This study was conducted for the final part of an MSc in Technology Management and was concerned with risk in web-development projects delivered by distributed development teams. It asks whether, and if so how, the use of geographically distributed teams affects the risk profile of projects when compared with those delivered by co-located teams. The practical output of this work is a risk framework for web-development Project Managers working with distributed teams that can act as a check-list to help them identify potential issues, plan risk-handling activities, and support them through the project lifecycle. A further goal of this work is to understand how this specific risk framework compares with frameworks developed for non-distributed teams in the web industry, and frameworks in the wider IT field.
To understand the scope of existing risk frameworks, an initial literature survey was conducted across research covering both the ‘traditional’ IT sector and web industry. Additionally, two questions were asked; does risk indeed change when teams are distributed; and secondly are web projects simply a subset of the ‘IT projects domain’ or are they unique? Research showed that authors such as Sudhakar (2013) and Betz (2007) propose that risk profiles in IT projects change when teams are distributed; affecting communication and development tools which are also core to the delivery of web projects implying likely change. Al-Rousan et al. (2009) and Steele & Carter (2001) have argued that whilst there are similarities between traditional IT development and web there are also significant differences, to the degree that web projects should be considered as unique. As part of a wider study Keshlaf and Riddle (2010) note that risks in web development are increased when distributed teams are used, however outside this reference no prior research has been identified that focuses purely on risks caused by distributed web development teams.
To investigate this further, and to gather data to build the proposed Risk Framework, research interviews were held with delivery team members working for a digital agency. This agency develops enterprise level content-managed websites for major corporate clients across many industries. The agency’s delivery and development teams are spilt across London, New York and Poland which makes the majority of their projects ‘distributed’.
Face-to-face and video interviews were held and recorded with eleven Project and Programme Managers based in the UK, Poland and New York to identify risks that they had encountered, or anticipated, on projects delivered using distributed teams. The recordings were analysed and the individual risks extracted and grouped into similar themes.
The identified themes formed the basis of the finalised Framework which consists of 108 discrete ‘Risks’ grouped into 25 ‘Risk Themes’, themselves contained within the following 10 ‘Major Themes’:
One of the key findings from this study is that all participants described the seventh item, ‘communication’, as the largest area affecting distributed web projects. Within this theme the most significant risks were felt to be caused by the increased time taken to brief staff, as well as a lack of face-to-face interaction leading to weaker teams. Equally ‘problems communicating change’ were raised repeatedly, together with ‘communicating progress to the project team’. Overall staff motivation and a shared sense of purpose were seen as hard to develop across a distributed team.
Interestingly, outside of ‘communication’, neither culture nor distance were considered major issues for the team on projects where the team was based in Northern or Central Europe. Equally as teams move south and east more issues were raised as the risks became more cultural, especially when the ‘common’ project language of English wasn’t spoken so well. US teams were felt to cause risk due partly to distance and time zone, but also for a higher likelihood of team fatigue in maintaining a shared window of working hours. Communication issues were raised with remote teams in India and Asia, with issues raised particularly by non-native English speakers around ‘hard to understand’ accents, as well as poor communications infrastructure.
On the positive, an observation that was raised repeatedly is the value of personal relationships and how they mitigate distance and cultural risk; this was stressed by the agency’s founder in his interview as he discussed how the company had been set up with teams in both the UK and Poland.
Comparing these results with a previous risk framework (McDowall, 2005) for non-distributed web development teams it can be seen there are some common themes across both studies e.g. ‘team’ and ‘communication’, and that some specific risks are the same e.g. ‘client's other commitments makes communication slow’, and “complex organisational decision making process’. Equally areas like ‘requirements uncertainty’ which figured heavily in the 2005 study of co-located teams is not raised as an impacted risk in this study.
Other researchers have looked at the wider area of offshoring general IT software development, and researchers like Islam et al. (2011) have identified similar themes including ‘communication’ and ‘project coordination’ as being essential in an offshore development context and when poorly managed increases risk significantly. This is clearly in line with this study’s findings and indicates commonality between distributed projects regardless of whether they are web focused or more general IT related.
The findings from this study strongly indicate that there is a change in the risk profile when teams are distributed. This is seen by the participants as overwhelmingly negative, in that risks are increased across all of the theme areas identified and with particular emphasis on Communication.
The risks identified above are summarised as a table here
| Home |