A Research Roadmap
Georg Peters, Florian Thoma
University of Applied Sciences - München
Department of Computer Science and Mathematics
Lothstrasse 34, 80335 Munich, Germany
Richard Weber
Universidad de Chile
Departamento de Ingeniería Industrial
Republica 701, Santiago, Chile
Keywords: Enterprise Resource Planning Systems, Network Externalities.
Abstract: Presently the ERP software market is characterized by a fierce competition of some of the largest software
producers of the world. While SAP tries to defend its position as the market leader, Oracle, founded as
database vendor, has heavily invested into its ERP business by taking over large related software producers
like Peoplesoft or Siebel. Microsoft also bought ERP vendors as a basis for its MBS (Microsoft Business
Solution) division. IBM and Sage are further players in the multi-billion market of ERP software. Currently
the market faces two big challenges. (1) The classic ERP market addressing large corporations has become
mature and is characterized by growth through displacement today. However the ERP market for SME still
promises big growth rates. (2) The classic monolithic ERP systems are presently replaced by new
middleware based technologies that provide open platforms to easily integrate business applications of third
party software producers. Network externalities will significantly influence which ERP vendor(s) will
dominate the future market. Therefore we will qualitatively analyse the current strengths and weaknesses of
the presently most promising ERP producer with respect to network externalities.
Currently the market for Enterprise Resource
Planning systems (ERP) is characterized by a fierce
competition of some of the biggest and strongest
software companies in the world. The competition
goes along with a market consolidation that makes
many even former large software companies to
While SAP is considered to be the market leader,
Oracle is aggressively investing in the market with
the objective of bypassing SAP.
In the last few years Oracle took over several
major players in the ERP sector. In late 2004, Oracle
acquired Peoplesoft for USD 10.3 billion that
already bought J.D. Edwards in 2004 (La Monica
2004). Siebel, the pioneer and former market leader
in CRM software, was bought for USD 5.85 billion
(including Siebel's cash of USD 2.24 billion)
Ihlenfeld 2005). Recently, in 2007, Oracle
announced to take over Hyperion, a specialist for
BPM - Business Performance Management software
for about USD 3.3 billions (Kahn et al. 2007).
In contrast to Oracle, until recently, SAP's
strategy has been to grow organically instead of
aggressively buying big software company to
complete its own portfolio. However, in 2007, it
announced to buy the French software company
Business Objects for EUR 4.8 billions (more than
USD 6.5 billions) to strengthen its position in the
field of Business Intelligence (Prodhan 2007b).
In the recent years Microsoft has bought several
ERP vendors, namely Axapta, Great Plains,
Navision and Solomon, which have formed the core
Peters G., Thoma F. and Weber R. (2008).
In Proceedings of the Tenth International Conference on Enterprise Information Systems - DISI, pages 97-104
DOI: 10.5220/0001674900970104
of its MBS division (Microsoft Business Solutions).
The MBS division is considered as one the core
areas within Microsoft's future grow. Like Sage a
leading Norwich UK based ERP producer
Microsoft's products focus on smaller business
(SME) rather than the core products of SAP and
Oracle which address large organisations.
And finally, very recently IBM announced to
take over the Canadian software company Cognos
for USD 5 billions (Wu 2007).
The competition and market consolidation is
mainly driven by the following factors.
(1) The ERP market for large organisations has
creased to grow rapidly. It has become a mature
market that is characterized by growth through
displacement. Only the market for ERP systems for
SME is considered to generate high growth rates in
the next years. Therefore new and established ERP
vendors introduce software products especially
designed for this clientele. For example -besides its
already established product lines for SME like
Business One - SAP recently introduced its Business
ByDesign applications (Prodhan 2007a) and has
started to even advertise its software on television.
(2) There is a technological shift from monolithic
ERP systems like SAP R/3 to open middleware
based systems that allow integrating and running
small software pieces from third parties. Besides
ERP vendors like SAP that introduced Netweaver
and Microsoft that promotes its .NET technology
software company, like IBM, that has no close
affiliation to ERP software are offering middleware
technology as platform to run business applications.
