Situational Fit in Incremental Method Engineering

Introduction

Many software companies struggle with improving their information system development methods (cf. Conradi, Fernström, & Fuggetta, 1993; Niazi, Wilson, & Zowghi, 2005; Coleman & O’Connor, 2008). Pino, García, and Piattini (2008, p. 253) found in a systematic literature review that “the software engineering community has shown an ever-increasing interest in tackling software process improvement in SMEs”. However, they also found that small companies find it hard to use the standard SEI and ISO models such as CMM and SPICE as they need to be adapted in order to be used successfully.

The domain in which we carry out our research is the domain of software product management (SPM). SPM is the discipline that governs a software product over its whole life cycle, from its inception to customer delivery, in order to generate the biggest possible value to the business (Ebert, 2007). Also in the case of SPM, companies find it hard to bring their processes to a higher level (van de Weerd, Versendaal, & Brinkkemper, 2006). To improve their processes, they need to design new methods or adapt the existing ones, while there is little education available in SPM domain (van de Weerd, Brinkkemper, Nieuwenhuis, Versendaal, & Bijlsma, 2006). Several authors proposed method engineering approaches to create or adapt methods to the company or situation at hand (cf. Brinkkemper, 1996; Aydin & Harmsen, 2002; Ralyté, Deneckère, & Rolland, 2003; Ågerfalk et al., 2003). Van de Weerd, Brinkkemper, and Versendaal (2007) introduced the term incremental method engineering to describe the implementation of method adaptations over time, in order to improve the overall performance of a method. A method increment is defined “a collection of method fragments that have been introduced in the method during the method adaptations between tⁱ and t^i-1” (van de Weerd, Brinkkemper, & Versendaal, 2007, p. 474). An important factor here is whether the methods or method fragments fit the situation of the company or project for which it is selected. We call this term situational fit. Situational fit is not a new term. In literature on organizational design, situational fit is defined as “the balance of the environmental conditions, the strategy, the management style, the climate, the size, and the technology” (Burton, Lauridsen, & Obel, 2002, p. 1463). Translating this to the method engineering domain, we define situational fit as the balance of the environmental characteristics, company characteristics, and information system development methods.

The aim of this study is to deliver a proof of concept of how situational fit and incremental method engineering can be used as a selection mechanism for process improvement in software companies. The main research question in this study is formulated accordingly:

By answering this question we aim to contribute to the field of method engineering by showing how the adaptation of a method can lead to a higher maturity level of that method. We elaborate on how method increments, based on situational fit of the method and the company, can improve the processes in the software companies.

In the following section, we first give an overview of related work. Then, we describe our research approach. In the subsequent sections, we present the results of the different steps followed in this approach: method selection and analysis; an exploratory case study in which we explore how incremental method engineering can be used to support process improvement; and a quantitative analysis to evaluate the concept of situational fit. Finally, we present our conclusions and directions for future research.