Materials Today: Proceedings xxx (xxxx) xxx

Contents lists available at ScienceDirect

Materials Today: Proceedings

journal homepage: www.elsevier.com/locate/matpr

Agile adoption issues in large scale organizations: A review

Dasari Venkatesh, Manik Rakhra ⇑
Department of Computer Science and Engineering, Lovely Professional University Phagwara, Punjab 14411, India

a r t i c l e i n f o a b s t r a c t

Article history: Recently agile methods are turning out to be very crucial approach for many different software develop-
Received 1 November 2020 ment companies to increase the quality factor in their products, however the method was originally
Accepted 10 November 2020 designed to help the self-organizing teams that are small to medium size but not for large-scale compa-
Available online xxxx
nies. So, using agile in large scale sectors could cause a new set of problems such work synchronization by
the team members and collaboration with other organizations as well. In this paper, we are going to show
Keywords: how agile methodology is implemented in large scale organizations and we are also going to take a look
Software requirement specification (SRS)
at all the problems that were reported by those large scale companies when they implemented agile
Agile software methodology
KLOC
methodology in their products. We are going to show the problems faced and also the success achieved
by those large-scale companies through literature review. From all the issues reported by those compa-
nies, it is clear that the major challenges in adopting agile methodology are choosing the method, training
and certification, team member spirit and attitude.
Ó 2020 Elsevier Ltd. All rights reserved.
Selection and peer-review under responsibility of the scientific committee of the Emerging Trends in
Materials Science, Technology and Engineering.

1. Introduction vey conducted by version one, 62% of the companies that took the
survey had well over 100 team members on their projects. It also
Back in 2005, agile methodology was first introduced by Boehm showed that 43% of the companies that took the survey have more
and Turner [1]. Originally, it was only designed to be implemented than 50% of their teams using agile methods. Obviously, the survey
in small scale self-organizing teams. Since it was a really great suc- was only conducted for specific number of companies. But this sur-
cess, developers thought it would be a great idea to implement vey showed that majority of the companies now a days are imple-
them in large scale companies to improve the quality of the prod- menting agile methodologies in their large-scale projects. Even
ucts despite the fact that implementing them in large-scale pro- though the literature reviews contain many reports shared by the
jects would be really hard. Unlike smaller projects, coordination experts and a lot of case studies on agile software development,
and activity synchronization is crucial in large-scale projects [1]. none of them mentioned a systematic approach and method for
One of the major issues with this method was coordination among the implementation in large-scale projects. In an XP conference
team members. Since large-scale projects involve additional that was held in 2010, Freudenberg and Sharp asked those large-
aspects such as marketing and sales, team and project manage- scale companies to make a backlog for all these suggestions and
ment, and human resources. Not to mention, compared to small topics that they need to do research upon to solve this research
scale projects, larger companies usually have less customer inter- gap. Majority of the companies voted that agile implementation
action with the developing team. This could end up making the in large-scale projects is the biggest problem they were dealing
customer unsatisfied at the time of delivery. Although we have with. They are voted agile distribution in large-scale projects
many problems like this, there is a trend which the companies among the top 10 problems in the conference. Even after a few
are following to make this agile methodology work in large-scale years, in the XP conferences that were help in 2013 and 2014,
projects. A survey was conducted back in the year 2007, asking implementation of agile methodology was the topic discussed
the companies about the problems they have been facing by the mostly. It was clear that even in a span of 6–7 years, the research
implementation of agile methods in their products. As per the sur- gap was not studied all too well (Table 1). This concludes that even
though there is tremendous development in agile software devel-
⇑ Corresponding author. opment and implementation, the basic systematic approach is still
E-mail address: rakhramanik786@gmail.com (M. Rakhra). not improved and studied yet. In this paper, we are going to be

https://doi.org/10.1016/j.matpr.2020.11.308
2214-7853/Ó 2020 Elsevier Ltd. All rights reserved.
Selection and peer-review under responsibility of the scientific committee of the Emerging Trends in Materials Science, Technology and Engineering.

