See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/303632370

Issues and Challenges In Scrum Implementation

Article  in  International Journal of Scientific and Engineering Research · August 2012

CITATIONS READS

42 6,110

2 authors, including:

Hammad Majeed
National University of Computer and Emerging Sciences
45 PUBLICATIONS   566 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Understanding Citizen Issues through Reviews: A Step towards Data Informed Planning in Smart Cities View project

Smart Living View project

All content following this page was uploaded by Hammad Majeed on 29 May 2016.

The user has requested enhancement of the downloaded file.

International Journal of Scientific & Engineering Research, Volume 3, Issue 8, August-2012 1
ISSN 2229-5518

Issues and Challenges in Scrum Implementation

Akif, R., H. Majeed

Abstract— The aim of this research paper is to bring the challenges and issues in Scrum implementation to light and proposing solutions
for these. For this, a survey is conducted in two companies named Digital Prodigy Limited (DPL) and Bentley Systems Pakistan.
Participants include experienced and inexperienced scrum personals from both companies. The analysis of the survey results exposed
several issues that affect scrum implementation directly or indirectly and resulting in violation of Scrum rules. Quality items pileup, module
integration issues, code quality, disruption in team work, mature vs. immature scrum, sprint duration, lack of Scrum training , release
process, backlog management, no technical practices, multiple teams, metrics, risk management, documentation and over idealistic are
some examples of these issues. During this study, it has been observed that proper training of Scrum can eliminate half of th e issues such
as, disruption in team work, immature Scrum, sprint duration and backlog management. At the end, suggestions to address these issues
are also provided.

Index Terms— Agile, Agile Issues, Agile Methodology, Distributed Scrum, Issues and Challenges in Scrum, Scrum, Scrum Implementation.

—————————— ——————————

1 INTRODUCTION

S oftware Engineering is the technique for the development

of software step by step to produce the high quality prod-
uct at the end. It guarantees the different aspects of soft-
propose solutions to some of them. This research paper focus-
es on the identification of hidden challenges and issues in
Scrum implementation and then the formulation of techniques
ware to work as designed by the engineers and allows the arc- for tackling such issues. It also proposes solutions to these
hitects to improve the quality of the software development issues and challenges mentioned in existing work by [6] such
efforts. It provides different ways to implement the software as documentation, communication, user involvement, scrum
with the help of policies, procedures and processes which is ceremonies and client involvement [13]. It also includes solu-
called software development methodology (SDM). There are tion to social and cultural differences issue which was hig-
various software development methodologies and each me- hlighted by [24].
thodology has its own software development life cycle This research paper also mentions the formal sessions with
(SDLC). When a team adapts any SDLC for the software de- two organizations to highlight and compare nature of issues
velopment, it actually adapts the policies, procedures and pol- they faced in scrum implementation in development. These
icies within the team to take best outputs of the life cycle such two organizations are ―Bentley Systems Pakistan‖ and ―Digi-
as quality and reliability. SDLC itself does not guarantee the tal Prodigy Limited‖ Each has more than 3 years and 9 months
success of the project but helps the team in success. There are experience in implementation of scrum framework. The dif-
several SDLC models e.g. Waterfall, Spiral and Agile but this ference in experience will help in the analysis of the difference
research is focused on Scrum which is the further classification between the issues faced by experienced Scrum teams versus
of Agile Development. the issues faced by inexperienced Scrum teams.

1.1 Drawbacks
2 MATERIALS AND METHODS
Substantial research has been done to perfect the different
aspects of Scrum framework such as [26], [6], [12]. Still there 2.1 Research Methodology
are some open research areas which need the attention of the In this research work, Mixed Approach is used to identify
community e.g. [22], [4]. In Scrum Framework, there are some and highlight issues and challenges of Scrum implementation
limitations with respect to nature of the work and are hig- with the help of existing work and from experiences of two
hlighted in [13] such as training, management, involvement, companies by survey. A survey was designed for this study.
access to external resources, corporate or organization size [6], The survey was reviewed by peers for readability and by other
[12], distributed area, sub contraction, developing large and Scrum experts for content validity. It was distributed to differ-
complex systems, but still there are open areas where no sig- ent Scrum experts for more than two weeks. Data was col-
nificant research work has been done. According to Boehm lected from two sources i.e. survey and face to face interview
and Turner (2005) [5], areas of limitations in agile processes with 20 employees from two companies.
are: distributed environments, building reusable artifacts, These employees include project managers, scrum masters,
large teams, and developing safety-critical software. Similarly development team and quality assurance team.
Sutherland et al. (2009b) [26] also considered distributed
2.2 Survey
Scrum as a challenging area where more research work is re-
quired. Survey of this research paper takes an in-depth view of the
challenges involved in implementation of scrum into two or-
1.2 Objectives ganizations.
Questionnaire for this survey was derived from the views
The objective of this research paper is to minimize the is-
of scrum experts. Survey included 23 questions and these
sues and challenges of Scrum in Software Development and
questions are related to scrum formal training, implementa-
IJSER © 2012
http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 3, Issue 8, August-2012 2
ISSN 2229-5518

