The ATA Chronicle - September/October 2017 - 23

quote (i.e., one of the main goals of having
a website in the first place!). In contrast,
referrals from bad bots, such as rankchecker.online and website-analyzer.info,
lead to sessions lasting just a few seconds,
with most visits to only one page, and
never result in a quote request.

HOW TO FILTER OUT
CONTAMINATED INFORMATION
FROM GOOGLE ANALYTICS REPORTS
Unfortunately, there is no universal fix or
global solution for eliminating bad bots
and spam referral from Google Analytics
data. Nevertheless, we provide three ways
you can "decontaminate" Google Analytics
data. They range in cost and complexity
and the choice will ultimately depend
on your objectives and the resources you
want to devote to the effort.
Apply Filters: The more advanced
Google Analytics users and aspiring cyber
security sleuths can visit the ADMIN
section of Google Analytics. (See Figure 5
on page 24.) Under the "View" column,
you will find a section called "Filters."
Here, filters can be added to remove
additional unwanted bot traffic and spam
referrals. At this point, the steps involved
can be stressful and time-consuming
and could have unintended results such
as accidently removing a legitimate
website crawler that is indexing your site
and which could increase its visibility.
Consequently, you should consider taking
advantage of Google's automatic exclusion
feature discussed in the next paragraph,
or contracting with experts on the matter.
Keep in mind that the creation of these
filters will not clean up historical data.
Check the Box: In July 2014, Google
Analytics announced a new feature to
automatically exclude bots and spiders
that appear on what is known as the
International IAB/ABC Spiders and Bots
List8 that is updated continuously. Since
every Google Analytics account requires
a unique sign in, there is no direct link
to the box you need to check to exclude
these nasty critters. You can find this
feature by going to the "ADMIN" section
of analytics and clicking on "View
Settings" under the "View" column.
(See Figure 5 on page 24.) Under "Basic
Settings" you can check the Bot Filtering
www.atanet.org

2009

2011

2016

Language

Sessions

Language

Sessions

1. English (U.S.)

1,819

1. English (U.S.)

3,283

Language

Sessions

1. Russian

1,643

2. English

105

2. English (U.S.)

1,417

3. Chinese

3. English (U.K.)

3. (not set)

817

4. German

4. Spanish

4. Russian (Russia)

474

5. Spanish

5. Chinese

5. Brazilian Portuguese

173

6. Italian

6. Spanish
(Spain)

6. Secret.google.com You are
invited! Enter only with
this ticket URL. Copy it.
Vote for Trump!

114

7. Brazilian
Portuguese

7. Chinese

8. French

8. c (the "c" indicates a bot)

9. Spanish
(Spain)

9. German

9. Spanish

10. Russian

10. English (U.K.)

Average Session Duration:
112 seconds

Average Session Duration:
56 seconds

Average Session Duration:
40 seconds

Figure 3: Audience Overview: Language-2009, 2011, and 2016 (unfiltered data)

Source

# of
sessions

Average # of pages Average session duration
per session
(minutes:seconds)

# of quotes
requested

atanet.org

553

2.99

1:38

cprintalliance.com

182

2.13

2:40

rank-checker.online

800

1.01

0:12

site-auditor.online

791

1.18

0:17

Figure 4: Contrasting profiles of visitors from legitimate websites and profiles of the bad bots that
generate spam referrals (2009 to 2016)

box labeled "Exclude traffic from
known bots and spiders." Unfortunately,
"Checking the box" will not clean up
historical data, but should filter data
going forward. It may take a week or two
to begin working. You may want your
website administrator to set up a test view
first to prevent any accidental damage to
unfiltered historical data you may need
later. Note: You will need admin access to
your Google Analytics account to apply
filters or to check the Bot Filtering box.
Manually Recreate the Source/Medium
Report: Since you cannot apply filters to
the contaminated historical data, you may
want to recreate the Source/Medium report
for one or more past periods. This can
be done by opening a new document in
Word and creating a simple table. Copy the
valid sources of site visits and data from
the contaminated Source/Medium report,

while purging data left behind by the bad
bots and spam referrals. You can then
recalculate session numbers and re-rank
the sources. Bad bots and spam referrals
can be found by looking for characteristics
such as a 100% bounce rate, session
durations of one second or less, and the
telltale words mentioned earlier in this
article. The Source/Medium page may be
the only one you can recreate, but it's one
of the most important reports that provides
actionable information for marketing your
translation business.
In Figure 6 on page 24, we compare
our company's manually filtered 2016
Source/Medium data with 2016 unfiltered
data. Note how atanet.org regains its
historic position in the top five sources
of referrals, as compared with the 13th
position in the unfiltered data.
American Translators Association

http://www.website-analyzer.info http://secret.google.com http://www.atanet.org http://www.cprintalliance.com http://www.atanet.org http://www.atanet.org

Table of Contents for the Digital Edition of The ATA Chronicle - September/October 2017