The Impact of ERP Systems On Firm and Business Process Performance
The Impact of ERP Systems On Firm and Business Process Performance
The Impact of ERP Systems On Firm and Business Process Performance
www.emeraldinsight.com/1741-0398.htm
The impact of
ERP systems
13
Introduction
The global success of enterprise resource planning systems (ERPS) has not only attracted
the interest of researchers from the IT/IS-discipline, but from all major disciplines in
business research including accounting. The emergence of ERPS has moved the topic
computerised business information systems increasingly from the IT/IS domain to the
business domain; from systems design and programming to business configuration,
process mapping and reengineering. Using the systems development life cycle (SDLC) as
conceptual reference, we observe that the efforts and weight of the critical success factors
of ERPS projects (Sumner, 1999; Stefanou, 2001) moved from systems design to
systems implementation, and implementation teams are now usually led and dominated
by classic business roles rather than IT-staff. ERPS impose their logic on organisations
and force employees to think in terms of integrated processes and to change the way they
do accounting, production planning and control, etc.
JEIM
19,1
14
All these changes brought about by ERPS made them a very attractive research
object of many business disciplines, and accounting is by nature the one from which
we expect an answer to the question which ultimately counts in business: Are those
systems worth the money?
Earlier approaches to this question identified non-financial performance measures
as proxies for financial value (Bhatt, 2000; Beretta, 2002; Irani, 2002). Only relatively
recently, the accounting discipline has come up with studies which for the first time
provided evidence of the economic impacts of ERPS (Poston and Grabski, 2001;
Matolcsy et al., 2002; Hunton et al., 2003; Nicolaou et al., 2003; Matolcsy et al., 2005).
The results of those studies point into the same direction, but are not entirely
consistent. They all find some evidence of immediate or delayed increases in firm
performance after ERPS adoptions, but the performance increases identified are
evident only in very few KPIs.
We argue that one of the reasons for these inconsistencies is that just relating
ERPS-adoption to publicly available financial key performance indicators (and
controlling for some other dependent variables) fails to explain the increasing diversity
and complexity of the life cycle of enterprise systems. In the emerging generation of
enterprise application integration (EAI) enabled enterprise systems, traditional
ERP-functionality cannot be isolated from the rest of the whole system and has to be
evaluated in the broader context using a multi-dimensional system of performance
measurement.
The purpose of this article is to provide further insights into the impacts of several
aspects of ERPS adoption and ERPS use on firm performance and business process
performance. It aims at challenging existing claims of ERP vendors with regards to the
benefits of their products, and at providing evidence of the impact of bundling ERPS
with supply chain management systems on performance.
We use ERPS adoption, ERPS history and ERPS extension (with a supply chain
management system) as independent variables and KPIs for supply chain performance
and firm performance as dependent variables.
Our key results contradict the claims of ERPS vendors insofar as we found no
significant performance differences between ERPS adopters and non-adopters, neither
at the supply chain level, nor at the overall firm level. While we could confirm the
existence of a learning curve (i.e. the longer the experience of firms with ERPS, the
higher their overall performance), we could not confirm this effect with regards to
business process (supply chain) performance. The latter result contradicts benefits
widely associated with ERPS and supports the arguments in favour of the adoption of
supply chain management systems (SCMS), because the latter proved to have a
positive impact on business process (supply chain) performance in our study.
The main part of the article is structured as follows: First, we develop our research
model and predictions; then we explain our research method and sample and provide
the results of our statistical analysis. The interpretations of the results and limitations
are outlined in the concluding section.
Research framework and model development
Review of literature on the impact of ERPS on performance
Recent research on the impacts of ERPS on financial performance has found some
evidence supporting the claim of all major ERPS vendors that their products enable
The impact of
ERP systems
15
JEIM
19,1
16
Figure 1.
Framework for evaluating
research on the effects of
IT investments
path 1), bypassing many other important factors. They also argue convincingly that
this mainstream one-dimensional approach fails to explain IT-performance and leaves
a gap between anecdotal evidence of IT-success and results of mainstream empirical
research. The IT-productivity paradox discussion and the often conflicting results of
research following alternative paths in their framework can be quoted as the most
important examples supporting their call for a more comprehensive multi-path
approach.
In the subsequent sections, we first operationalise the three objects in our model in
the context of ERPS performance measurement: IT-measures, business process
performance measures and firm performance measures. Then we make predictions
about relationships (paths) between those objects.