IBM promotes its middleware platform Websphere
as neutral and truly open since it has no ERP
package to sell with it like classic ERM vendors
The network externalities are having a significant
influence how the ERP market of tomorrow will
look like, in particular which software companies
will dominate it.
The objective of this article is to qualitatively
analysis the strengths and weaknesses of leading
ERP software companies with respect to network
externalities and provide a roadmap for future
research. The complexity of this topic requires it to
limit our presentation to an overview on network
externalities and a short description of each factor
with respect to each company. So the paper also
functions as a guidebook for our ongoing research
where each factor is address in depths.
The remaining paper is organized as follows. The
next Section gives an introduction to the theory of
network externalities. In Section 3 we present the
results of a qualitative analyse of the positions of
some leading companies related to network
externalities. The paper concludes with a summary
in Section 4.
2.1 Foundation of Network Theory
Often superior quality and state of the art technology
are regarded as the most crucial factors for a product
to become a success. However there are many
examples where inferior products and technologies
eventually dominate the market after they squeezed
out better products with superior technologies. So, in
literature, the question was addressed if such
occurrences can be considered as market failure (e.g.
Liebowitz, Margoli 1995b).
A classic example is the "battle" between several
video cassette recorder technologies (VCR) in the
eighties of the last century. Although Sony's
Betamax technology was considered to be the best
but it failed to become market standard. Instead the
VHS format of JVC squeezed Betamax out of the
consumer market and dominated the VCR area until
recordable DVD eventually replace video cassettes
(for simplicity we restrain from the consideration of
further VCR technologies like the European format
This, at the first sight, "odd" market behaviour
can be explained by network externalities. Basically
the idea of network externalities is as follows. The
utility of a technology is not only determined by its
quality but also by its diffusion, or in other words by
the size of its network:
u = q + n
u = utility, q = quality, n = size of the network.
Consider the video cassettes. The more
consumers own a certain format the more cassettes
are offered in video stores, the easier it is to
exchange cassettes with friends etc. In the end the
advantages generated out the network size are more
important than a marginally crisper video with a
slightly better sound of a technology that rarely
anybody else has.
JVC simply managed to achieve more users for
its VHS technology by quicker and better identifying
the consumers preferences and demands (Liebowitz,
Margoli 1995b) so that its inferior technology was
overcompensated by its larger network. Therefore,
ICEIS 2008 - International Conference on Enterprise Information Systems
in the end, the utility for the consumer of the VHS
technology was higher than the utility of Sony's
Betamax system.
Another famous example for an "odd" market
outcome in the presence of network externalities is
the QWERTY keyboard (David 1985) that is
regarded as ergonomically inefficient but seems to
be set in stone. A current example, quite similar to
the VCR case and again under the participation of
Sony, is the battle of the two new DVD
technologies, BlueRay and HD DVD. Network
externalities have gained special attention to explain
market outcomes in the software sector (e.g. the
classic paper of Brynjolfsson, Kemerer 1996 or
Succi et al. 1998).
Formally “Networks are composed of
complementary nodes and links. The crucial
defining feature of networks is the compatibility
between the various nodes and links” (Economides
2003). The key characteristic of network
externalities is the fact that the utility of each
member in a network increases with the size of a
network. (Katz , Shapio 1985).
Katz and Shapiro (1985) defined two types of
network externalities, namely direct and indirect
network externalities. (Liebowitz and Margolis
1998) give the following example: "Direct network
effects have been defined as those generated through
a direct physical effect of the number of purchasers
on the value of a product (e.g. fax machines).
Indirect network effects are ‘market mediated
effects’ such as cases where complementary goods
(e.g. toner cartridges) are more readily available or
lower in price as the number of users of a good
(laser printers) increases”.
Network externalities can result out of different
effects. In our paper we deal with:
Learning effects
Economies of scale
Information and communication
Technological standards
which will be introduced in the following
2.2 Sources of Network Externalities
2.2.1 Learning Effects
Learning plays an important role in industry. It is
one essential precondition for any progress and long
term growth. One of the core objectives in a
company is to improve efficiency of the production
process based on experiences gained in the past. Or
as (Henderson 1974) says “costs […] do continually
decline as a function of experience." The famous
related concept is the learning curve which goes
back to (Wright 1936). Later the more involved
Crawford model also gained big attention (Teplitz
In the context of network externalities we
distinguish between two types of learning, (1)
learning by doing and (2) learning by using.
