Integrating UX work with agile development through user stories: An action research study in a small software company

https://doi.org/10.1016/j.jss.2020.110785Get rights and content

Highlights

  • UX integration is complex, contextualized, and difficult in practice.

  • Verbose user stories sustain defensive behavior between the two practices and.

  • Deliberation based on concise user stories helps improve the integration of UX.

Abstract

The integration of user experience (UX) work with agile software development has been addressed in extensive research of challenges and process models. However, in-depth research of context-specific improvements of this integration with actual UX professionals and agile developers in their actual practice is limited. This study examines how the integration of UX work with agile development can be improved in the context of a small Danish Software-as-a-Service (SaaS) company. We used the problem- and solution-oriented action research method over 12 months in the company. During this period, we initially carried out extensive participant observations, recorded 32 semi-structured interviews, and finally conducted two improvement iterations with evaluations of their effect on agility. We identified user stories as an essential indicator of UX integration. Verbose user stories imply problems in collaboration and trust, while concise user stories and deliberation improve integration of UX work with agile development. The conclusion is that integrating UX work with agile development in practice is complex, contextualized, and difficult even for only a small part of it. We propose that concise user stories and deliberation can be useful and well-defined focuses for integrating UX work with agile software development without sacrificing their agility.

Introduction

During the last two decades, both agile development and user experience (UX) work have become standard practices in the software industry (Brhel et al., 2015, Bruun et al., 2018, Chilana et al., 2010, Larusdottir et al., 2017). Both domains strive to build quality software (Ferreira et al., 2011) and contribute to economic success in highly competitive markets (Brhel et al., 2015). However, agile development and UX work utilize different approaches, values, and views on what quality software is Ferreira et al., 2011, Larusdottir et al., 2017. Even though both domains are iterative, the rigor of UX up-front activities clashes with the rapidness of agile development (Larusdottir et al., 2017), which imposes a challenge for integrating UX with agile development (Chilana et al., 2010, Ferreira et al., 2011).

UX integration has been an area of interest for the academic community for over a decade (Brhel et al., 2015) and has generated numerous studies of processes, challenges, and success factors (Brhel et al., 2015, da Silva et al., 2013, Kuusinen et al., 2012). These studies are valuable in gaining an understanding of the current state of UX integration; however, they have captured only a glimpse of a practice, lacking an understanding of events that have an effect on the transformation of a practice (Kashfi et al., 2019).

One of the main challenges within UX integration is to build a common ground between UX professionals and agile developers (Garcia et al., 2019). Artifacts, such as prototypes and user stories, aid the establishment of a common ground and a common understanding (Brhel et al., 2015, Garcia et al., 2019, Jones and Thoma, 2019). Artifacts are boundary objects between UX professionals and agile developers, and are used during collaborative activities (da Silva et al., 2018, Garcia et al., 2019). Artifacts are therefore seen as a fundamental part of the software development process (Zaitsev et al., 2020). A user story is a popular artifact within software development (Lucassen et al., 2016). The user story method stems from agile discipline and is used to describe user values and needs (Cohn, 2004). A user story is short, comprehensible, and negotiable in order to mediate information to any kind of stakeholder regardless of their department, educational background, or technical insight (Cohn, 2004). Thus, user stories are beneficial for the interdisciplinary work.

Generalizable and rational solutions are not effective because these are rarely followed in practice due to local circumstances within each organization (Ferreira et al., 2011). Thus, UX integration research needs a more nuanced view of how user stories can be useful in practice, which requires long-term studies (Kashfi et al., 2019) that emphasize the situated practice in which the issue of UX integration is embedded (Ferreira et al., 2011). A case study, a prevalent research method used when studying UX integration with agile development, makes it difficult to uncover what aspect of a practice improves UX integration (da Silva et al., 2015). Thus, researchers ought to make interventions that are tailored for the local practice (Mckay and Marshall, 2001).