IT (ERPS) measures refined
The definitions of ERPS provided by the literature vary slightly, but they all have one
common theme: enterprise-wide integration of data/information and business
processes. For Markus et al. (2000), for example, ERP-systems are commercial
software packages that enable the integration of transaction-oriented data and
business processes throughout an organization, and according to Shang and Seddon
(2002) ERP software integrates management information and processes, such as
financial, manufacturing, distribution and human resources, for the purpose of
enterprise-wide management of resources. They also emphasise that ERPS are an
example, but not the only form of enterprise systems (ES). They discuss common
features of enterprise software which are often associated with ERPS[1], such as set of
packaged application software modules, real-time information processing, based on
deep knowledge of business practices gained by the vendors from their customers,
semi-finished product which has to be configured, etc.
The first comprehensive empirical analysis of the performance of ERPS uses ERPS
adoption as the only independent variable, and then compares the performance of
organisations pre and post-adoption (Poston and Grabski, 2001). Later research considers
and investigates the time-factor in more detail, or more precisely: the time-difference
between the adoption of an ERPS and when performance is measured (e.g. Hunton et al.,
2003; Matolcsy et al., 2002; Nicolaou et al., 2003; Matolcsy et al., 2005). The results of this
more recent research suggest the existence of a time-lag of approximately two years
between the adoption of an ERPS and the realisation of benefits. We therefore include the
time factor in our analysis of performance impacts of ERPS.
The second aspect of ERPS to be considered is the co-existence of other components
of enterprise systems (ES), in particular specialised supply chain management systems
(SCMS). Already back in 2001, a survey (n 89) in the US found that 59 percent of the
respondents either had already or were planning to extend their ERPS with a
SCM-solution. Of the respondents, 84 percent indicated that for any such expansion
they would prefer to use the SCMS provided by their ERP-vendor (CIO Magazine,
2001). SCM-software packages are more specialised forms of ES which integrate all
logistics functions within organisations and also support inter-organisational business
and systems integration (Tarn et al., 2002). In contrast to ERPS, SCMS usually contain
advanced planning systems (APS) functionality that supports mathematical
optimisation procedures and heuristics. Used in combination with ERPS, they partly
substitute and partly complement ERPS functionality, with the level of
substitution/complementation being primarily determined by the functionality of
either package and partly by the preferences of the organisation using them
(Davenport and Brooks, 2004; Stefanou, 2001).
We expect that the advanced features of SCMS and their functional focus lead to
higher performance along the primary activities of the value chain (Porter, 1985), and
used in conjunction with ERPS, to higher overall firm performance.
Business process performance measures
All major ERP-products on the market support logistics processes and methods in
operations management as part of their core-functionality. Therefore, we expect that
the adoption of ERPS leads to better performance in terms of both primary and
secondary[2] (support) business processes in the value chain.
Financial KPIs are doubtlessly the most important and most widely used KPIs for
firm performance, and they can also provide some valuable insights into business
process performance (Dehning and Richardson, 2002; Matolcsy et al., 2005). These
insights are, however, limited, especially if only publicly available data (from annual
reports, etc.) are used. These highly aggregated data can only be used for a very
general analysis of core processes in the value chain, e.g. the ability to turn over
inventory quickly, but do not provide deeper insights into performance at the business
process level (Kaplan and Norton, 1992). Obtaining these more detailed data on
business process performance, however, requires research beyond secondary data
analysis, i.e. a field study.
The operations management and supply chain management (SCM) literatures
provide many examples of SC-performance measurement, and the supply-chain
operations reference model (SCOR-model) published by the Supply Chain Council
(www.supply-chain.org) is considered to be the most established model for measuring
performance along this dimension; it will therefore be used for identifying and
measuring business process performance in this study. The model comprises a mix of
The impact of
ERP systems
17
JEIM
19,1
18
Study
Table I.
Financial and market
performance
measurements used in
prior literature
Measures
used
Number of
employees/
revenues
Return on assets
Return on assets
Return on
investments
Asset turnover
Return on sales
Current ratio
Asset turnover
Notes: SG&A = Selling, general and administrative expenses; COGS Costs of goods sold.
(The measures in italics Oare deemed to be firm performance measures.)
Predictions
As mentioned in the introduction, our key objective is to investigate impacts of several
aspects of ERPS adoption on firm performance and business process performance. We
use ERPS adoption, ERPS history and ERPS extension with a SCMS as independent
variables and KPIs for supply chain performance and firm performance as dependent
variables.
The model shown in Figure 2 comprises all main relationships investigated in our
study. Our model development leads to the following predictions:
P1a.
We predict better firm performance for ERPS users than for non-ERPS
users.
P1b.
For ERPS users we predict that the longer the experience with ERPS,
the better their firm performance.