tion of basic rules of scrum, managerial and team response, ture in Scrum implementation have relatively fewer issues
meetings, deployment, testing and release process. than immature teams. But the difference is not very big. This is
A total of thirty five individuals were asked to participate because Digital Prodigy Limited (DPL) is trying to follow
in the survey. From these thirty five individuals, responses scrum rules properly.
from twenty individuals were received making return rate of It has also been observed that issues identified from DPL
57.14%. The survey was composed of a total of twenty three survey results are 14% more than the issues identified from
questions. The individuals were not offered any incentive to Bentley Systems Inc. survey data, which is 43%.
answer these questions. These questions covered the areas of VI. Sprint Duration
Development, Testing and Release Process. The participants Sprint duration plays an important role in effectiveness of
belong to the teams that have adopted Scrum Framework. All Scrum. Scrum Master has to wisely select sprint duration to
roles related to software development such as project manag- get maximum benefits of this framework.
ers, scrum masters, development team and quality assurance Results clearly highlight the number of issues at different
team responded to the survey. The participant’s experience sprint duration. 6% of survey participants are facing 2% and
ranged from 1 year to 15 years. All of the interviews were au- 8% issues for two and five weeks of sprint duration respective-
dio taped and logged. The received data was stored in a data- ly. Minimum sprint duration which is identified from survey
base and later data mining technique was applied on it to find is one week and their identified issues are more than any other
the hidden relation between different concepts. In the process sprint duration issues. 50% of team members are following 1
of data analysis grounded theories were used to derive con- week sprint and their identified issues are 55% of total. The
structs from collected raw data. second most issues are identified from the team members who
have three weeks of sprint duration. They are 28% of the sur-
vey participants and are facing 29% of total issues which are
3 RESULTS AND DISCUSSIONS identified from the survey. It has also been identified that 11%
3.1 Identified Issues
of team members are following six weeks of sprint duration
Several issues are identified from survey results and from but issues identified from their responses is only 6%.
interview sessions. These issues are directly affecting Scrum VII. Lack of Scrum Training
implementation and are related to management, development Results also revealed that 50% team members lack formal
and release process. Here is a detailed description of each. training of scrum and are unaware of the scrum process. The
knowledge that they have gained is either because of the other
I. Quality Items Pileup team fellows or from their scum masters.
Due to agility of work in Scrum, teams have an obligation VIII. Release Process
to present something in a short time regardless of their scope Another major issue in scrum is the Release process/ dep-
in ―Sprint Planning‖. Because of this sometimes teams ignore loyment process. Agility in work is introduced by Scrum;
quality of software and create a pileup of quality related items. sprint deployment is the major concern for every team.
This may relate to performance improvement steps as well. IX. Backlog Management
This is the biggest issue with Scrum Methodology and for that Scrum provides insufficient guidance with respect to the
matter with any Agile Methodology. structure of the backlog. The scrum management tools that are
II. Module Integration Issues available in the market are either too complex (which in turn
As the products are released frequently during a sprint are more expensive) or too simple and are therefore not useful
module integration testing cannot be performed properly all for the team. Due to this the teams that have recently adapted
the times, as it requires a lot of time for testing and quality Scrum framework for their projects, avoid large investments at
assurance. Inadequate time allocation for QA of large or com- initial level of framework implementation.
plex systems can easily generate critical issues. X. No Technical Practices
III. Code Quality Although scrum shows a good project management capa-
Teams have a short term deadlines due the agility of the bility, there aren’t any technical practices that can be called
teams. To cope up with a lagging deadline, the developer ―best‖.
needs to put extra hours. This will create Code Quality related XI. Multiple Teams
issues. However, one cannot write bug free code all the time The survey showed that working with multiple teams is a
especially when he is working under pressure. tough job, unlike the traditional SDLC methods because scrum
IV. Disruption in Team Work doesn’t have strong advices on it. Although scrum talks about
From survey results, it is noted that Product Owner and ―Scrum of Scrums‖ but this technique doesn’t work well when
Scrum Master interfere with team members by asking statuses the teams are specially distributed. In fact agile is virtually
like they used to ask in traditional SDLC. It has also been ob- silent on this issue. When working in distributed environment
served that client adds the requirements during the sprint. It Scrum Masters require lot of work and coordination among all
has been observed that from survey results that interruptions the teams. This is not a failing of scrum but a lack of strong
by managers have been reported by 44% of team members, process when working with multiple teams.
whereas the number of people/employees satisfied by their XII. Metrics
manager’s roles is about 56%. Burn down charts and velocity is used for measure metrics
V. Mature vs. Immature Scrum in Scrum. These metrics can only be useful in the presence of a
This is also proved from results that teams which are ma-
IJSER © 2012
http://www.ijser.org
 International Journal of Scientific & Engineering Research, Volume 3, Issue 8, August-2012 3
ISSN 2229-5518

