our sites, to the present.