Please cite this article as: D. Venkatesh and M. Rakhra, Agile adoption issues in large scale organizations: A review, Materials Today: Proceedings, https://
doi.org/10.1016/j.matpr.2020.11.308
D. Venkatesh and M. Rakhra Materials Today: Proceedings xxx (xxxx) xxx

Table 1 into traditional software development. In addition to that, the

Searches found on different databases on agile adoption. teams should also take care of the aspects such as HR, marketing
Database URL No.of searches and sales, which are not so agile friendly [7]. If we take HR as an
Scopus http://www.scopus.com 1542 example, it demands that the teams should be following a strict
IEEE http://ieeexplore.ieee.org 720 schedule the entire time whereas agile methodology wants the
Web of knowledge http://apps.ewbofknowledge.com 146 team members to be more flexible towards the schedule and be
ACM http://dl.acm.org 752 ready for any kinds of changes that might occur (See Table 1).
Agile methodology also forces the organization to get rid their
traditional life cycle models and start going towards incremental
bridging this gap in research and also represent some steps models. This would affect the management and customers since
through the literature review. the customers would not be able to understand what to expect
in their next installation of their ongoing long term projects.
2. Methodology
2.3. Large-scale agile teams and organizations
2.1. Agile software development
According to the research done by Dingsoyr in the year 2014
According to the agile manifesto, Agile methodology is an incre- [2,3], the term size was usually defined by number of persons in
mental and procedural software development method [8]. The a team, cost of the project estimation, time taken for completion
main goal of agile software development was to replace the tradi- and number of KLOC in the coding section. A project could be ter-
tional software development which takes the SRS documentation med as large-scale if it has around 40 people and 7 teams [9], cost
from both the team and customers, then go with design and imple- estimation of around 10 million dollars and the code size of almost
mentation phases. This kind of traditional approach only focuses 5000KLOC according to the study conducted by Paasivaara in the
on documented planning and approach, which makes it very reluc- year 2008 [4]. All of these studies on agile software development
tant to any kind of changes. Not only this takes a lot of time and mainly focused on number of persons in a single team.
effort, but also has poor customer interaction. When agile was first So, to finalize these studies, an organization or a project could
introduced, it was criticized by a lot of people about how it could be termed as large scale if it has more than 50 people and around
lead to both success and a huge failure if things don’t go right. 6 teams working on it [9,3]. These people may or may not be a part
However, when it was first implemented in small self-organizing of the development team but they must belong to the project man-
teams, it showed really great potential. According to Dyba and aged by the organization. We can consider scrum masters as a
Dingsoyer [2], agile was not supposed to be implemented in large prime example for this. The scrum master is responsible for orga-
scale projects [2,3]. One of the first solutions ever proposed for the nizing the team and the project but does not necessarily belong to
implementation of agile methods in large-scale projects was that the development team. Hence, he/she will be included in the team
the organization should maintain a strict plan and should be able size. Although some surveys consider number of teams rather than
to maintain their balance with agile methods whilst taking care the team size, the difference it makes is not all that significant.
of other aspects such as human resources, marketing and sales. Even if there are multiple teams working on their individual seg-
The two important agile methods namely scrum and extreme ment of the project, they all have to combine the modules of the
programmings are still being used in the current decade. Scrum final product for integration. This states that the teams having
is basically organizing the entire team and the project by tracking more than 50 people and around 6 teams [8,2] or multiple number
the progress of the project, having daily scrum meetings of 15– of teams summing up to have 50 people yields the same result and
20 min in which the next steps to be taken are discussed. Scrum thus be considered as large-scale organization.
master is responsible to take care of the scrum teams which means,
the teams cannot be self-organizing like they do in small scale pro-
3. Literature review
jects. On the other hand, XP also known as extreme programming
is basically doing the incremental process in the most efficient way
Muhammad Sohail Khan et al. (2019) explained how the adop-
possible by using good coding skills which helps the team to
tion of agile methods takes place in large-scale projects. This paper
reduce bugs and errors whilst consuming less resources.
also explained how to find search for literature study to under-
stand the concept of agile. An efficient procedure to maintain the
2.2. Implementation of agile in large-scale projects quality the software and consume as less time as possible was also
mentioned in this paper.
Usually, the difficulty level of introducing agile methodology Christoph Segler et al. (2019) explains how frequent the
scales up with the size of the organization those choses to adopting changes are actually done in continuous architecture by many soft-
it. It means, the larger the size of the organization is, the harder it is ware developing companies and automotive industries. All the
to adopt agile methodology in their large-scale products [6]. The add-ons and implementation of new components is heavily relying
main reason for this kind of situation to arise is the fact that larger on non-functional attributes of that particular pre-installed soft-
organizations are a lot more reluctant to any kind of changes com- ware. In this case, component-based software engineering could
pared to smaller self-organizing teams. help tackle the situation fairly easily by making the new compo-
One of the biggest differences between large-scale and small- nents compatible with the old ones.
scale teams when they try to adopt agile methodology is that the Aldo Dagnino et al. (2014) clearly explains the analytical statis-
large-scale organizations have a lot of interdependencies among tics of the production line of all the components required to be
multiple teams and the projects they deal [5]. This means, when embedded into the system with highest efficiency and in very
one of the factors of those aspects change, the entire functioning cost-effective way. He states that the agile methods would help
culture of the organization along with the teams should also the software developing team to find the capability of a system
change. This makes the team members to be more dependent upon and its limits which would help them to find the perfect compati-
the formal documentation which reduces the agility. This defeats ble components to be implemented in the new system without
the purpose of agile software methodologies entirely and turns it having any errors.
2
D. Venkatesh and M. Rakhra Materials Today: Proceedings xxx (xxxx) xxx

