![](bg5.png)
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.
ICEIS2014-16thInternationalConferenceonEnterpriseInformationSystems
482