P1c.
P2a.
We predict better supply chain performance for ERPS users than for
non-ERPS users.
P2b.
For ERPS users we predict that the longer the experience with ERPS,
the better their supply chain performance.
P2c.
The impact of
ERP systems
19
Figure 2.
Research model
JEIM
19,1
20
Table II.
Industry classification of
respondents
(%)
SCMS
Diversified resource/mining
Others (primary sector)
Energy
Total: primary sector
Assembly-line production
Batch production
Process production
Project/unit production
Total: secondary sector
Retail/wholesale and transport
Consulting
Financial services, banking/finance, insurance
Healthcare, tourism and leisure
Media, telecommunications, software, IT-services
Property, infrastructure and facilities
Others (services sector)
Total: tertiary sector
15.7
1.0
2.9
19.6
5.9
13.7
10.8
1.0
31.4
18.6
2.0
11.8
6.9
2.9
1.0
5.9
49.0
U
U
U
U
U
U
U
U
U
U
The impact of
ERP systems
21
JEIM
19,1
22
test for normal distribution of the test group and control group was negative,
nonparametric two-independent sample tests (Mann-Whitney U test and two-sample
Kolmogorov-Smirnov test) were used.
Table III summarises the results of our tests of P1a and P2a. Despite the fact that
ERPS users on average scored higher in all dimensions of performance, we could not
find any significant differences between the two samples, neither in terms of firm
performance, nor in terms of performance along the supply chain. Even a 1-tailed
interpretation of our tests does not change this fact.
These overall results are partly contradicting those of recent studies quoted in the
literature review, which actually concluded that there are some significant differences
in financial performance between ERPS adopters and non-adopters (Hunton et al., 2003;
Nicolaou et al., 2003; Matolcsy et al., 2005). What was even more surprising was the fact
that ERPS users did not score significantly higher than the control group in the area of
logistics/supply chain management, one of the core domains of ERPS. A closer look at
all individual scores within the aggregate measure supply chain KPIs revealed that
ERPS users did actually score higher in all 23 individual KPIs (see Appendix), but it
also shoed that these differences were only significant in the following four individual
categories (at the 0.05-level): Relative level of manufacturing labour costs, relative level
of finished goods inventory holding costs, time required to produce an item and
percentage of shipping errors.
In the case of firm performance indicators, the ERPS users also showed higher
means in all individual categories, but none of these differences were significant. We
expected further insights into the reasons for the partly surprising results by testing
for performance differences with regards to ERPS history and ERPS adoption (use).
ERPS history and performance (P1b and P2b). Three types of tests were performed
to identify differences between companies with longer ERP-experience and more recent
adopters. In the first test, an ERPS history scale was used which measures the number
of months between the go-live date and the day the survey was completed. As the
observations on this scale were normally distributed, we used standard correlations
(Pearson, Kendall and Spearman) to test our predications.
The results of this first test confirm our predictions with regards to the relationship
between ERPS history and firm performance, but not about the impact of ERPS history
on supply chain performance.
Grouping variable: ERPS use
Table III.
Mann-Whitney U test and
two-sample
Kolmogorov-Smirnov test
for ERPS/non-ERPS
Mann-Whitney U
Wilcoxon W
Z
Asymp. Sig. (two-tailed)
n
Kolmogorov-Smirnov Z
Asymp. Sig. (two-tailed)
n
Note: a Only SCM-companies included
841.5
1,661.5
2 0.841
0.400
87
0.717
0.683
87
242.5
395.5
21.073
0.283
52
0.847
0.470
52
In the second test, we split the group of ERPS users into two sub-groups of mature
ERPS users and recent ERPS adopters. We used the median (five years) as a cut-off
point, which also complies with recent research which suggests that there is a time-lag
of benefit realisation and that the positive impact on performance starts to become
significant after approximately two years and tends to increase over years. The results
of the nonparametric two-sample tests confirmed the results of the correlations shown
in Table IV.
To gain further insights into the time-dimension of performance of ERPS, we used
trend analysis (linear and polynomial), as shown in Figure 3.
The performance impact of ERPS-history as indicated by the linear trendline is
approximately 6 percent per annum, which is quite remarkable. The third order
polynomial trendline partly confirms existing literature on the short and long-term
performance implications of ERPS implementations, namely, a performance dip after
going live, a recovery and finally a stage of net benefit realisation. In contrast to
ERPS use in months
Pearson correlation
Correlation coefficient
Sig. (one-tailed)
n
0.394 * *
0.004
44
0.173
0.143
40
Kendalls tau_b
Correlation coefficient
Sig. (one-tailed)
n
0.243 *
0.012
44
0.073
0.278
33
Spearmans rho
Correlation coefficient
Sig. (one-tailed)
n
0.382 * *
0.005
44
0.061
0.368
33
Notes: *Correlation is significant at the 0.05 level (one-tailed); * *Correlation is significant at the 0.01
level (one-tailed); a Only SCM-companies included
The impact of
ERP systems
23
Table IV.
