Towards a General Framework for Business Tests
Marijke Swennen
1
, Benoît Depaire
1,2
, Koen Vanhoof
1
and Mieke Jans
1
1
Business informatics research group, Hasselt University, Agoralaan Building D, 3590 Diepenbeek, Belgium
2
Research Foundation Flanders (FWO), Egmontstraat 5, 1000 Brussels, Belgium
Keywords: Business Test, Problem Definition, Performance, Compliance, Risk, Design Science.
Abstract: Testing and controlling business processes, activities, data and results is becoming increasingly important
for companies. Based on the literature, business tests can be divided into three domains, i.e. performance,
risk and compliance and separate domain-specific frameworks have been developed. These different
domains and frameworks hint at some aspects that need to be taken into account when managing business
tests in a company. In this paper we identify the most important concepts concerning business tests and their
management and we provide a first conceptual business test model. We do this based on an archival
research study in which we analyse business tests performed by an international consultancy company.
1 INTRODUCTION
In classical management theory, Fayol (1949)
identified five ‘elements’ of management, among
which Controlling is one. While the validity of
Fayol’s work has been subject of academic debate
(Fells, 2000), there is little arguing that companies
are increasingly confronted with incentives and
obligations to test and control their business (Wade
and Recardo, 2001; Shamsaei et al., 2010). These
business tests entail any kind of test on business
objects such as a process, an activity, an employee
or a product.
The need for business testing stems from different
origins, such as legislation compliance requirements,
Service Level Agreements (SLAs) or performance
management (Wade and Recardo, 2001). We argue
that due to the diverse nature of business test
incentives, management of business tests is often
without a holistic overview, fragmented and as a
consequence possibly inefficient.
The objective of this paper is threefold: we
provide a rationale for a consolidated view of
business tests, a working definition of business tests
and identify various dimensions of business tests.
This research is a first step towards a general
framework to identify, model and manage business
tests.
Section 2 and 3 will respectively describe the
applied research methodology and the data, while
section 4 introduces the rationale for a consolidated
business test framework. Next, section 5 will discuss
the definition and several dimensions of business
tests. Conclusions and future work will be discussed
in section 6.
2 RESEARCH METHODOLOGY
The research described in this paper is part of a
bigger research project with the objective to develop
models, methods and a language for a holistic
business test framework. The nature of these overall
research objectives demand for the scientific
paradigm of Design Science research (Hevner et al.,
2004, Peffers et al., 2007). This paper presents the
first steps of this paradigm, i.e. the problem
identification, the research motivation and a first
iteration to define a business test artefact.
All results were derived through literature review
and archival research. Firstly, a literature review of
existing work on business tests was performed
according to the steps by Fink (2005). Based on the
gained insights, the business test artefact was
developed.
Next, archival research on a set of 156 business
tests, provided by 4 different business units of an
international consultancy company, was performed
to evaluate to what extent the different elements of a
business test artefact can and have been applied.
The archival research findings were further
enriched by qualitative interviews with various
478
Swennen M., Depaire B., Vanhoof K. and Jans M..
Towards a General Framework for Business Tests.
DOI: 10.5220/0004969204780483
In Proceedings of the 16th International Conference on Enterprise Information Systems (ICEIS-2014), pages 478-483
ISBN: 978-989-758-029-1
Copyright
c
2014 SCITEPRESS (Science and Technology Publications, Lda.)
employees actively working with business tests at
different levels in the business test cycle, such as
developing, implementing, performing, evaluating
and improving business tests. The seven stages of
conducting interviews, stated by Kvale and
Brinkmann (2008), were applied.
3 DATA
For our archival research study, we selected an
international consultancy company as our case study
as it consists of different business units performing
business tests for different reasons. In total 4 units
were selected, i.e. Audit & Risk, Forensics, Tax and
Consulting, from which a convenience sample of
156 was drawn. Please note the exploratory nature of
this research which justifies the use of a convenience
sample.
