Professional Documents
Culture Documents
Agile and Software Project Management Antipatterns: Clarifying The Partnership
Agile and Software Project Management Antipatterns: Clarifying The Partnership
fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/MS.2020.3001030, IEEE Software
Ana M. Moreno
Universidad Politécnica de Madrid, Spain
Lawrence Peters
Universidad Politécnica de Madrid, Spain
Iowa State University, U.S.A
Abstract— Agile methods have been recognized as providing tangible benefits, some of them directly
related to software project management (SPM). In this paper we address in detail the relationship between
Agile and SPM, discussing the extent to which the use of agile practices helps to reduce or avoid the
occurrence of different SPM antipatterns. Our results may be particularly useful tor organizations in the
daunting journey of becoming agile.
0740-7459 (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more
Authorized licensed use limited to: University of Exeter. Downloaded information.
on June 22,2020 at 16:52:52 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/MS.2020.3001030, IEEE Software
0740-7459 (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more
Authorized licensed use limited to: University of Exeter. Downloaded information.
on June 22,2020 at 16:52:52 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/MS.2020.3001030, IEEE Software
They can encourage organizations to literature published over the last ten years that
bring to surface their SPM malpractices collected agile practices. The details are in [11].
and identify the agile practices that can We confirmed that all practices provided by [10]
address them. were included in the scientific sources (although
sometimes different names were used) and
They can be used in trade-off analysis to
there were not any extra agile practices included
define action plans containing the agile
in the scientific literature that were not
practices to be adopted according to
considered in [10]. Table 2 contains all the
their potential effects on the most
practices discussed in [10] with one exception.
critical antipatterns, among other
We included Kanban under the general practice
variables.
of Agile portfolio planning, as, according to [10],
This information can provide decision Kanban is a specific way to address this task, and
making support for senior management both generated similar results in relation to the
and teams in order to evaluate the SPM antipatterns.
potential benefits of transitioning to
The validity threats for this study relate to the
Agile.
risks suggested by Petersen et al. [12]. Regarding
search coverage we worked with the most well-
known scientific databases and defined the
AGILE PRACTICES search string with several keywords to obtain the
A quick internet search provides several lists of most accurate secondary papers within our
agile practices. One of the most well-known scope. We might have used the list of agile
catalogs is reported in [10]. It was gathered from practices suggested by [10], but we wanted to
organizations in the global software confirm such a list with the existing literature. To
development community. The agile practices avoid bias in the study selection, we defined
described in this report are classified into two detailed inclusion and exclusion criteria, and the
groups: agile techniques and engineering first two authors independently agreed on the
practices. In this work we will focus on the agile final selection of the secondary papers according
techniques, as the engineering practices to these criteria. Then, the search was
reported generally refer to technical activities complemented with snowball sampling. Finally,
(e.g. unit testing, refactoring) not directly linked to deal with the data extraction and classification
to management issues. threat, a data extraction form was filled by the
first author with the information about the
In order to confirm that the agile techniques in secondary papers, and analyzed and evaluated
[10] are also in accordance with the scientific by the second author to ensure that the selected
community we performed a tertiary study of data were meaningful for our study.
0740-7459 (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more
Authorized licensed use limited to: University of Exeter. Downloaded information.
on June 22,2020 at 16:52:52 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/MS.2020.3001030, IEEE Software
Table 2. Consolidated list of agile practices (ordered by their frequency of use in industry) [10]
Agile practice Description
Daily Stand-ups is one of the most common practices followed by agile teams. Is a meeting that takes
Daily standup
place every day, preferably face-to-face, to report updates on the daily work.
Sprint planning sessions held at the beginning of the sprint (iteration cycle), are intended to review
Sprint
and analyze possible changes and to define the sprint backlog, with the tasks to be completed in the
planning
sprint.
Retrospectives are the meetings held after the completion of an iteration to suggest process
Retrospectives
improvements for the following sprint.
Sprint review is a meeting where the team usually presents the sprint work to the product owner for
Sprint review
their feedback.
Short iterations typically seek to maintain a consistent sprint length. Its implementation behaves as a
Short
support for agile team accuracy while calculating the amount of work that they would be able to
iterations
perform.
Release Release planning is performed as an independent session for planning every release. Its purpose is to
planning estimate which features will be delivered in the established release deadlines.
Planning Planning poker is shaped by numbered cards in a Fibonacci series. It is used by the agile team as a tool
poker to estimate the value of each task that should be performed.