Correlations for ERPS
history and performance
Figure 3.
ERPS-history and firm
performance
JEIM
19,1
24
existing literature which stipulates a recovery period of about two years (Hunton et al.,
2003; Nicolaou et al., 2003; Matolcsy et al., 2005), the data collected from our sample
suggests that firms recover by far later, namely, approximately four to five years after
the implementation.
We acknowledge that the R2 for both trendlines is quite low, which suggests that
there are many other factors which have an impact on performance. However,
ERPS-history clearly has some impact, and it seems that firms need much longer for
recovering from the aftershock of ERPS implementations than often stipulated.
ERPS extensions (SCMS) and supply chain performance (P1c and P2c). For the
reasons mentioned before, the weak impact of ERPS history on supply chain
performance was surprising. So we hoped to gain further insights into this result by
comparing the performance scores of ERPS users with SCMS (test group) and without
(control group).
The fact that only 12.3 percent of all ERPS users had also SCMS in place limits the
statistical explanatory power of our analysis. However, the results of our analysis
might still give useful indications and provide paths for future research. The results of
the independent sample statistics are depicted in Table V. In the absence of positive
results from tests for normal distribution, nonparametric two-independent sample tests
were used again.
These results provide an indication that ERPS users that also adopted SCMS
perform by far better along the supply chain (mean 5.0) than ERPS users that did not
use additional SCMS (mean 4.24). However, we could not confirm prediction 1c,
namely, that the adoption of a SCMS in a ERPS user company also directly leads to
firm performance increases.
Supply-chain performance and firm performance (P3). Prediction 3 was tested only
for control purposes rather than for gaining new insights, because unless important
additional factors impact negatively on firm performance, high performance in the core
business processes would very likely lead to higher firm performance. Correlation
results indeed confirmed our predication (one-tailed):
.
Pearson correlation: 0.363, p 0.007;
.
Kendalls tau_b: 0.213, p 0.024; and
.
Spearmans rho: 0.304, p 0.022.
Grouping variable: SCMS & ERPS use
Table V.
Extended ERPS (ERPS
SCMS)
Mann-Whitney U
Wilcoxon W
Z
Asymp. Sig. (two-tailed)
n
Kolmogorov-Smirnov Z
Asymp. Sig. (two-tailed)
n
130.0
964.0
20.640
0.522
6
0.670
0.761
6
34.5
737.5
22.253
0.024 *
6
1.418
0.036 *
6
Notes: * Significant at the 0.05 level (one-tailed); * * Significant at the 0.01 level (1one-tailed)
The impact of
ERP systems
25
Figure 4.
Summary of main results
JEIM
19,1
26
processes (supply chain) than companies with a shorter ERPS history (and non-ERPS
adopters). We can summarise that the ERPS-users in our study failed to achieve higher
supply chain performance, both in the short and long-term, and they failed to achieve
higher overall firm performance, although the latter improved with time since the
adoption (learning curve). The shape of this learning curve suggests that it takes
companies (in Australia) much longer to recover from the aftershock of
ERP-implementations than often suggested.
Finally, we found some evidence supporting our prediction that extending ERPS
with additional solutions for supply chain management actually has a positive impact
on an organisations performance along the supply chain.
We acknowledge that there are several limitations to our research, the most
important one being the low response rate to our survey and the small sample size. We
are aware that this limits explanatory power of our results.
We also made some observations which are not explicitly included in our analysis,
but which might also create incentives for further research. We noticed, for example,
that there is some confusion about the term ERPS and its relation to other forms of
enterprise systems. We assume this confusion is stimulated by the ERP-vendors in two
ways: On the one hand, we found that more and more vendors of small functional
software packages seem to use the label ERP inappropriately, e.g. for small
accounting packages with integrated inventory management. On the other hand, it is
observed that the large ERP-vendors SAP, Oracle and Peoplesoft start to change their
interpretation and use of the term ERP and package their traditional products in a
variety of solutions, which make it increasingly difficult to classify them as ERP, SCM,
CRM, etc. In fact, future research in enterprise systems will have to redefine its
underlying typologies of enterprise systems based on this ongoing changes in the
business systems landscape. This will open a pathway for research into, e.g. the
optimum scope of integration and advantages and disadvantages of various
integration approaches (ERPS, SCMS, CRMS, DW, Enterprise Application Integration,
etc.).