The tests performed in the first business unit,
Audit & Risk, define the risk level of a client
company. The second business unit performs all
tests related to Forensics, which concerns the risk
and compliance level of a company. Tax is a
business unit in which all kinds of tests concerning
VAT and applied VAT rates are performed. These
are defined as compliance tests that result from rules
or laws the client companies need to obey to.
Finally, the Consulting business unit performs tests
to provide advice to client companies about different
elements such as pricing or customer orientation.
These tests can be categorized as performance tests.
The number of tests performed by each business
unit, is given in table 1.
Table 1: Overview of all tests per business unit in the case
study.
Business Unit
Number of tests
Audit & Risk 31
Forensics 75
Tax 36
Consulting 14
Total 156
4 AN INTEGRATED BUSINESS
TEST FRAMEWORK
Research on business tests has been done in three
business domains, i.e. performance management,
compliance management and risk management.
Within each domain, several separate models and
frameworks have been developed. For a
comprehensive overview of trends and different
frameworks in each domain, the reader is
respectively referred to Bourne (2001) or Bititci et
al. (2011), Mahmoud (2010) and O’Donnell (2005).
The fact that each research domain presents its
own frameworks is also reflected in the business
reality where an integrated approach towards
business tests is often lacking. Although each
business unit of the company in our archival
research study identifies its tests as exclusively
related to either performance, compliance or risk,
some tests can actually be assigned to multiple
domains. However, two lines of arguments can be
developed in favour of a more integrated approach.
Firstly, from a management perspective,
isolating business test efforts from each other, could
lead to inefficiencies. Shamsaei et al. (2010)
mention that different business rules might be
correlated and actions to improve the results for one
rule might have side effects on other rules. Being
unaware of such correlations leads to
suboptimisation. Also, measuring and evaluating
highly similar and even duplicate tests for different
perspectives, will create an administrative burden.
Furthermore, Bardoliwalla et al. (2009) state that
risk management is often hampered by
organizational silos, causing a lack of consistent
taxonomies, measurement, and reporting. This
results in obscured visibility, preventing managers to
obtain a true picture of the overall enterprise. It is
expected that their observation gains even more
importance when expanding to the full range of
business tests.
A second line of arguments in favour of an
integrated business test framework originates from
the information systems perspective. Business tests
deal with measuring and gathering the correct data
and providing it timely to the right manager,
preferably in an automated way. Consequently,
business tests should be an integrated part of the
information architecture and have a direct impact on
the data and system requirements. Strangely, while
several frameworks, models and languages exist to
provide an overview of the data architecture and the
business process architecture, no such framework
exists for business tests. Consequently, if a manager
asks the IT department for the implementation of a
new business test, the development team lacks a
complete overview of the existing business tests.
The idea of an integrated approach is not entirely
new and recently a few researchers have hinted at a
similar idea. A first attempt is the GRC concept,
TowardsaGeneralFrameworkforBusinessTests
479
which combines governance with risk and
compliance management and is an emerging topic in
the business domain (Racz et al., 2010).
Bardoliwalla et al. (2009) further combine the GRC
concept with performance management, merging the
main three sources of business tests. While the work
of Bardoliwalla et al. (2009) unifies the three
concepts of performance, risk and compliance into a
coherent strategic management process framework,
they do not provide an integrated framework how to
design, implement and document business tests.
5 THE CONCEPTUAL MODEL
FOR A BUSINESS TEST
A test can be defined as the evaluation of a
measurement against a predefined target, where
the evaluation process results in a conclusion
about the measured object.
For example, to test whether we should stop at
the current gas station entails measuring the current
fuel level and evaluating it against the fuel level
target required to drive to the next gas station. Based
on this evaluation we can conclude if we make it to
the next station.
A Business Test is then a test that relates to any
kind of business object, such as e.g. a product or
process, and is typically related to some kind of
performance, risk or compliance purpose (or a
combination of them as we discussed in section 4).
Note that we define a business test at the lowest
(most detailed) measurement level.