Jianguo Chen et al. (2009) explains the significant job of pro- rapidly which would even start resistance. Proper management of
gramming measurements engaged with component- based pro- the schedule is usually the only way to handle the workload.
gramming building. Programming measurements is for the most
part utilized for the estimation of nature of the product and its 4.2. Difficulties in implementation of agile
unpredictability. It is likewise clarified how measurements vary
in customary programming advancement from component-based It is a well-known fact that agile could be really hard to adapt
turn of events. He additionally disclosed how to draw a control to, even for a seasoned trained team of engineers. This is mostly
stream outline utilizing cyclomatic intricacy which causes us to due to the fact that agile methods are a lot different compared to
quantify what number of linear autonomous ways are these in the traditional software development life-cycle techniques. The
the source code. usual problems that arise are:
Jen Knodel et al. (2017) explained how the evaluation of soft-
ware architecture could be done using various methods such as  Not understanding the concept of agile properly. Majority of
architecture compliance check, documentation quality check and the issues are caused due to poor understanding of the concept
drivers used in architecture. He mentioned that using the above of agile adaptation. Many development teams view it as a short-
stated methods could greatly help the software development team cut to ignore documentation, directly going with development
to examine and evaluate the final product or software before deliv- by ignoring design phase and thinking that agile method pro-
ering it to the customer, thus improving the quality of the motes generalist. If the team proceeds to use agile in this way,
software. it would only end up leading to poor-quality software that ruins
Laxmi Ahuga et al. (2019) explained the how much of an impact the market value of the organization.
does the implementation of agile methodology effect the quality of  Need of filling the research gap: - Although agile methodology
a software. It is mentioned how important maintaining the quality is improving every year, there are many questions that have not
of a software is. This could be compromised when the development been solved for the past decade. The problems faced by large-
process is rushed too fast which would ultimately result in cus- scale organizations when they implement agile methods are
tomer being unsatisfied. yet to be solves. As Beavers stated, there is no guide or a manual
Tien N. Nguyen et al. (2008) proposed an exploration that clar- that helps us to guide through the whole process of adoption of
ifies the various types of translations of a product update when the agile methodology. In order to solve this issue, there is a specific
framework is made by component-based programming designing. filtering technique to help us to find the appropriate work
The two sorts of translations are standard understanding and quickly.
unique translation. He likewise clarified how states and activities  Lack of customization of agile: -This is directly related to
work semantically when the framework gets an update or a hot misunderstanding of the concept of agile. When organizations
fix. Rules of activity during the framework update have addition- try to customize the product to their liking, since there is no
ally been clarified in this paper. proper documentation in agile methodology, sometimes the
project might end up being non-feasible. In this kind of scenar-
4. Discussion io, they end up skipping the practices which leads to poor cus-
tomization of the final product which causes the customer to be
When an organization implements agile methodology in their unsatisfied thus destroying the market value. This leads to cus-
projects, the problems that arise could be in multiple areas and tomizing the final modules of the product in a poor way.
is not confined to the developing team’s issue. The following states
the various areas that are affected. 4.3. Problems with inter-team coordination