Scrum Master or Product Manager who would help in analyz- Re-scoping for tasks which are missing deadlines can solve
ing the burn down chart for its peaks. However, these metrics code quality issue very easily. The tasks can be divided into
are often ignored and not analyzed. Burn Down chart is good subtasks after proper re-scoping. Progress on the task can be
to show overall status of the sprint in terms of the number of shown by doing some sub sections in current sprint moving
hours that have been spent and the number of hours that are the rest of the sections to the next sprint. However those sec-
left. However, these burn down charts are not enough to pro- tions should be finished in the next sprint as they would be
vide in depth detail of that particular day. considered as a lag.
As far as velocity is concerned, it is defined as number of IV. Disruption in Team Work
tasks/story points completed. There are cases when in one Disruption in Team work doesn’t allow a team to be self-
sprint velocity is high and on the very next sprint it is low. organized and self-empowered, which is a strong benefit of
This is because it can never be used to identify low or high scrum. If team is not mature in their work and responsibilities,
velocity in a project. then it should be made mature first. Proper training and scru-
XIII. Risk Management tinizing process plays a major role in the maturation of the
There is no plan or strategy in scrum to handle risks. It is an team. Managers and Scrum Masters should avoid interfering
important factor for any project and hence work is required in in team work. Let the team do their job themselves.
this area of Scrum. V. Sprint Duration
XIV. Documentation It is very important to set sprint duration to the rate at which
Agile believes in no documentation so does Scrum; howev- requirements change and the rate at which the team can deliv-
er this phenomenon is still not successful in real environment er a functionality. Adding more work than the team can do
where things come in and out through email or any other makes it difficult to meet commitments and to measure
source, which is quite difficult to track. From the survey, it has progress and velocity of the team, and new "high-priority"
been found that requirements get changed through emails work can disrupt flow.
without relating them properly in Product Backlog which VI. Lack of Scrum Training
creates issues for team in traceability. Properly trained teams are far better than the ones having a
XV. Too Idealistic lack of formal training. They are better in terms of self-
Scrum assumes teams are of same self-managing. As per organization and self-empowerment. Formal Scrum training
rules of Scrum, Scrum master job is to remove all impediments helps in acquainting the teams about the working of scrum
for his team and product owner job is to provide the require- and in increasing awareness about team rights. Through prop-
ments in stories which will help to build or complete the er training, teams will be able to better understand the work-
project incrementally. However in practical this scenario is not ing of Scrum and they will eventually know how to execute
valid. It is not possible to make your team self-organized and and extract the benefits of Scrum for their team and product.
self-empowered all the time or all team members are self- VII. Release Process
organized and self-empowered. For instance, a new team Teams have more deployments because the work that they
member who does not know about scrum cannot have those have done is now divided into sprints. Extra care must be tak-
attributes unless he is properly trained and has gained some en in case of deployment. A process that can be of great help
experience in Scrum process. in this case is the Release Management process. Release Man-
XVI. Communication/Scrum Ceremonies agement also requires policies which can best suit to Scrum.
Due to increase in communication, some team members This issue can be resolved by better policies and their en-
feel disturbance in their work. They are not able to concentrate forcement in teams.
properly. Effective communication also adds frequent meet- VIII. Multiple Teams
ings. Some team members think attending all meetings are The major challenge in execution of any agile process includ-
unnecessary for them, hence they do not show their interests ing Scrum across distributed environment is the communica-
in those meetings which are not directly related to them. tion issue. Resolution for this is provided in this research
where frequent informal meetings can help teams to involve
3.2 Solution for the Identified Issues and share ideas other than the work. Introducing extra-
In this section, solutions are proposed to some of the issues curricular activities such as debates (only global matters or
which are identified in the above section. interesting matters), conferences and re-allocation can drasti-
I. Quality Items Pileup cally help in resolving this issue. On the other hand, manager
Special attention is required to quality related items to in- needs to realize how critical this issue is and how it can affect
crease reliability and in achieving goals such as maintainabili- the overall performance of the teams. Managers also need to
ty and performance. Quality related tasks always helps in define policies which can help in reducing social and cultural
achieving factors such as reliability, maintainability and scala- issues. Racism must be prohibited in any environment and
bility hence it should not be ignored at any time. effective policy defining can surely help in this matter.
II. Module Integration IX. Documentation
Teams should spend more time in testing of final internal re- Changes should be tractable from Scrum Backlog. If any
lease. It is highly suggested to dedicate a full sprint for change comes, it should be added in task story properly. It
projects which are having tightly coupled modules for testing. also helps in referencing and makes the requirement tractable.
III. Code Quality One more advantage of documentation is to help out the other
IJSER © 2012
http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 3, Issue 8, August-2012 4
ISSN 2229-5518