Notes
1. Similar examples are provided by Lee and Lee (2000) and Tarn et al. (2002).
2. Secondary (support) processes such as accounting, IT, human resource management, are not
considered in this study.
3. The individual performance indicators are listed in the Appendix.
4. The measure was, however, relative in the sense that we asked respondents to benchmark
their company against the industry.
5. The fact that 269 questionnaires were returned by the receivers because of the addressed
companies having moved location gave reason to assume that a very large number of other
questionnaires has not reached the addressee, but was not voluntarily returned by the
receiver. This assumption was confirmed in the second stage of the survey, a telephone
survey, which revealed that many of the addressees selected in this second stage had never
received the questionnaire by mail, although we had not got their letters back marked
return to sender.
References
Beretta, S. (2002), Unleashing the integration potential of ERP systems: the role of process-based
performance measurement systems, Business Process Management Journal, Vol. 8 No. 3,
pp. 254-77.
Bhatt, G.D. (2000), An empirical examination of the effects of information systems integration
on business process improvement, International Journal of Operations & Production
Management, Vol. 20 No. 11, pp. 1331-59.
Booth, P., Matolcsy, Z. and Wieder, B. (2000), The impacts of enterprise resource planning
systems on accounting practice the Australian experience, Australian Accounting
Review, Vol. 16 No. 1, pp. 4-18.
CIO Magazine (2001), ERP progress report, available at: www2.cio.com/research (accessed
13 May 2004).
Davenport, T.H. and Brooks, J.D. (2004), Enterprise systems and the supply chain, Journal of
Enterprise Information Management, Vol. 17 No. 1, pp. 8-19.
Dehning, B. and Richardson, D.J. (2002), Returns of investments in information technology:
a research synthesis, Journal of Information Systems, Vol. 16 No. 1, pp. 7-30.
Hunton, J.E., Lippincott, B. and Reck, J.L. (2003), Enterprise resource planning (ERP) systems:
comparing firm performance of adopters and non-adopters, International Journal of
Accounting Information Systems, Vol. 4 No. 3, pp. 165-84.
Irani, Z. (2002), Information systems evaluation: navigating through the problem domain,
Information and Management, Vol. 40 No. 1, pp. 11-24.
Kaplan, R.S. and Norton, D.P. (1992), The balanced scorecard measures that drive
performance, Harvard Business Review, January-February, pp. 71-9.
Lee, Z. and Lee, J. (2000), An ERP implementation case study from a knowledge transfer
perspective, Journal of Information Technology, Vol. 15 No. 2, pp. 281-8.
Markus, M.L., Axline, S., Petrie, D. and Tanis, C. (2000), Learning from adopters experience
with ERP problems encountered and success achieved, Journal of Information
Technology, Vol. 15 No. 4, pp. 245-65.
Matolcsy, Z., Booth, P. and Wieder, B. (2002), The economic benefits of enterprise resource
planning systems: some empirical evidence, paper presented at the AAANZ Conference,
Perth, July.
Matolcsy, Z., Booth, P. and Wieder, B. (2005), The economic benefits of enterprise resource
planning systems: some empirical evidence, Journal of Accounting and Finance
(forthcoming).
Nicolaou, A.I., Stratopoulos, T. and Dehning, B. (2003), Financial analysis of potential benefits
from ERP systems adoption, Journal of Business and Information Technology, Vol. 2 No. 1,
pp. 40-50.
Oracle (2003a), E-Business suite, Vol. 8, available at: www.oracle.com/applications/ (accessed
8 August 2003).
Oracle (2003b), Oracle customers, available at: www.oracle.com/customers/index.
html?ebsquotes.html/ (accessed 8 August 2003).
Peoplesoft (2003), available at: www.peoplesoft.com/corp/en/products/line/scm/inv_plan/
(accessed 8 August 2003).
Porter, M.E. (1985), Competitive Advantage: Creating and Sustaining Superior Performance, Free
Press, New York, NY/London.
The impact of
ERP systems
27
JEIM
19,1
28
The measurement of the SCM-KPIs was based on the following question: Please rate
your firms performance in the following attributes compared to other similar firms
(1 not at all satisfactory, 4 satisfactory, 7 outstanding):
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Corresponding author
Bernhard Wieder can be contacted at: Bernhard.Wieder@uks.edu.au
The impact of
ERP systems
29