Software engineering in start-up companies: An analysis of 88 experience reports

Abstract

Context Start-up companies have become an important supplier of innovation and softwareintensive products. The flexibility and reactiveness of start-ups enables fast development and launch of innovative products. However, a majority of software start-up companies fail before achieving any success. Among other factors, poor software engineering could be a significant contributor to the challenges experienced by start-ups. However, the state-ofpractice of software engineering in start-ups, as well as the utilization of state-of-the-art is largely an unexplored area. Objective In this study we investigate how software engineering is applied in start-up context with a focus to identify key knowledge areas and opportunities for further research. Method We perform a multi-vocal exploratory study of 88 start-up experience reports. We develop a custom taxonomy to categorize the reported software engineering practices and their interrelation with business aspects, and apply qualitative data analysis to explore influences and dependencies between the knowledge areas. Results We identify the most frequently reported software engineering (requirements engineering, software design and quality) and business aspect (vision and strategy development) knowledge areas, and illustrate their relationships. We also present a summary of how relevant software engineering knowledge areas are implemented in start-ups and identify potentially useful practices for adoption in start-ups. Conclusions The results enable a more focused research on engineering practices in startups. We conclude that most engineering challenges in start-ups stem from inadequacies in requirements engineering. Many promising practices to address specific engineering challenges exists, however more research on adaptation of established practices, and validation of new start-up specific practices is needed.

Exploration of technical debt in start-ups

“…Demographic information from respondent answers shows that our sample is skewed towards active companies, respondents with little experience in start-ups, young companies and small development teams of 1-8 engineers. In these aspects our sample fits the general characteristics of startups, see for example, Giardino et al [10,11] and Klotins et al [18]. However, there clearly is a survivorship bias, that is, failed start-ups are underrepresented, thus our results reflect state-of-practice in active start-ups.…”

An anatomy of requirements engineering in software startups using multi-vocal literature and case survey

“…To find the cases for this study, a questionnaire was created that aimed at collecting information regarding the use of software engineering methods and practices in software startups [34]. The questionnaire was inspired by the co-authors' earlier work [35] on software engineering in startup companies, which was based on a multi-vocal exploratory study of 88 startup experience reports, and thus served as a foundation for our case survey. The questionnaire contained 10 sections that addressed different software engineering aspects of software startups and retrieved meta-information about the cases.…”

How to Design a Successful Digital Product? An Analytic Hierarchy Process (AHP) Analysis of Expert Opinions from the Berlin Start-up Scene

“…How an organization manages its programming languages and codes, develops its software, and operates its server, database and architecture [38], [39].…”