Available A product owner is responsible for managing the customer requirements, transforming them on
product backlog items and prioritizing it for then communicate them to the agile team.
owner
Single team is a practice in which theoretically, every team member is open to take up any work
Single team
irrespective of their skill set.
Frequent Frequent releases allow the team to see the state of the system and track if the deadlines are being
releases meet.
Common work Common working area supports frequent meetings, that lead to informal communication among
area stakeholders, which aids to the constant evolution of the requirements
Product Planning the roadmap of a product is part of the practices that help to have a general overview of what
roadmapping is expected from the product and how the system is being constructed based on the features.
Story mapping Story mapping is a technique to organize the product backlog by release and functionalities.
Agile portfolio An agile portfolio planning is a support to keep track of the activities that are in process. It shows the
planning status of each of the tasks that were planned for the sprint.
Agile UX seeks to integrate the user experience practices into the agile development cycle. Performing
Agile UX corporate design standards and getting feedback from clients to refine the requirements and
prototypes.
0740-7459 (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more
Authorized licensed use limited to: University of Exeter. Downloaded information.
on June 22,2020 at 16:52:52 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/MS.2020.3001030, IEEE Software
In order to check if there exists a more recent list papers), like [13][14]. We confirmed that the set
of SPM antipatterns we performed another of antipatterns found in these sources is a subset
ternary study of scientific literature dealing with of the ones in [6]. However, these publications
SPM antipatterns published from 2015 till June are evidence of the interest of the community on
2019. The details can be found in [11]. We did not SPM antipatterns.
find any secondary study with SPM antipatterns
In sum, the 22 SPM antipatterns identified in [6]
published later than 2015. However, we found
represent the consolidated list of SPM
information about SPM antipatterns in grey
malpractices that we worked with. They are
literature (sources different from scientific
described in Table 3.
documents like books, journals or conference
0740-7459 (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more
Authorized licensed use limited to: University of Exeter. Downloaded information.
on June 22,2020 at 16:52:52 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/MS.2020.3001030, IEEE Software
0740-7459 (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more
Authorized licensed use limited to: University of Exeter. Downloaded information.
on June 22,2020 at 16:52:52 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/MS.2020.3001030, IEEE Software
combinations of this agile practice have of the matching mainly providing a better
been identified with the Myopic Delivery or rationale for specific combinations. Later this
Dry Waterhole antipatterns, which will be matching was used for the interviews with the
discussed later. rest of participants. Interviews discussed the
details of the Agile/SMP antipatterns mapping,
Table 4 was created through detailed analysis
and the potential benefits of the results for the
and discussion among the authors according to
industry.
the definition of agile practices and antipatterns
provided in Table 2 and Table 3, respectively; and Regarding the Agile/SPM relationship, personal
the evidence gathered from our experience in experience of each interviewee had a relevant
the software industry for more than 40 years in impact of the interpretation of each
software project management and consulting combination. Even if a consensus was reached, it
with firms ranging from high technology startups was clear that although keeping its essence, each
and business management to large multi- agile practice can be implemented and adapted
national corporations. Additionally, we surveyed in different ways by different organizations, or
a total of 30 agile practitioners through face to even by different projects in the same
face interviews to discuss the matching results. organization. Consequently, the matching
The interviewed practitioners came from a wide process might provide slightly different results
range of domains (banking, stock market, for different organizations. Regarding specific
transport, e-health and education). They are combinations, interestingly, six participants,
geographically distributed in North America, explicitly recognized the negative effects of
South America and Europe, have a mean number Single team and Domino Effect inside their own
of six years of experience in Agile, and more than organizations.
80% of them are certified as Scrum Masters. We
Regarding the potential benefits, interviewees
surveyed 30 professionals as this number is used
highlighted the idea of the matching to put on
in empirical studies in software engineering to
the table potential SPM issues and explicitly
have a representative sample [15].
discuss how Agile can contribute to address
Unstructured interviews were used to discuss them. More than half of the participants
the matching. This matching was distributed in explicitly highlighted the need for commitment
advance to the participants and later the face to and support from the management in the
face interviews were set for about one hour. In transitioning process and how these results
the first, round, we interviewed ten participants could contribute in that direction.
that contributed to consolidate the first version
0740-7459 (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more
Authorized licensed use limited to: University of Exeter. Downloaded information.
