704
Views
16
CrossRef citations to date
0
Altmetric
Articles

Developing software beyond customer needs and plans: an exploratory study of its forms and individual-level drivers

ORCID Icon, ORCID Icon, ORCID Icon & ORCID Icon
Pages 7189-7208 | Received 10 Dec 2017, Accepted 03 Feb 2019, Published online: 23 Feb 2019
 

Abstract

Excessive software development is the tendency to develop new software above and beyond the requirements of the market and/or planned specifications. It is a widespread phenomenon involving both risks and flexibility advantages. As it represents a challenging dilemma for software developers, it is important to study its human origins. Drawing on the tripartite model of individual attitudes, this study investigates the influence of developers’s cognitive (intuitive and rational thinking styles), affective (emotional attachment) and behavioural (reliance on past experiences) traits on two forms of excess, beyond needs and beyond plans. Using survey data on 307 software developers, this study shows that different manifestations of excess are associated with distinct traits of software developers. Emotional attachment drives beyond needs excess. A positive (negative) association is found between relying on past experiences and beyond needs excess (beyond plans excess). An intuitive cognitive style fosters the inclusion of extra features in the new product scope, whereas a rational style might lead to developing one-size-fits-all software that targets the needs of a broad user base. These findings contribute to research on the development of digital new products and production technologies by offering a comprehensive yet fine-grained picture of excessive software development’s nature and drivers.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Another tactic that offers development flexibility is the use of a modular architecture for the new product (Peng and Mu Citation2018). Modular designs enhance the ability to accept changes and additional features in the course of a project and throughout the product lifecycle. We thank an anonymous reviewer for this comment. While potentially a limitation, the role of a modular design for the new software is not included in the scope of this study.

2. Another tactic that offers development flexibility is the use of a modular architecture for the new product (Peng and Mu Citation2018). Modular designs enhance the ability to accept changes and additional features in the course of a project and throughout the product lifecycle. We thank an anonymous reviewer for this comment. While potentially a limitation, the role of a modular design for the new software is not included in the scope of this study.

3. A person’s cognitive style refers to his/her preferred way of organizing and processing information to reach a decision (Messick Citation1976). A rational style involves slow, conscious, and analytical processing of information, while an intuitive style is fast, unconscious and instinctive (Kahneman Citation2003). Emotional attachment is defined as an emotion-laden target-specific bond between a person and a specific object (Thompson, Hamilton, and Rust Citation2005). Reliance on past experiences is here defined as using knowledge gained in previous projects as reference information to make decisions about the current initiative (Shmueli, Pliskin, and Fink Citation2015).

4. Software development projects are naturally characterized by heterogeneous levels of uncertainty and volatility. Because the unit of analysis in this study is the aggregate development process, it does not measure project-level factors. Although this is a limitation, which is common to other studies (Tripp and Armstrong Citation2016), Furr and Dyer (Citation2014) rank the software industry as a whole in the top three for technical and market uncertainty.

5. A first round of data collection targeted four communities, Italia JavaScript (152 of the total 870 members responded to the questionnaire, a raw response rate of 17.4%), Google Development Group Slack Milan (27 of the total 83 members responded, a raw response rate of 32.5%), Google Development Group Florence (40 of the total 219 members responded, a raw response rate of 18.3 %), ASP.NET Italia (30 of the total 552 members responded, a raw response rate of 5.4%). This dataset was used for the empirical study by Bianchi, Marzi, and Guerini (Citation2018), which addresses a different research question and draws on different concepts and theories than the present study. A second round of data collection shortly after targeted the remaining six communities: WordPress Italia (53 of the total 209 members responded to the questionnaire, a raw response rate of 25.4%), Gnome (45 of the total 649 members responded to the questionnaire, a raw response rate of 6.9%), Grav Developers (42 of the total 2096 members responded to the questionnaire, a raw response rate of 2%), WordPress International Slack (39 of the total 1574 members responded to the questionnaire, a raw response rate of 2.5%), Gimp Developers (33 of the total 92 members responded to the questionnaire, a raw response rate of 35.9%), Joomla Core Code Developers (18 of the total 54 members responded to the questionnaire, a raw response rate of 33.3%). These additional data are not used in previous publications.

6. The final sample consists of: 117 members of Italia JavaScript; 21 members of Google Development Group Slack Milan; 26 members of Google Development Group Florence; 17 members of ASP.NET Italia; 24 members of WordPress Italia; 28 members of Gnome; 27 members of Grav Developers; 23 members of WordPress International Slack; 7 members of Gimp Developers; 17 members of Joomla Core Code Developers. Because the sample size is low compared to the large population of Internet software developers and because individual developer effects can be nested in other higher-level, e.g., community-level, effects, results should be interpreted with caution.

7. The elimination of item BNE6 from factor FD-BNE should actually strengthen the validity of this factor, because its formulation only implicitly related to the inclusion of extra, non-core features, differently from the other FD-BNE items that contain explicit links.

 

Log in via your institution

Log in to Taylor & Francis Online

PDF download + Online access

  • 48 hours access to article PDF & online version
  • Article PDF can be downloaded
  • Article PDF can be printed
USD 61.00 Add to cart

Issue Purchase

  • 30 days online access to complete issue
  • Article PDFs can be downloaded
  • Article PDFs can be printed
USD 973.00 Add to cart

* Local tax will be added as applicable

Related Research

People also read lists articles that other readers of this article have read.

Recommended articles lists articles that we recommend and is powered by our AI driven recommendation engine.

Cited by lists all citing articles based on Crossref citations.
Articles with the Crossref icon will open in a new tab.