Learning-by-Doing. Leaning-by-Doing is related to
improvements of productivity. The more a product is
produced the more the production process can the
optimized with respect to time and costs. See (e.g.
Leiby, Rubin and Lu 1997) for an overview of
studies related to Learning-by-Doing.
Learning-by-Using. In contrast to Learning-by-
Doing Learning-by-Using (Rosenberg 1982) is not
related to the production process but directly to the
usage of the product itself.
The more a product is used the more ideas for
product improvements will be generated.
The more a product is used the better it is
understood. Therefore it can be used more
Often Learning-by-Using results in a lock-in of a
technology (Arthur 1989). For example, the
QWERTY keyboard is considered to be a locked-in
technology (David 1985). There are better keyboard
layouts. However, since the QWERTY layout is the
standard it would be to expensive (training costs
etc.) to replace it with a superior keyboard layout.
Therefore the suboptimal technology QWERTY
keyboard is locked-in.
2.2.2 Economies of Scale
Economies of scale can be divided into supply side
and demand side economies of scale. Supply side
economies of scale are related to the production side
and to decreasing production costs when production
increases. In economic literature these effects are
often not considered as network effects.
Network effects are, in a narrow sense, no
economies of scale, or they are defined as demand
side economies of scale. For the context of our
survey we will restrain from these definitions and
will smatter of economies of scale meaning both
2.2.3 Information and Communication
When a consumer is in the decision process what to
buy, she/he collects information of suitable products.
Normally it is easier to obtain information on a
popular product than on a product that is not widely
spread. Therefore, due to its larger network, a (good)
popular product has already a head start in
comparison to an exotic product characterized by a
smaller network. It is even rational to stop searching
for information on exotic product when the search
costs exceed the savings obtained by buying the
exotic instead of the popular product.
In related literature the concept is known as
bandwagon effect (Leibenstein 1950). Customers
mainly buy a product because other customers
bought the same product before. Or in trust theory,
network based trust (Henning-Thurau 2004) can be
observed when consumers regard products as
trustable because other consumers trust the same
product also.
2.2.4 Sub-Technologies
A sub-technology network emerges when additional
products or services accompany the core product.
These products and services increase the utility of
the core product.
For example, regard the recent announcement of
Google, Qualcomm, Samsung, T-Mobile and
another 30 companies to promote the open operating
system Android for mobile phones. It is considered
that one advantage of Android is that many small
applications will be developed for this platform
(Auchard et al 2007). These applications are sub-
technologies which increase the utility of the core
product. The larger the network is the more
applications will be produced and the more useful is
the core product.
2.2.5 Technological Standards
Last but not least, the most important source for
network externalities is probably (technological)
standardization (Katz, Shapiro 1985), (Farrell,
Saloner 1985). In contrast to sub-technologies we
address here the direct compatibility within a
network. For example users of video tapes can only
exchange movies if they have chosen the same
technological platform/standard (e.g. VHS).
There are many examples in history where
companies "fought" to make their technology the de
facto industry standard. Besides VRC and the
already addressed competition of the new DVD
standard standardization is of great importance in the
software sector also.
Well known examples are the Windows
operating system of Microsoft or the current
discussions on an industry-wide standard for office
documents like word processing or spreadsheets and
others: OASIS Open Document Format for Office
Applications vs. Microsoft's Office Open XML.
In this Section we qualitatively analysis the strength
and weaknesses of five leading ERP software
vendors - IBM, Microsoft, Sage, SAP, Oracle - with
respect to their network externalities.
Please note, that we also consider IBM as the
producer of the middleware WebSphere without any
classic ERP solution. However, middleware
platforms are regarded as central technologies for
ERP systems. Their significance can be compared
with the importance of an operating system for
personal computers.
Therefore, currently, the competition in the ERP
sector for large organisations is determined by the
emphasis of the vendors to dominate the market of
middleware solutions.
3.1 Learning Effects
3.1.1 Learning-by-Doing
We consider learning-by-using related to the
expertise in developing relevant software systems.