Figure 1: The conceptual model of a business test.
With this general definition of a business test, we
proceeded to analyse the literature to develop a
conceptual model for a business test, which is
illustrated in Figure 1. We identified concepts from
the performance, compliance and risk literature that
were relevant to a general business test concept.
Next, archival research evaluated if the different
elements of a business test made sense in practice.
Simultaneously, we explored to what extent our set
of business tests were complete in the sense of our
conceptual model.
Business Object
5.1
From our conceptual model in Figure 1 we can argue
that a business test should always be indirectly
linked to a business object via a measurement. A
measurement is an observation of a business object,
i.e. the number of steps in a process or the average
weight of a specific product, but is never a
comparison between two things, which is the
evaluation. A measurement is always related to only
one object but more than one measurement can be
necessary to perform a test. As a result, a single
business test can be concerned with several business
objects which means that several business tests may
be required to evaluate a specific aspect of the
company. The evaluation is the formula or metric
and the target with which the measured value will be
compared to define a conclusion.
For almost all business tests performed at the
international consultancy company a link with a
business object can be found. The Forensics and
Consulting business units, for example, both define
eleven business tests to check the quality of master
data of a client company. This implies that different
tests are performed to evaluate the quality of master
data. Furthermore, in the Audit & Risk business unit,
all tests perform a check on data from the purchasing
process in client companies. For each test a certain
measure of this process is stored. To conclude we
can state that for almost all tests at the different
business units of our case study company a link to a
business object is defined.
5.2 Business Goal
One of the frameworks in the performance
measurement domain, defined by Neely et al.
(2000), states that performance measures should
always be linked to the company’s strategic
objectives. Following this thought, we can argue that
business tests should be connected to one or more
business objectives or goals and that these goals
should be properly documented.
Most of the tests in the different lists of our
archival research do not include any objective or
goal. Only the list of tests from the Forensics
business unit includes for every set of tests an
ICEIS2014-16thInternationalConferenceonEnterpriseInformationSystems
480
objective. However, after analysis, these objectives
are found to be explanations of how the tests will be
performed and cannot be regarded as business goals.
For example, for the quality of master data tests, the
given objective is to identify missing critical master
data for customers and suppliers. This only gives an
overview of the different checks that will be
performed. From this we can conclude that the
objectives of the business tests in our sample are not
documented. Some additional questions at the
consultancy company tell us that the overall goal of
the different business units is to make sure the client
company is satisfied with the delivered service.
Furthermore, the interviews tell us that at Audit &
Risk the objectives are mostly to look for fraud and
irregularities in client data. This is something they
only started documenting very recently and is not
standardized or implemented yet.
5.3 Evaluation Scales
We defined a test as the evaluation of a
measurement against a predefined target, where the
evaluation process results in a conclusion about the
measured object. The risk management framework
developed by Kaplan and Mikes (2012) presents a
categorization of risks that allows to tell which
actions to take for and how to evaluate each risk
category. The first category are preventable risks,
which arise from within the organization. They are
controllable, should be eliminated or avoided and
are best managed through active prevention. This
implies a binary evaluation, the risk should be
corrected immediately if it occurs.
Category two are strategy risks that are quite
different from preventable risks because they can be
assumed to be not inherently undesirable.
Sometimes it is required to take on significant risks,
and managing those risks can be an important factor
in obtaining potential benefits. A risk management
system is required to reduce the probability that the
assumed risks occur and to improve the ability to
contain or monitor the ratio between performance
and risk. Here a ratio evaluation is implied which
requires measurements of different objects.
The third category are risks which arise from
events outside the company and are therefore
beyond its influence or control. Causes may be
natural or political disasters and major
macroeconomic shifts. With an interval evaluation
management should focus on measuring and
mitigating the impact of these risks because they
cannot be prevented from occurring.
