A METHODOLOGY TO FINALIZE THE REQUIREMENTS FOR
A PROJECT WITH MULTIPLE STAKE- HOLDERS
Presenting Software Engineering Workshop as a Solution
Ashutosh Parashar and Selvakumaran Mannappan
Cognizant Technology Solutions, SDBIII, 5/535, Old Mahabalipuram Road, Chennai 600096, India
Keywords: Software Requirements, Software scope definition, Software Engineering Workshop, Requirements
Gathering, Requirements prioritization, stakeholder alignment.
Abstract: Implementing software projects for large corporations,
more often than not, involves large number of
stakeholders, each with their own set of requirements, which makes requirements finalization very difficult.
The authors propose Solution Envisioning Workshop (SEW) as a solution and present the practice from the
context of a large project executed for a European banking giant. The project had a very large and diverse
set of stake-holders- around 300 member banks as the client organizations, interfacing requirements with
around ten separate systems/ projects, active involvement of central departments of the organization as
active stakeholders. The paper elaborates on the approach taken towards implementing the SEW, the
preparatory & follow-up activities, the benefits, limitations and the lessons learnt. They conclude that the
SEW approach results in creating better understanding, much faster requirement finalization. Quantitative
and qualitative inputs are provided to corroborate the findings.
1 THE PROBLEM CONTEXT
Scope management, requirements prioritization and
ensuring user acceptance at the end of the project are
constant concern from requirements perspective.
This is more so in the case of large software project
that affect one entire department or the complete
enterprise, due to the simple fact that, each such
software project implementation results in affecting
large number of organizational resources, processes
and practices. Each of these entities becomes a
stakeholder in the project success and joins the
project with their own specific agenda. As the
number of stake holders increases, the complexity
the requirement management process also increases.
In some cases this is because the requirements from
various stakeholders are contradictory but mostly it
is because there is lack of agreement on the
priorities of the requirements. Since for each of
the stakeholder their own set of requirements are the
most desirable ones and a project can satisfy only a
limited set of requirements, the task of scope
definition and requirements finalization becomes
very difficult. And if the project fails to satisfy any
of the requirements of any of the stakeholders, the
acceptance of the project becomes uncertain,
resulting in continuous scope creep and time, effort
overruns and even project scrapping/ failure.
2 INTRODUCING SOLUTION
ENVISIONING WORKSHOP
Enter Solution Envisioning Workshop or SEW.
SEW is a systematic approach to counter the
problem of requirements finalization in the above
context. It gathers all key stakeholders together for a
short but intensely focused period (Leffingwell,
Widrig 1999). This results in better understanding
and faster decision on issues.
The SEW makes the process much more
syste
matic and ensures that the maximum benefits of
the exercise can be taken out with minimum effort
and time. SEW is a rapid, high impact process
designed to align the organizational stakeholders
behind a business opportunity, the requirements that
will enable that opportunity, and a credible delivery
plan (Please refer to the figure 1). The SEW
approach is ideal for certain software project
scenarios as given in the figure 2.
449
Parashar A. and Mannappan S. (2007).
A METHODOLOGY TO FINALIZE THE REQUIREMENTS FOR A PROJECT WITH MULTIPLE STAKE- HOLDERS - Presenting Software Engineering
Workshop as a Solution.
In Proceedings of the Second International Conference on Software and Data Technologies - SE, pages 449-452
Copyright
c
SciTePress
Figure 1: SEW Triple convergence of the benefits.
Figure 2: SEW usage scenarios.
3 INTRODUCING THE PROJECT
AND COMPLEXITY
The SEW approach is discussed here from the
perspective of a project implemented by authors’
organization. This was large development project for
a European bank. The proposed system was a
centralized, loan processing system having
automated interfaces with multiple systems.
From the perspective of the requirements
management and other issues, the project complexity
was so severe that the earlier attempts had failed two
times. This complexity came mainly from large
number of stakeholders and complex interfacing
requirements with the adjoining systems/ projects
(need to finalize and constantly update the system-
interface contracts).
An important factor that increased the number of
stakeholders was the ‘consortium structure’ of the
bank, which consists of a network of around 300
member banks, the Central Organization and several
daughter organizations. In most cases the member
banks work as independent organizations.
With regards to using and maintaining the legacy
software system also these banks were working
independently. Even though each member bank
started with identical software, gradually difference
of contexts necessitated diverse changes and finally
ICSOFT 2007 - International Conference on Software and Data Technologies
450
they had, more or less, a different system. From
requirement management point, this meant bigger
and very diverse set of requirements.
The project had involvement of a large number of
stakeholders including departments for Central
organization for IT strategy, and the departments for
legal & compliance issues, Quality, Usability etc.