on June 22,2020 at 16:52:52 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/MS.2020.3001030, IEEE Software
0740-7459 (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 22,2020 at 16:52:52 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/MS.2020.3001030, IEEE Software
Figure 1 quantifies the effect of each agile However, the specific impact, including possible
practice on the antipatterns. Sprint planning collateral harm, of each SPM malpractice in an
followed by Planning poker and Agile UX are the organization or even project, should be carefully
agile practices that contribute to address the analyzed to evaluate the criticality of each
highest number of antipatterns (15, 13 and 13, malpractice. This information, along with its
respectively). Among them Fire Drill, Mushroom implementation cost should be part of the
Management or Myopic Delivery have special decision making process about adopting an agile
negative implications on the customer and practice. In this sense, it is worth mentioning
therefore should be carefully considered [6]. Agile UX, which addresses multiple antipatterns,
Other antipatterns that these practices address but according to [10], is the least used agile
are Detailitis Plan, Process Disintegration or practice. Its implementation is costly and has
Project Mismanagement which according to [6] important implications in the agile process itself
have an important effect on developers and the as well as the team.
project manager itself. Therefore, the previous
agile practices might be good candidates to be
incorporated early in an Agile transition process.
18
16
Number of SPM antipatterns
14
12
10
0740-7459 (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more
Authorized licensed use limited to: University of Exeter. Downloaded information.
on June 22,2020 at 16:52:52 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/MS.2020.3001030, IEEE Software
Retrospectives is the agile practice that most impact on solving SPM risk scenarios does not
contributes to identifying SPM malpractices. It is mean that it will contribute less to the success of
aimed at reviewing the whole development the project. For example, Available product
process in searching problems and potential owner has low direct impact on SPM
improvements. Consequently, in the case of SPM malpractices. However it is essential from the
problems, retrospectives is the place where they requirements perspective and therefore, for
should come out. Additionally, Retrospectives project success.
help to reduce the impact of a few antipatterns,
Absentee Manager, All you Have is Hammer,
PRACTICAL IMPLICATIONS AND FINAL
Appointed Team and Process Disintegration. So, REMARKS
in general it has a positive effect over all SPM Project management malpractices need to be
malpractices. This practice can be implemented brought to surface, discussed and addressed.
with low cost and therefore is a good candidate Table 4 can be used to identify agile practices
to be incorporated in the first phases of an agile which can be useful to address most relevant
transition. In fact, it is one of the most used agile SPM problems. But each organization is
practices [10]. responsible of determining the negative effects
An unexpected result was the possible negative that particular SPM malpractices have on its
contribution of a specific agile practice, Single business value and consequently the risk
team. We have already mentioned the associated. The decision of whether to
combination Single team/Domino Effect. This incorporate an agile practice in a software
practice can also have other negative effects, for development process is a tradeoff between
example wrongly assuming that having a team different variables, among them, its ability to
where members can take different roles allows address the most harmful and frequent SPM
setting a fixed delivery date (Myopic Delivery) as malpractices and the cost of their
multidisciplinary teams will be able to recover implementation.
the work. Additionally, the idea that team We do not mean that one practice will solve a
members need to be able to perform any kind of particular anti-pattern. Most generally, the
technical activity, might raise the expectations of implementation of several practices working
the organization defining very strict altogether are needed for this purpose. Again
requirements and incrementing the effects of each organization, needs to monitor the process
the Dry waterhole antipattern, resulting in a very and take the corresponding corrective actions, if
limited pool of available talent. Even when the needed.
application of this agile practice has recognized
positive effects and it is an essential element of Finally, this paper discusses the link between
the agile philosophy, the case of these Agile and SPM malpractices, providing
combinations represents possible risky situations organizations a preliminary mapping between
that should be carefully considered. them, which might need to be adapted according
to each organization’s specifics. This information
In this study we are focusing on agile practices can provide decision support in an agile
and their potential effect on SPM malpractices, transition process. However, the adoption of
but the fact that one agile practice can have less agile practices will not free agile teams of SPM
10
0740-7459 (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more
Authorized licensed use limited to: University of Exeter. Downloaded information.
on June 22,2020 at 16:52:52 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/MS.2020.3001030, IEEE Software
11
0740-7459 (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more
Authorized licensed use limited to: University of Exeter. Downloaded information.
on June 22,2020 at 16:52:52 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/MS.2020.3001030, IEEE Software
12
0740-7459 (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more
Authorized licensed use limited to: University of Exeter. Downloaded information.
on June 22,2020 at 16:52:52 UTC from IEEE Xplore. Restrictions apply.