 Teams interface coordination: - This is one of the biggest

4.1. Lack of investment
issues when a large-scale team adopts agile methods. Since
each team works on their individual tasks, some team will have
4.1.1. Poor training
to collaborate with other teams to finish their task. However,
A large-scale project that is using XP (extreme programming)
each team has its own schedule to follow, skill level.so the rate
method of agile, it demands that the members of the development
of completion of their daily tasks may not match with the other
team are well trained and are ready to take on hard challenges that
team. This could lead to improper functioning of the teams
come with the adoption of agile methods. If the team members are
causing delay in the completion of the project. This states that
not trained enough for this, it turns into a glaring issue that need to
agile adoption did not solve the problem of dependencies and
be solves as soon as they could.
dependency makes the project to go slower.
 Consistency in technical skill: - As stated above, not all team
4.1.2. Poor coaching working on the same large-scale project has same amount of
The scrum master is usually responsible for managing the team technical skills. Some teams may be slower than the others.
and guides them to accomplish their designated tasks. However, Due to this inconsistency, the problem arises during the final
many large-scale organizations are failing to arrange an experi- integration phase. One module might be coded a lot better than
enced coach to guide their teams. Silva and Doss stated that many the other module. In order to make both of them compatible
organizations that fail to find a good coach, people who are less with each other, either we could increase the standard of the
experienced are appointed to fill that job. This could create many lower-tier module or make a moderate tier module by sacri-
problems when it comes to guiding the team during crucial ficing some efficiency.
situations.
5. Result
4.1.3. Heavy workload
Although many large-scale organisations are willing to adapt to In Fig. 1, we have explained how to search the database to
the changes caused by agile methodology, it may not always be gather research work on agile adoption in large-scale projects in
feasible with the work environment of the development teams. an efficient way. Out of a total of 1832 papers, using the abstract
People would go under lot of stress if the changes are taking place filtering, we have removed the duplicated and finalized it into
3
D. Venkatesh and M. Rakhra Materials Today: Proceedings xxx (xxxx) xxx

Fig. 1. Process of database search.

References

[1] B. Boehm, R. Turner, Management challenges to implementing agile processes

in traditional development organizations, IEEE Softw. 22 (5) (2005) 30–39,
https://doi.org/10.1109/MS.2005.129.
[2] T. Dyba, T. Dingsoyr, What do we know about agile software development?,
IEEE Softw 26 (5) (2009) 6–9, https://doi.org/10.1109/MS.2009.145.
[3] T. Dingsøyr and N. B. Moe, ‘‘Towards principles of large-scale agile
development,’’ in Proc. Int. Workshops. Rome, Italy: Springer, May 2014, pp.
1–8.
[4] M. Paasivaara, S. Durasiewicz, C. Lassenius, Using scrum in a globally
distributed project: a case study, Softw. Process Improve. Pract. 13 (6) (2008)
527–544, https://doi.org/10.1002/spip.402.
[5] A. Elshamy and A. Elssamadisy, ‘‘Divide after you conquer: An agile software
development practice for large projects,’’ in Proc. 7th Int. Conf. Oulu, Finland:
Springer, Jun. 2006, pp. 164–168.
[6] S. Chandra Misra, V. Kumar, U. Kumar, Identifying some critical changes
Fig. 2. literature surveys done on agile.
required in adopting agile practices in traditional software development
projects, Int J Qual Reliabil. Manage. 27 (4) (2010) 451–474, https://doi.org/
10.1108/02656711011035147.
160 papers. We have then used full text filtering and narrowed it [7] P. Rodríguez, J. Markkula, M. Oivo, and K. Turula, ‘‘Survey on agile and lean
down to 49 papers which are completely related to the research usage in finnish software industry,’’ in Proc. ACM-IEEE Int. Symp. Empirical
Softw. Eng. Meas., Sep. 2012, pp. 139–148.
work done on the area of agile adoption in large-scale project. In [8] M. Fowler and J. Highsmith, ‘‘The agile manifesto,’’ Softw. Develop., vol. 9, no. 8,
Fig. 2, we have explained the rate of studies being done in each pp. 28–35, 2001.
area of agile adoption. This includes literature surveys with their [9] T. Dybå, T. Dingsøyr, Empirical studies of agile software development: a
systematic review, Inform. Softw. Technol. 50 (9-10) (2008) 833–859, https://
count on y-axis. This shows how significantly the work is being doi.org/10.1016/j.infsof.2008.01.006.
done to improve the implementation of agile adoption on large-
scale projects every year. We can also see thesis work is not being
done too often. Further Reading

