Information Technology Reference
In-Depth Information
now that you are familiar with the long tail, you should be able to use it to your
advantage in any number of ways. although there are plenty of metrics to play with,
i like to showcase two interesting ideas for optimizing your keywords using long tail
data from within yahoo! web analytics.
let's look at a traditional search phrases report, as shown in Figure 13.2.
327
Figure 13.2
Search Phrases report
this is a great report by itself, and you can do some exciting stuff with this
dimension. however, take a closer look, and you'll notice that the data you see on
screen is very limited and only represents a small section of not only the organic traffic
influx, but the total number of unique search phrases.
listed on the screen are 10 unique search phrases, accounting for 4,492 visits
(just below 20 percent of all traffic). if we believe in the traditional long tail and that
this dimension follows it, we should expect this to be the head of our distribution.
thus there should be a long tail of unique search phrases generating very little traffic,
but altogether generating over 80 percent of the total traffic influx.
an easy way to test this is by showing 100 rows per screen and scrolling into the
subsequent pages. i suggest that in such cases you use yahoo! web analytics' export
functionality and export all the search phrases. when you export, just go with the
maximum number when asked; why settle for anything less? For our example, i chose
to export 200,000 rows in a microsoft excel format. the excel file for the data in
Figure 13.2 is shown in Figure 13.3.
scroll to the bottom to get a sense of the true volume of unique search phrases.
in this scenario, we have 8,834 unique search phrases.
our first optimization opportunity involves correlating this complete list of
8,834 phrases with our entire catalog of paid search listing keywords. this will create
a list of phrases that we are not bidding on, and we need to determine why that is.