At the Tax business unit all tests are evaluated
with ‘ok’ or ‘not ok’. This implies no toleration or
trade-off for the results of these tests. In the list of
tests from Forensics, some guidance steps for the
client company are provided in case risky or
unreliable results occur. All test results are evaluated
by checking their importance or impact, which
implies a ratio evaluation. From this we can
conclude that for some business units of our archival
research company a notion of ways to evaluate the
results is present. However, this is only true for two
of the four business units in the study.
5.4 Follow-up Actions
Based on the different categories of risk presented in
the risk management framework of Kaplan and
Mikes (2012), we can argue that all categories have
one thing in common. For a proper management of
risks not only measuring their impact is of
importance, also the identification of which actions
to take in mitigating or managing these risks should
be included. This refers to the conclusion in our
general definition of a business test. We can transfer
this idea to the business concept by stating that for
every business test the possible follow-up actions
and triggers to activate these actions should be
defined. Based on the three types of risks, we see
that different types of risks can have a different
influence on how each test should be evaluated and
which follow-up action should be performed. We
can add the notion that a follow-up action of a
business test can be another business tests. This is
also discussed in section 5.6.
In the Forensics and Tax business units we find
that for only a small amount of tests follow-up
actions are included. At Tax, most of the defined
follow-up actions are manual checks or the delivery
of a list of transactions with standout results to the
client company. In five of the tests performed by the
Tax business unit a business test is followed by
another business test if a certain result occurs.
However, there is no documentation about which
actions to take to prevent, manage or mitigate certain
risk events.
5.5 Weights
In the compliance management domain we find the
approach developed by Shamsaei et al. (2010) which
enables organizations to measure the current
compliance level of their processes and track down
and analyse compliance problems. Measuring the
importance of organizational rules allows to
distinguish the most important problems that need to
TowardsaGeneralFrameworkforBusinessTests
481
be assigned first from the less important problems.
By expanding these conclusions to the business test
concept, we can argue that giving weights to
business tests can provide an insight in which tests
and which resulting outcomes need to be covered
first. Moreover, an overall measure of compliance
can be found by providing more important tests with
a higher weight.
The notion of different weights given to business
tests could not be found in the lists of business tests
we received from the different business units of the
consulting company. The tests are presented as
being all at the same level, implying that they all
have the same weight or importance. Some
additional questions at the company tell us that these
weights and their documentation appear to be
interesting but are not yet developed or
implemented.
5.6 Intertest Relationships
Besides the notion of weights, Shamsaei et al.
(2010) also argue that more than one rule possibly
applies to a single process, and hence a change to
enhance the compliance level of one rule may have
side effects on the compliance level of other rules.
Transferred to the concept of business tests, we can
argue that the result or outcome of a business test
can be influenced by the implementation or the
result of another business test. This adds to the
findings that business tests can be follow-up actions
of other business tests. In the conceptual model in
Figure 1, this interest relationship is indirectly given
by the many-to-many relationship between objects
and tests. The business objects represent the link
between different business tests.
We already mentioned that some business tests
from the international consulting company are
followed by other business tests, but there is no
information about the influence of different business
tests and their results on each other.
6 CONCLUSIONS
From the literature review we can infer that the
different business test domains, i.e. performance,
compliance and risk, are only recently brought in
connection to each other. Most existing frameworks
concern only one of these domains. However, in this
research we found that some business tests are not
exclusively assignable to just one of these domains.
Furthermore, we noticed that in many cases,
business testing and the management of these tests
fall victim to organizational silos, lacking consistent
taxonomies, measurement, and reporting, which
obscures visibility.
In this first step of our overarching research
project to develop a holistic business test framework
we define a business test as an evaluation of a
measurement against a predefined target, where the
evaluation process results in a conclusion about the
measured object. Furthermore we state that business
tests are grouped around a business object and
connected to a higher business objective. The
evaluation scale, weight and follow-up actions,
which can be other business tests, should be defined
and intertest relationships may be present.
The need for a proper management of business
tests becomes clearer when adding the findings from
the archival research study. In this study we found
that the objectives of the business tests are not
documented properly and that no weights are
provided to the business tests as suggested by the
literature. Besides that, only a small amount of the
tests in the study include a notion of follow-up
actions and ways to evaluate the results of the tests.
