X hits on this document

PDF document

Introducing B-COP: A Methodology to Benchmark Competitive Online Performance - page 9 / 21





9 / 21

Then, we invoke the Alexa.com page for each of the direct competitors and once again

study the “related links”. Our interest here is in investigating whom the direct competitors

compete with. A competitor of a competitor is counted as an indirect competitor. Frequently,

we find that indirect competition is strong and ignoring it leads to misleading results.

Then, we compile all direct and indirect competitors into a list. The top ten members of

this list in terms of traffic rank over the last 3 months are chosen as the competitors of the

company. This approach allows us to identify competitors based on actual data.


Computing absolute and relative ranks, reach and page views. .

We calculate the traffic rank for each of the ten competitors. Relative ranks are

calculated by dividing a competitor’s absolute rank by the subject company’s absolute rank. For

instance, if we are benchmarking Google’s competition, the relative rank of Yahoo is given by

absolute rank of Yahoo/absolute rank of Google. A relative rank higher than the numeric value 1

is an indication of a strong competitor. For example, in the Amazon.com case study provided

below you will find eBay’s relative rank is 1.9 indicating that it is stronger. A relative rank less

than 1 is an indication of a weak competitor. Similar calculations are performed for reach and

page views. The formulas are shown below-

  • Relative Rank = Subject Absolute Rank / Competitor Absolute Rank.

  • Relative Strength = Competitor Reach / Subject Reach

  • Relative Page Views = Competitor Page Views / Subject Page Views

Usually, competitiveness among brands is defined using an us/(us+them) index. Our

point is that doing that on the Internet is not always effective since competitive boundaries are

amorphous. There is no clear way of identifying a universal set of competitors even- for

instance, the managers at New York Times may not have considered BBC.co.uk as an important


Document info
Document views91
Page views91
Page last viewedSun Jan 22 18:32:54 UTC 2017