Strengths. IBM is one of the biggest players in
information technology with a high expertise
especially in large scale systems.
Weaknesses. Within the scope of its market
IBM has no significant weaknesses.
Strengths. Microsoft has a high expertise in the
development of office software which it utilises
for the design of ERM software for small
organisations. Since it bought several ERP
vendors it has expertise in this sector too. Its
technology .net has gained increasing
Weaknesses. In the beginning Microsoft has
struggled to successfully develop software for
ICEIS 2008 - International Conference on Enterprise Information Systems
larger IT infrastructure (e.g. Windows Server).
However, in the meantime they have gained
knowledge in this field.
Strengths. Sage has a high expertise in the
development of ERP applications for small
Weaknesses. There is no information available
if Sage ever considers entering the ERP market
for large organisations.
Strengths. SAP is the market leader in ERP
systems for large organisations. Therefore it has
the highest expertise in its development of all
Weaknesses. Although SAP has a high expertise
in the development of ERP systems for large
corporations it has to proof that it can transfer
this knowledge to the development of ERP
applications for small organisations and
middleware technology.
Strengths. Oracle, as the second largest software
company worldwide, has a high expertise in the
development of complex software systems,
particularly databases. It has also gained high
expertise in the ERP sector by buying several
large ERP vendors like Siebel or Peoplesoft.
Weaknesses. The integration of the products
could divert from an enhancement of the
3.1.2 Learning-by-Using
Here we analyse if the user feed helps to improve
the product.
Strengths. IBM is a well established producer of
middleware systems. Therefore it should have
high expertises gained by user feedbacks.
Weaknesses. We could not identify any
significant weakness related to this topic.
Strengths. Due to the distribution of its products
Microsoft has constant and high feedbacks of its
Weaknesses. The feedback in the ERP section is
smaller since the recent start of this products
and the smaller number of customers. However
its customer basis is increasing rapidly.
Strengths. Sage has a large customer base and
long experience in ERP sector. Therefore it has
got a number of high feedbacks.
Weaknesses. SME might not provide the same
level of feedback as large organisations do.
Strengths. As the market leader in the sector for
large organisations SAP gets constant and
frequent user feedbacks.
Weaknesses. We could not identify any
significant weakness related to this topic.
Strengths. As the second largest ERP vendor
Oracle gets constant and frequent user
feedbacks. It also gets feedback of its ERP
Weaknesses. The integration of recent
acquisitions may distract from the feedback.
3.2 Economies of Scale
As discussed above economies of scale are supplier
side network effects. Software production is
characteristic by large fix costs and small variable
costs. The effect for software development was
addressed by (Banker et al. 1989).
Strengths. As a large software producer IBM
should generate reasonably economies of scale.
Weaknesses. We could not identify any
significant weakness related to this topic.
Strengths. As the largest software company of
the world Microsoft should be able to generate
significant economises of scale.
Weaknesses. We could not identify any
significant weakness related to this topic.
Strengths. As leading company of EPR
solutions for SME Sage should generate
reasonably economies of scale.
Weaknesses. Sage has a decentralized structure.
Strengths. SAP is the market leader in the ERP
market. Therefore it should generate economies
of scale in its core market.
Weaknesses. We could not identify any
significant weakness related to this topic.
Strengths. Oracle is the second largest software
company of the world. Therefore it should
generate significant economies of scale.
Weaknesses. We could not identify any
significant weakness related to this topic.
3.3 Information and Communication
In this overview we do not distinguish between
information and communication.
Strengths. As the largest IT company of the
world IBM has an excellent image and is
considered to produce high quality solutions for
large organisations.
Weaknesses. IBM may have the conviction to its
customers that a neutral platform has significant
advantages in comparison to more integrated
Strengths. As market leader in the fields of
desktop operating systems and office products
Microsoft has a high reputation and can argue
that it can smoothly integrate ERP products
Weaknesses. Microsoft may have to convince
medium and, in the long run, large size
organisations that their products suit them too.
Strengths. Sage ERP products for SME have a
good image.
Weaknesses. Sage must try to keep the image
that it will survive the competition of Microsoft
and SAP because have entered the ERP market
of SME.