We contribute to the calls for research (Kashfi et al., 2019) with an action research (AR) study (Hayes, 2011, Mckay and Marshall, 2001) conducted in a small Danish Software-as-a-Service (SaaS) company – ServiceSoft. Our aim was to determine how we can improve the integration of UX work with agile software development. The use of we implies not only we as researchers but also the practitioners, thus emphasizing the collaborative nature of AR. The verb improve refers to change, which is a core principle of AR, to benefit practice and contribute to research (Mckay and Marshall, 2001). Integration implies that we recognize UX and agile software development as distinct but mutually dependent practices and areas of research. Segregating them into two independent areas would undermine their interdependent relationship in making useful software (Detweiler, 2007, Ferreira et al., 2011). Thus, in this study, we as researchers collaborated with practitioners on making changes that improve the integration of UX work with agile software development while maintaining the distinctive strengths of the two practices.

In the following, we first contextualize our study in related research. Afterward, we describe our AR approach to address our research question and the problematic situation in a specific company, ServiceSoft. ServiceSoft is a pseudonym used throughout the article to ensure the anonymity of the company. In the findings section, we present our two interventions at ServiceSoft and note what consequences they had on the agility of its practice. Finally, we discuss the contribution of our findings, limitations, followed by a conclusion.

Section snippets

Related research

In the following section, we appreciate the problem of UX integration in agile software development through related research. In Section 2.2, we present user stories, and discuss artifact mediated communication as a possible solution to the problem of UX integration.

Methodology

To further our understanding of how we can improve the integration of UX work with agile software development, we used an AR method in “the seeking of solutions or improvements to ‘real-life’ practical problem situations” (Mckay and Marshall, 2001 p. 47). AR is a problem- and solution-oriented method of initiating a sustainable change that emphasizes both theory and practice with the objective to apply, as well as add to, the existing body of knowledge (Mckay and Marshall, 2001), in this case

Findings

In this section, we present the problem situation at ServiceSoft and our two interventions to improve the integration of UX work with their agile software development.

Discussion

Integrating UX work with agile development is a considerable challenge that has been a research interest for over a decade (Brhel et al., 2015, da Silva et al., 2018) with many studies of its challenges and success factors (da Silva et al., 2013, Brhel et al., 2015, da Silva et al., 2018, Kashfi et al., 2019). However, few studies are concerned with the problem identification and improvement in practice through AR. Against this backdrop, we present two key contributions from our two AR

Limitations

Our study focused on how user stories and deliberation affect collaboration on the activities prior to software implementation in exploring and establishing a common understanding of user needs and technological limitations. The user’s perspective is a continuous concern; therefore, future work is needed to understand how concise user stories and deliberation affect other activities in agile software development. Furthermore, in follow-up interviews with the product manager and the head of

Conclusion

Integrating UX work with agile software development is complex, contextualized, and difficult in practice. From two AR iterations in a small software company, we showed how concise user stories and deliberation are useful focuses. Verbose user stories with voluminous context descriptions are artifacts that, as such, should support collaboration between UX professionals and agile developers (Garcia et al., 2019). However, our study showed that verbose user stories became a wall between the two

CRediT authorship contribution statement

Alisa Ananjeva: Conceptualization, Data curation, Formal analysis, Investigation, Writing - original draft. John Stouby Persson: Methodology, Conceptualization, Supervision – ensuring that data curation is focused, Writing - review & editing. Anders Bruun: Conceptualization, Supervision – ensuring that data curation is focused, Writing - review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Alisa Ananjeva, M.Sc., is a Ph.D. fellow at the Department of Computer Science at Aalborg University, Denmark. Alisa’s research interest lies in the interface between Information Systems (IS) and Human–Computer Interaction (HCI). She is interested in problem-based research and research methodologies such as Action Research and Action Design Research. Alisa aims at finding and solving problems in collaboration with the industry practitioners.

References (48)

  • AndreiG. et al.

    Artifacts for agile user-centered design: a systematic mapping

    HICSS

    (2017)
  • AvisonD.E. et al.

