Disclosing the network structure of private companies on the web. The case of Spanish IBEX 35 share index
| Date | 08 June 2015 |
| Published date | 08 June 2015 |
| Pages | 360-382 |
| DOI | https://doi.org/10.1108/OIR-11-2014-0282 |
| Author | Enrique Orduña-Malea,Emilio Delgado López-Cózar,Jorge Serrano-Cobos,Nuria Lloret Romero |
| Subject Matter | Library & information science,Information behaviour & retrieval |
Disclosing the network
structure of private
companies on the web
The case of Spanish IBEX 35 share index
Enrique Orduña-Malea
EC3 Research Group, Instituto de Diseño y Fabricación,
Polytechnic University of Valencia, Valencia, Spain
Emilio Delgado López-Cózar
EC3 Research Group, University of Granada, Granada, Spain, and
Jorge Serrano-Cobos and Nuria Lloret Romero
Instituto de Diseño y Fabricación, Polytechnic University of Valencia,
Valencia, Spain
Abstract
Purpose –It is common for an international company to have different brands, products or services,
information for investors, a corporate blog, affiliates, branches in different countries, etc. If all these
contents appear as independent additional web domains (AWDs), the company should be represented
on the web by all these web domains, since many of these AWDs may acquire remarkable performance
that could mask or distort the real web performance of the company, affecting therefore on the
understanding of web metrics. The purpose of this paper is to determine the amount, type, web impact
and topology of the AWDs in commercial companies in order to get a better understanding on their
complete web impact and structure.
Design/methodology/approach –The set of companies belonging to the Spanish IBEX-35 stock
index has been analysed as testing bench. The authors proceeded to identify and categorise all
AWDs belonging to these companies, and to apply both web impact (web presence and visibility)
and network metrics.
Findings –The results show that AWDs get a high web presence but relatively low web visibility, due
to certain opacity or less dissemination of some AWDs favoring its isolation. This is verified by the low
network density values obtained, that occur because AWDs are strongly connected with the corporate
domain (although asymmetrically), but very weakly linked each other.
Research limitations/implications –The categories used to classify the various AWD, although
they are clearly distinguishable conceptually, have certain limitations in practice, since they depend on
the form adopted by companies to publish certain content or to provide certain services or products.
Otherwise, the use of web indicators presents certain problems of accuracy that could be softened
if applied with caution and in a relational basis.
Originality/value –Although the processes of AWDs creation and categorisation are complex
(web policy seems not to be driven by a defined or conscious plan), their influence on the web performance
of IBEX 35 companies is meaningful. This research measures the AWDs influence on companies under
webometric terms for the first time.
Keywords Spain, IBEX 35, Online business information, Social network metrics, Web visibility,
Web metrics
Paper type Research paper
Online Information Review
Vol. 39 No. 3, 2015
pp. 360-382
©Emerald Group Publishing Limited
1468-4527
DOI 10.1108/OIR-11-2014-0282
Received 26 November 2014
First revision approved
31 March 2015
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/1468-4527.htm
This research has been funded under the project APOSTD/2013/002 from the Regional Ministry
of Education, Culture and Sport (Generalitat Valenciana) in Spain.
360
OIR
39,3
Introduction
The application of web indicators to the performance of companies through cybermetric
techniques is relatively low when compared with other applied areas of this discipline,
especially those studies focused on universities (Aguillo et al., 2006), academic journals
(Vaughan and Thelwall, 2003) and, more recently, alternative metrics (Thelwall, 2014).
However, during the last decade there has beenlimited but increasing research focusing
on commercial companies (Romero-Frías, 2011).
An early focus of this research was the search for significant correlations between
web indicators, especially web visibility (the number of links that target a particular web
domain) of companies’corporate web sites, with financial variables (Vaughan, 2004).
Other studies looked at the benefits of certain web indicators, especially those that can
be used to understand the relationships between companies in homogeneous and
heterogeneous sectors (Vaughan and Wu, 2004). Particularly noteworthy are the studies
using co-links –web sitesA and B are co-linked “x”times when “x”websites link to both
A and B (Vaughan and You, 2006) –and co-words: words A and B are co-worded “x”
times when “x”web sites mention both words A and B (Vaughan and You, 2010).
In parallel there have been numerous studies based on the web platforms likely to
provide the necessary data to construct indicators, for example, general search engines,
such as Alltheweb, Altavista, MSN and Google (Vaughan and Wu, 2004) and Yahoo!
(Vaughan and You, 2006); specific search engines, such as Google News and Google
Blogs (Vaughan and Romero-Frías, 2012); and web traffic sources, such as Alexa,
Compete and Google Trends (Vaughan and Yang, 2013).
These works expanded the application of cybermetrics outside academic
environments (where the objects of study are academic content or academic-related
agents, such as universities, journals, authors, etc.), exploiting web data for competitive
intelligence purposes (Reid, 2003) under the assumption that link and web content
analysis can be useful both for gathering soft information (non-statistical intelligence
gathered from non-traditional sources) in general, and for identifying competitors’
relationships (official or unofficial) in particular.
Nonetheless, the application of web metrics to commercial companies has a number
of important shortcomings. For example, the analysis of direct links between firms is
not recommended since companies rarely tend to generate links to their competitors
(Shaw, 2001). Moreover, they do not publish trade secrets, and their web sites’impact
(web visibility, web traffic, etc.) may depend on many external factors (stock sales
and purchases of businesses, event sponsorship, etc.) which are difficult to identify and
correct, and make their web impact widely distributed and complex to measure accurately
(Thelwall, 2000, 2001).
One of the biggest limitations is determined by the complexity of company web
sites, due not only to their content structure (determined by parameters of information
architecture and interaction design), but on the existence of multiple web doma ins in
which content is located, that indirectly reflect the real structure of these firms.
This complexityis determined not only by the existenceof sub-domains or sub-folders
(which obviously exist), but by the existence of multiple web domains, technically
independent of the official corporate web site, intended to provide targeted information
directly or indirectly related to the company.
In this sense it is common for an international company to have different brands,
specific products or services, information web sites for investors, a corporate blog,
affiliates or branches in different countries(with their own products, services, etc., acting
as a fractal), etc. If all theseweb sites appear as separate web domains or additional web
361
Network
structure of
private
companies
To continue reading
Request your trial