Strengths. As a market leader SAP is the choice
that has the least risk. If one chooses the market
leader the decision cannot be wrong.
Weaknesses. Small companies consider SAP as
complex and expensive. SAP must convince
SME that their ERP suites for small
organisations are easier to install and use than
its classic applications.
Strengths. Oracle is known for its expertise in
database systems.
Weaknesses. During the integration phase of
Peoplesoft, Siebel and others Oracle faces the
challenge that it has to manage the integration
and takes care of its old clients.
3.4 Sub-Technologies
We consider to sub-technologies, add-on software
and consulting expertise to run and maintain the
Strengths. As one of the biggest information
technology companies the world IBM
demonstrates consulting expertise and has
related products that function as sub-
technologies for its middleware platform.
Weaknesses. We could not identify any
significant weakness related to this topic.
Strengths. As the largest software company
offers a wide range of sub-technologies for its
ERP products.
Weaknesses. Often Microsoft is regarded to
have a close-door so that third party
applications might not be developed in the
numbers as in the open source community.
However it very successfully communities etc.
around its .NET technology.
Strengths. We could not identify a special
strength of Sage in this category.
Weaknesses. We consider Sage's position
related to sub-technologies not as strong as the
positions of its competitors.
Strengths. SAP has a wide range of products
that function as sub-technologies. Furthermore
it has a large network of consultants and
software solutions by third parties around its
core products.
Weaknesses. We could not identify any
significant weakness related to this topic.
Strengths. As a large software company Oracle
can offer sub-technologies to their ERP product.
Weaknesses. The network of consultants might
not be as strong as the one of SAP.
3.5 Technology Standards
Setting a (technological) standard is the most
important network effect. The middleware
technologies are advertised as open with a high
degree of interoperability.
ICEIS 2008 - International Conference on Enterprise Information Systems
Strengths. Since IBM offers the platform only
its police must be to disclose as much
information on its standard as possible to attract
third parties to run applications on it.
Weaknesses. The above said is a strength and a
weakness as well.
Strengths. Microsoft has a successful history of
setting proprietary standards. Furthermore the
end users already know Microsoft products and
how to use them. Its .NET platform is very
widely used.
Weaknesses. Because of Microsoft's history in
setting proprietary standards customers might
look for other products.
Strengths. We could not identify a special
strength related to this topic.
Weaknesses. We could not identify a special
weakness related to this topic.
Strengths. SAP is the market leader and sets the
standard for ERP systems.
Weaknesses. It has to demonstrate if it can
leverage this standard to its products for SME
since SAP R/3 and successors, the product for
large organisations is considered as very
Strengths. Oracle is the second largest software
company and also the second largest ERP
Weaknesses. SAP is reasonably bigger in the
ERP market.
In this paper we applied the theory of network
externalities to the vendors of ERP software and
analysed how these effects may influence the future
shape of the market for ERP software.
In our analysis shows that SAP still has a very
strong position in the core sector of ERP software
for large organisations. However, in the last few
years, Oracle has enlarged its ERP business
dramatically by aggressively buying big software
companies in this section. Furthermore it can draw
on its core business, database systems, and possibly
leverage this "sub-technology" to further strengthen
its position in the ERP sector.
Microsoft can possibly leverage its strong
position in operating systems and office products to
help to grow it ERP business for SME.
In our survey IBM has a special position since it
offers WebSphere as platform for ERP solutions.
However, it seems to have not intensions to enter
this market. This approach avoids conflicts of
interest, as they are discussed in the case of
Microsoft and its operating system Windows and
applications running on this platform. Therefore it
relies on the applications of third party vendors.
In the long run Sage seems to be the weakest
vendor although it presently has a strong position in
the market section for SME. However the other
companies, besides their financial power, can draw
on strong networks of sub technologies besides other
effects that will be real threats for Sage.
As the main purpose of our paper is to serve as
an overview of important network effects for ERP
vendors, in our future research we address each
network effect in separated research projects.
Support by the Scientific Millenium Institute
"Complex Engineering Systems'' is greatly
acknowledged; see
Arthur, B. 1989. Competing Technologies, Increasing
Returns and Lock-in by Historical Events. Economic
Journal 99, 106-131.