However, most of the tests can be divided in
different groups around objects in the company.
Finally, we can conclude that valuable information
is trapped in the different departments of the
organization and is not aggregated with information
from other departments. Especially for the steps
before actually performing the tests, the data
collection and data cleaning, a lot of redundancy is
present.
In general, these results empirically validate the
need for an integrated framework for defining and
implementing business tests, as only some of the
business test elements are present.
After performing this first step in the Design
Science Research Methodology Cycle, the next steps
can be executed. In cooperation with some experts
working with the business tests in our archival
research the tests can be transferred into our
conceptual model. Appropriate enhancements or
modifications can be implemented.
However some challenges and different
perspectives can provide an even better basis for this
first step. First of all, we can assume that different
companies can be in a different maturity level in
terms of the development and implementation of
business tests and the structure of business tests.
Also, the same research can be carried out on tests
performed at companies on their own data instead of
consultancy companies who perform tests at client
data.
ICEIS2014-16thInternationalConferenceonEnterpriseInformationSystems
482
REFERENCES
Bardoliwalla, N., Buscemi, S., Broady, D., 2009. Driven
to Perform: Risk-Aware Performance Management
From Strategy Through Execution. Evolved
Technologist Press. New York, 1
st
edition.
Bititci, U, Garengo, P., Dorfler, V., Nudurupati, S., 2011.
Performance measurement: challenges for tomorrow.
In International Journal of Management Reviews 14
(3). p. 305–327.
Bourne, M., 2001. Handbook of Performance
Measurement. Gee Publishing Limited, London.
Fayol, H., 1949. General and Industrial Management. Sir
Isaac Pitman & Sons Ltd. London.
Fells, M. J., 2000. Fayol stands the test of time. In Journal
of Management History, 6 (8), p. 345–360.
Fink, A., 2005. Conducting Research Literature Reviews.
From the Internet to Paper. Sage Publications. 2
nd
edition.
Hevner, A. R., March, S. T., Park, J., Ram, S., 2004.
Design Science in Information Systems Research, In
MIS Quarterly, 28 (1), p. 75-105.
Kaplan, R. S., Mikes, A., 2012. Managing Risks: A New
Framework. In Harvard Business Review. 90 (6), p.
48-60.
Kvale, S., Brinkmann, S., 2008. InterViews: Learning the
Craft of Qualitative Research Interviewing. SAGE
Publications, Inc.
Mahmoud, A., 2010. A compliance management
framework for business process models. Ph.D. Thesis.
Germany, University of Potsdam.
Neely, A., Bourne, M. Kennerley, M., 2000. Performance
measurement system design: developing and testing a
process-based approach. In International Journal of
Operations & Production Management. 20 (10). p.
1119-1145.
O’Donnell, E., 2005. Enterprise risk management: A
system-thinking framework for the event identification
phase. In International Journal of Accounting
Information Systems. 6 (3). p. 177-195.
Peffers, K., Tuunanen, T., Rothenberger, M., Chatterjee,
S., 2007. A Design Science Research Methodology for
Information Systems Research, Journal of
Management Information Systems. 24 (3). p. 45-77.
Racz, N., Weippl, E., & Seufert, A., 2010. A frame of
reference for research of integrated governance, risk
and compliance (GRC). In Communications and
Multimedia Security, p. 106–117.
Shamsaei, A., Pourshahid, A., Amyot, D., 2010. Business
Process Compliance Tracking Using Key Performance
Indicators. In 6th Int. Workshop on Business Process
Design (BPD 2010), BPM 2010 Workshops, LNBIP
66, Springer, p. 73-84.
Wade, D., Recardo, R., 2001. Corporate Performance
Management: How to build a better organization
through measurement-driven, strategic alignment.
Butterworth-Heinemann.
TowardsaGeneralFrameworkforBusinessTests
483