developers to understand the business values and goals. [10] Hneif, M. and S. H. Ow. 2009. Review of Agile Methodologies in Software
X. Communication/Scrum Ceremonies Development. Int. Journal of Research and Reviews in Applied Sciences. (Is-
Impediment resolution meetings should be conducted with lamabad, Pakistan) 1(1): 1-10.
only those team members who are directly related to the im- [11] Jyothi, V.E. and K.N. Rao. 2011. Effective Implementation of Agile Practices -
pediments. After the resolution, other team members should Ingenious and Organized Theoretical Framework. Int. Journal of Advanced
also be informed about the conclusions of the discussions or Computer Science and Applications (IJACSA). (New York, USA) pp. 41-48.
meetings. Resolutions should also be added in task stories. [12] Kontio, J., M. Höglund, J. Rydén and P. Abrahamsson. 2004. Managing
Maintaining stories/task descriptions help other team mem- Commitments and Risks: Challenges in Distributed Agile Development ICSE.
bers to track requirement in later sprints. 26th Int. Conf. on Software Engineering. (Helsinki Univ. of Tech., Finland) pp.
732-733.
[13] Livermore, J. 2008. Factors that Significantly Impact the Implementation of an
4 SUMMARY
Agile Software Development Methodology. Journal of Software. (Oulu, Fin-
This paper determines that a large number of factors have an land) 3(4): 31-36.
impact on Scrum implementation which are also valid for oth- [14] Mahanti, A. 2006. Challenges in Enterprise Adoption of Agile Methods: A
er agile development methodologies. Issues identified from Survey. Journal of Computing and Information Technology, JCIT. (Zagreb,
this survey are Quality Items Pileup, Module Integration Is- Croatia) 14(3): 197-206.
sues, Code Quality, Disruption in Team Work, Mature vs. [15] Marchenko, A. and P. Abrahamsson. 2008. Scrum in a Multi project Envi-
Immature Scrum, Sprint Duration, Lack of Scrum Training, ronment: An Ethnographically-Inspired Case Study on the Adoption Chal-
Release Process, Backlog Management, No technical practices, lenges. Agile Int. Conf. (Toronto, Canada) pp. 15-26.
Multiple teams, Metrics, Risk Management, Documentation, [16] Newkirk, J. 2002. Introduction to Agile Processes and Extreme Programming.
Too Idealistic Scrum and Communication/Scrum Ceremonies. Int. Conf. on Software Engineering (ICSE). (Orlando, FL., USA) pp. 695-696.
Whereas solution were provided for the following issues [17] Rising, L. and N. Janoff. 2000. The Scrum Software Development Process for
which are Quality Items Pileup, Module Integration Issues, Small Teams. IEEE Computer Society Press. (Los Alamitos, CA., USA) 17(4):
Code Quality, Disruption in Team Work, Sprint Duration, 26-32.
Lack of Scrum Training, Release Process, Multiple Teams, [18] Rychly, M. and P. Ticha. 2008. A Tool for Supporting Feature-Driven Devel-
Documentation and Communication/Scrum Ceremonies. opment. Balancing Agility and Formalism in Software Engineering. Springer-
Verlag. (Berlin, Heidelberg, Germany) pp. 196-207.
[19] Salo, O. and P. Abrahamsson. 2008. Agile Methods in European Embedded
REFERENCES Software Development Organizations: A Survey on the Actual Use and Use-
fulness of Extreme Programming and Scrum. Institution of Engineering and
[1] Abrahamsson, P., O. Salo, J. Ronkainen and J. Warsta. 2002. Agile Software