Auchard E., Carew. S. 2007. Google enters mobile phone
market, shakes industry. Reuters 05.11.2007
ndia-30345320071105, retrieved 11.11.2007).
Banker, R., Klemperer, C. 1989. Scale Economies in New
Software Development. IEEE Transactions on
Software Engineering 15, 1199-1205.
Brynjolfsson, E., Kemerer, C. 1996. Network Externalities
in Microcomputer Software: An Econometric Analysis
of the Spreadsheet Market. Management Science 42,
David, P. A. 1985. Clio and the economics of QWERTY.
American Economic Review 75, 332-7.
Economides, N. 1996. The Economics of Networks.
International Journal for Industrial Organization 14,
Economides, N. 2003. Dictionary of Terms in Network
Economics. School of Business; New York University
retrieved 06.11.2007).
Farrell J., Saloner, G. 1985. Standardization,
compatibility, and innovation. Rand Journal 16, 70-
Henderson, B. 1974. The Experience Curve-Reviewed v.
Price Stability. Perspectives (Reprint) 149
veV.pdf, retrieved 06.11.2007).
Henning-Thurau, T. 2004. Warum Kunden anderen
Kunden im Internet zuhören. Jahrbuch der Absatz-
und Verbrauchsforschung 50, 52-75.
Ihlenfeld, J. 2005. Oracle kauft Siebel für 5,85 Milliarden
US-Dollar. 12.09.2005
(, retrieved
Kahn, M., Davies, M. 2007. Oracle says to buy rival
Hyperion for $3.3 billion. Reuters 01.03.2007
NAS264020070301, retrieved 12.11.2000).
Katz, M., Shapiro, C. 1985. Network Externalities,
Competition, and Compatibility. American Economic
Review 75, 424-440.
La Monica, P. 2004. Finally, Oracle to buy PeopleSoft. 13.12.2004
peoplesoft/, retrieved 12.11.2007).
Leibenstein, H. 1950. Bandwagon, Snob, and Veblen
Effects in the Theory of Consumers' Demand.
Quarterly Journal of Economics 64, 183-207.
Leiby, P., Rubin, J., Lu, C. 1997. Topics on Modeling
New Technology Introduction: Learning-by-Doing,
Irreversible Investment, Risk Aversion, and Limited
Foresight. TAFV Model Technical Memorandum
(, retrieved
Liebowitz, S., Margolis, S. 1995a. Are Network
Externalities a New Source of Market Failure?” Law
and Economics 17, 1-22.
Liebowitz, S., Margolis, S. 1995b. Path dependence, lock-
in and history," Journal of Law, Economics, and
Organization 11, 205-226.
Liebowitz, S., Margolis, S. 1998. Network Effects and
Externalities. Newman, P. (ed): The New Palgrave
Dictionary of Economics and the Law 2. Stockton
Press, New York, 671-74.
Prodhan, G. 2007a. SAP unveils Web software, new
business model. Reuters 19.09.2007
N1926461820070919, retrieved 11.12.2007).
Prodhan, G. 2007b. SAP to buy Business Objects for 4.8
bln euro. Reuters 08.10.2007
telco-SP/idUSL0768722820071008, retrieved
Rosenberg, N., 1982. Inside the Black Box: Technology
and Economics. Cambridge University Press,
Succi, G., Predonzani, P., Valerio, A., Vernazza, T. 1998.
Network externalities in software systems.
StandardView 6, 185 - 191.
Teplitz, C. 1991. The Learning Curve Deskbook. Quorum
Books, Westport.
Wright, T. 1936. Factors Affecting the Cost of Airplanes.
Journal of Aeronautical Science 4, 122-128.
Wu, T. 2007. IBM to buy software maker Cognos for $5
bln. Reuters 12.10.2007
telco-SP/idUSN1244717520071113, retrieved
SAP, SAP R/3, SAP Netweaver are trademarks or
registered trademarks of SAP AG. WebSphere is a
registered trademark of IBM Corp. All other
trademark names used within the article are
properties of their owners.
ICEIS 2008 - International Conference on Enterprise Information Systems