    Action research

    ACM

    (1999)
  • BarksdaleJ. et al.

    Software product innovation in agile usability teams: an analytical framework of social capital, network governance, and usability knowledge management

    Int. J. Agile Extreme Softw. Dev.

    (2012)
  • BrinkmannS.

    Doing without data

    Qual. Inq.

    (2014)
  • Brown, J., Lindgaard, G., Biddle, R., 2011. Collaborative events and shared artefacts: Agile interaction designers and...
  • BruunA. et al.

    The role of UX professionals in agile development: A case study from industry

  • ChilanaP.K. et al.

    Understanding usability practices in complex domains

  • CohnM.
  • ConboyK.

    Agility from first principles: Reconstructing the concept of agility in information systems development

    Inf. Syst. Res.

    (2009)
  • da Silva, T.S., Silveira, M., Maurer, F., 2013. Ten lessons learned from integrating interaction design and agile...
  • da SilvaT.S. et al.

    A systematic mapping on agile UCD across the major agile and hci conferences

  • DetweilerM.

    Managing UCD within agile projects

    Interactions

    (2007)
  • DourishP.

    The Stuff of Bits: An Essay on the Materialities of Information

    (2017)
  • FerreiraJ. et al.

    User experience design and agile development: Managing cooperation through articulation work

    Softw. Pract. Exp.

    (2011)
  • Cited by (24)

    • Agile software development and UX design: A case study of integration by mutual adjustment

      2022, Information and Software Technology
      Citation Excerpt :

      The conceptualizations provided a starting point for deliberation on their practices in the workshop. While this starting point for deliberation is more abstract by focusing on the overall integration process than previous research on, e.g., concise user stories [8] or technical debt items [38], it still allows positioning and discussing Agility in their specific situation. We use the three views on the Agility of software development and UX design (cf. Fig. 1) to explain our case study at Mjølner Informatics and as a basis for proposing the three distinct claims about Agility presented in Fig. 2.

    • Guiding the selection of research methodology in industry–academia collaboration in software engineering

      2021, Information and Software Technology
      Citation Excerpt :

      This is, in fact, the aim of TAR with its dual engineering and research cycles, focusing on action and generalized design knowledge, respectively. To illustrate AR, as such, we select Ananjeva et al.’s study, aiming to integrate user experience (UX) work with agile software development, as our illustrative example [83], referring to artifacts and activities in Fig. 6. The AR study is conducted jointly (Collaboration) between a small software-as-a-service company (Industry), an on-site observer (Master thesis student), and researchers with software engineering and UX competence (Academia) over 12 months.

    View all citing articles on Scopus

    Alisa Ananjeva, M.Sc., is a Ph.D. fellow at the Department of Computer Science at Aalborg University, Denmark. Alisa’s research interest lies in the interface between Information Systems (IS) and Human–Computer Interaction (HCI). She is interested in problem-based research and research methodologies such as Action Research and Action Design Research. Alisa aims at finding and solving problems in collaboration with the industry practitioners.

    John Stouby Persson, Ph.D., is an associate professor at the Department of Computer Science at Aalborg University, Denmark. He does collaborative practice research on managing the development of information systems. Following a pragmatist philosophy, his research interests pertain to the management of global software development, agility, digital transformation, business cases, and benefits realization. In this context, he has published qualitative studies and action research of agile software development over the past 10 years in predominantly Information Systems research outlets.

    Anders Bruun, Ph.D., is an associate professor at the Department of Computer Science, Aalborg University, Denmark. Anders’ research in HCI deals with theoretical, methodological and practical aspects of user experience and interaction design. This research emphasizes development and assessment of novel methods that reliably measure UX in digital products, including a combination of quantitative real time measurements from wearable sensors as well as post-hoc qualitative responses from users. Anders’ research interests also include studies of the challenges arising in the intersection between software development and interaction design practices. Industry collaborations have led to publications on e.g. identifying and overcoming challenges related to integrating user experience and software development practices.

    View full text