Technology. (Stevenage, UK) 2(1): 58–64.
Development Methods-Review and Analysis. Espoo. VTT Publications 478.
[20] Scotland, K. and A. Boutin. 2008. Integrating Scrum with the Process Frame-
Dept. Information Processing Science, (Univ. Oulu, Finland).
work at Yahoo Europe. Agile Int. Conf. (Washington, DC., USA) pp. 191-195.
[2] Bates, C. and S. Yates. 2008. Scrum Down: A Software Engineer and a Sociol-
[21] Smits, H. and G. Pshigoda. 2007. Implementing Scrum in a Distributed Soft-
ogist Explore the Implementation of an Agile Method. Int. Conf. on Software
ware Development Organization. Agile Int. Conf. (Washington, DC., USA)
Engineering (ICSE). (New York, USA) pp. 13-16.
pp. 371-375.
[3] Begel, A. and N. Nagappan. 2007. Usage and Perceptions of Agile Software
[22] Sutherland, J. 2005. Future of Scrum: Parallel Pipelining of Sprints in Complex
Development in an Industrial Context: An Exploratory Study. 1st Int. Symp.
Projects. ADC. Proc. of Agile Development Conf. (Denver, CO., USA) pp. 90-
on Empirical Software Engineering and Measurement (ESEM). (Madrid,
102.
Spain) pp. 255-264.
[23] Sutherland, J., C. R. Jakobsen and K. Johnson. 2008a. Scrum and CMMI Level
[4] Berczuk, S. 2007. Back to Basics: The Role of Agile Principles in Success with a
5: The Magic Potion for Code Warriors. Hawaii Int. Conf. on System Sciences
Distributed Scrum Team. IEEE Computer Society Press. (Washington, DC.,
(HICSS). (Washington, DC., USA) pp. 191-195.
USA) pp. 382-388.
[24] Sutherland, J., G. Schoonheim, E. Rustenburg and M. Rijk. 2008b. Fully Distri-
[5] Boehm, B. and R. Turner. 2005. Management Challenges to Implementing
buted Scrum: The Secret Source for Hyper productive Offshore Development
Agile Processes in Traditional Development Organizations. IEEE Computer
Teams. IEEE Computer Society Press. (Washington, DC., USA) pp. 339-344.
Society Press. (Los Angeles, CA, USA) 22(5): 30-39.
[25] Sutherland, J., G. Schoonheim and M. Rijk. 2009a. Fully Distributed Scrum:
[6] Cho, J. 2008. Issues and Challenges of Agile Software Development with
Replicating Local Productivity and Quality with Offshore Teams. Hawaii Int.
Scrum. IACIS. Issues in Information Systems. (Stillwater, OK., USA) 9(2): 188-
Conf. on System Sciences (HICSS). (Big Island, HI, USA) pp. 1-8.
195.
[26] Sutherland, J., G. Schoonheim, N. Kumar, V. Pandey and S. Vishal. 2009b.
[7] Cordeiro, L., C. Becker and R. Barreto. 2007. Applying Scrum and Organiza-
Fully Distributed Scrum: Linear Scalability of Production between San Fran-
tional Patterns to Multi-site Software Development. 6th Latin American Conf.
cisco and India. Agile Conf. (Washington, DC., USA) pp. 277-282.
on Pattern Languages of Programming. (Porto de Galinhas, Recife, Brazil) pp.
[27] Turk, D., R. France and B. Rumpe. 2002. Limitations of Agile Software
46-67.
Processes. 3rd Int. Conf. on eXtreme Programming and Agile Processes in
[8] Hansen, M.T. and H. Baggesen. 2009. From CMMI and Isolation to Scrum,
Software Engineering. (Sardinia, Italy) pp. 43-46.
Agile, Lean and Collaboration. Agile Conf. (Washington, DC., USA) pp. 283-
288.
[9] Hazzan, O. and Y. Dubinsky. 2008. Introduction to Agile Software Develop-
ment. Balancing Agile Software Engineering. Springer-Verlag. (London, UK)
pp. 1-24.
IJSER © 2012
http://www.ijser.org

View publication stats