[1] H. Berger P. Beynon-Davies The utility of rapid application development in

large-scale, complex projects 19 6 2009 549 570 10.1111/j.1365-2575.2009.
6. Conclusion 00329.x.
[2] K. Petersen, C. Wohlin, The effect of moving from a plan-driven to an
In this paper, we showed the status and advantages of agile incremental software development approach with agile practices: an
industrial case study, Empir Software Eng. 15 (6) (2010) 654–693, https://doi.
software development in software engineering. We have also org/10.1007/s10664-010-9136-6.
talked about various research works done on this topic for the past [3] E. Bjarnason, K. Wnuk, and B. Regnell, ‘‘A case study on benefits and side-effects
decade. We talked about the difficulties faced by large-scale orga- of agile practices in large-scale requirements engineering,’’ in Proc. 1st
Workshop Agile Requirements Eng. New York, NY, USA: ACM, 2011, p. 3.
nizations when they adopt agile methodologies. In the discussion [4] L. Williams, ‘‘Agile software development methodologies and practices,’’ in
area, various categories of challenges faced by both the organiza- Advances in Computers, vol. 80. Amsterdam, The Netherlands: Elsevier, 2010,
tions and its teams were explained. When it comes to future pp. 1–44.
[5] S. Nerur, RadhaKanta Mahapatra, G. Mangalaraj, Challenges of migrating to agile
research topics, it suggested to study on the problems mentions methodologies, Commun. ACM 48 (5) (2005) 72–78, https://doi.org/10.1145/
above and finds a proper solution on how to resolve these issues 1060710.1060712.
without costing the large-scale organization too many resources. [6] M. Fowler, Put your process on a diet-as a reaction to cumbersome approaches
to development, new methodologies have appeared. These methods attempt a
compromise between no process and too much process, Softw. Develop. 8 (12)
(2000) 32–39.
Declaration of Competing Interest [7] S. Jalali, C. Wohlin, Global software engineering and agile practices: a systematic
review, J. Softw. Evol. Process 24 (6) (2012) 643–659.
The authors declare that they have no known competing finan- [8] M. Kaisti, V. Rantala, T. Mujunen, S. Hyrynsalmi, K. Könnölä, T. Mäkilä, T.
Lehtonen, Agile methods for embedded systems development - a literature
cial interests or personal relationships that could have appeared review and a mapping study, J Embedded Syst 2013 (1) (2013), https://doi.org/
to influence the work reported in this paper. 10.1186/1687-3963-2013-15.

4
D. Venkatesh and M. Rakhra Materials Today: Proceedings xxx (xxxx) xxx

[9] S. Ali, L. Hongqi, and M. F. Abrar, ‘‘Systematic literature review of critical [11] D. Budgen and P. Brereton, ‘‘Performing systematic literature reviews in
barriers to software outsourcing partnership,’’ in Proc. 5th Int. Multi-Topic ICT software engineering,’’ in Proc. 28th Int. Conf. Softw. Eng. New York, NY, USA:
Conf. (IMTIC), Apr. 2018, pp. 1–8. ACM, 2006, pp. 1051–1052.
[10] M. Lacity L. Willcocks D. Feeny Commercializing the Back Office at Lloyds of [12] K. Power, L. Morgan, and K. Conboy, ‘‘Enabling open innovation through agile
London: European Management Journal 22 2 2004 127 140 10.1016/j. development,’’ in Proc. 11th Int. Conf. Agile Softw. Develop., 2010, pp. 1–3.
emj.2004.01.016