EVALUATION OF SEARCH ENGINE OPTIMIZATION
Experiments for a Business Site and a New Evaluation Measure
Julia Maria Schulz, Ralph Kölle, Christa Womser-Hacker and Thomas Mandl
Information Science, University of Hildesheim, Germany
Keywords: Search Engine Optimization, Internet Marketing.
Abstract: This paper reports experiments on search engine optimization (SEO) for a business site. Several search
terms have been optimized for three web search engines. From the business site, 300 pages have been
selected for optimization. In three phases several on- and off-page modifications have been carried out and
the results have been monitored. The results show that search engines do react to modifications and that the
target pages are ranked higher on average. The variance of the improvements is extremely large which
means that there is no guarantee that SEO activities are beneficial for one single page. We suggest a new
evaluation measure for the success of SEO which takes typical Web user behavior into account.
1 INTRODUCTION
Search Engine Optimization (SEO) is practical
information retrieval from the perspective of the
information provider. Web site authors and
providers are interested in attracting many visitors
mostly for economic reasons. They want that their
pages are ranked at top positions in search engine
results for relevant keywords. Relevance is defined
from the perspective of the provider in this case.
Usually, the user is considered as central for the
definition of relevance in information science.
SEO has become a prosperous business branch
for information professionals. Many companies have
been established and many guide books have been
published (e.g. Grappone and Couzin, 2006). There
is much anecdotic evidence on the effects of SEO on
single pages. However, surprisingly little reliable
empirical research is available for the topic. We
conducted an empirical study which showed the
effects of several SEO activities. The analysis
showed that new measures for analyzing the success
and value of SEO activities are necessary.
Search engines are ambivalent toward SEO. On
the one hand, they do not publish their algorithms
which would make SEO a more transparent task.
The search engines argue that such a step would lead
to much spam in the search results. On the other
hand, search engines companies give hints on
potential optimisation potential. In such suggestion
lists, the search engines clearly distinguish between
desired and undesired practices.
This makes SEO an information ethical issue.
Search engines draw the borderline between
tolerated and forbidden SEO activities. The
regulations are not publicly discussed but simply
proposed by the companies. The discovery of the
violation of these regulations may lead to
consequences. This can take the form of removal
from the index or lead to lower ranking positions.
Punishment in such a form is obviously problematic.
They are not transparent, information providers are
not properly notified and no appeal is possible.
Much rather, the violator needs to rely on the mercy
of the search engines.
Meanwhile, the internet is extremely important
for finding information and for being found from the
perspective of information providers. Search became
a basic service. There are even calls for a public
provision of search services (Maurer, 2007). From
the perspective of SEO, it would be preferable to
achieve a more democratic decision process for
regulations than to simply accept the regulation of
the search engine businesses.
2 EXPERIMENTAL SETUP
In a experiment, the effect of typical SEO services
were analyzed systematically. As a target site, a e-
commerce shop of a present vendor was selected
247
Maria Schulz J., Kölle R., Womser-Hacker C. and Mandl T. (2008).
EVALUATION OF SEARCH ENGINE OPTIMIZATION - Experiments for a Business Site and a New Evaluation Measure.
In Proceedings of the Tenth International Conference on Enterprise Information Systems - SAIC, pages 247-250
DOI: 10.5220/0001713602470250
Copyright
c
SciTePress