4 SEW IMPLEMENTATION IN
THE PROJECT
The project profile fitted the ‘Major Development
Project’ scenario for SEW implementation. (Please
refer to Figure 2)
Under the SEW approach, the actual workshop to
finalize the requirements takes place towards the end
of the requirement gathering stage. The workshop
requires the largest commitment of time and
resources in the preparation and participation of key
stakeholders whose time is at a premium (Polikoff,
Coyne, Hodgson 2005). A part of the time and effort
is spent on follow-up activities. The figure 3 would
clarify it further.
The Initial Overview stage involved study of the
legacy system, understanding its software context
(relation with adjoining applications), and the
business process behind the system. There were one-
on-one/ group discussions; based on these initial
Software Requirement Overview Document was
prepared. The final finding was presented to all
stakeholders.
The requirements were divided into functional
and non- functional (technical) tracks and there
onwards were pursued separately. Wherever needed,
the tracks were put in sync with each other.
During the requirement gathering, the documents
for these requirements were perfected, by several
rounds of review and rework, to the extent possible
(provisional approved).
To complement the requirements specification, a
prototype was also prepared and updated/ elaborated
continuously. The decision on the scenario to be
depicted in the prototype was made in the beginning
itself.
During final Workshop the output of all this
preparation is presented to all stakeholders. The
discussion was facilitated by the prototype.
The new/ diverging points coming up during the
workshop were handled in the following fashion
Specific observation on part of the system:
Logging the item as a specification defect.
Generic observation on any part of whole of the
system: Listing & prioritizing items.
o High priority: Logged as specification defects.
o Medium priority: Logged as Change requests.
o Low priority: ‘Wish- list’ items. The priority
for items was determined so that release plan
for them can be made.
Figure 3: SEW Approach in the project’s requirements gathering phase.
A METHODOLOGY TO FINALIZE THE REQUIREMENTS FOR A PROJECT WITH MULTIPLE STAKE- HOLDERS
- Presenting Software Engineering Workshop as a Solution
451
5 BENEFITS FROM SEW
In general, the SEW provides the following benefits
for the software projects:
Requirements workshops provide business
value by reducing the time it takes to gather
requirements, by increasing team productivity, and
by reducing risks associated with software projects
(Gottesdiener, 2002).
Build consensus on the problem scope
Validating business goals
Identifying the project objectives
Core functional requirements
From the perspective of SEW implementation in
the project under discussion the benefits are:
On the spot discussion helped in preventing
possibility of later disagreements.
Consent from ‘Senior User’ (representative of
member banks) ensured acceptance by the bank on
the scope decision.
Changing the document status from
‘Provisionally approved’ to ‘Approved’.
The workshop came out with 133 items.
Without SEW, such items would have surface very
resulting in big rework.
Figure 4: Workshop items- number breakup.
Please refer to Figure 5. Around two- third of the
items are of medium/ high severity.
Figure 5: Workshop items- severity.
6 LIMITATIONS OF SEW
The SEW implementation in the project put forward
the following disadvantages/ limitations of the
approach:
Time management: The ideal team for
workshop consists of people having thorough
understanding of processes and hence considerable
experience within the organization/ industry. It’s
very difficult to get such people in one place at the
same time.
Coordination problem: SEW requires active
participation of a project sponsor, who has authority
to make decision in case of persistent disagreement
among participants.
The use of an outside facilitator experienced in
requirements management can help ensure the
success of the workshop. (Leffingwell, Widrig
1999). For large group of stakeholders such
facilitator, with a convincing yet agreeable approach
for persuading stake- holders and coordinating the
proceedings, is rather a must.
Time to ponder: Difficult to have continuous
workshop sessions. Participants need to reflect over
items and even need to discuss it with their other
colleagues.
The elaboration and documentation of the items
discussed in workshop requires effort after
workshop.
7 THE WAY FORWARD WITH
SEW
The authors envision the following improvements:
Staggered workshop: Dividing the workshop
into multiple small workshops.
The workshop approach should have been
utilized for the beginning of the requirements as
well.
REFERENCES
Leffingwell Dean, Widrig Don, October 1999, Managing
Software Requirements: A Unified Approach,
Addison Wesley Professional
Polikoff Irene, Coyne Robert, Hodgson Ralph, July 2005,
Capability Cases: A Solution Envisioning Approach,
Addison Wesley Professional
Gottesdiener Ellen, April 2002, Requirements by
Collaboration: Workshops for Defining Needs,
Addison Wesley Professional
ICSOFT 2007 - International Conference on Software and Data Technologies
452