Characterising users through an analysis of on-line technical support forums

Users of software systems need support. When users choose to go directly to online forums rather than report issues to the source of the problem, the development organization loses out in terms of gathering information on where their systems could be improved, and can lose customer loyalty and goodwill. This in turn will have a negative effect on future sales and system enhancements. The objective of this study is to examine online forums in order to characterize technical support (TS) users, and create a framework that allows commercial organizations (and other interested parties) to identify types of users and how best to address their needs. One hundred and sixteen threads (3,064 messages) from eight online open source forums were analysed using a grounded theory approach. Also the literature on human factors, and personalisation was examined to elicit information on how people can be categorized.We found that users of TS systems can be grouped according to their level of expertise and what they value. Additionally we identified characteristics of the communication handling process that influence desirable and undesirable outcomes. The contribution of this research is an empirically derived framework that identifies TS users according to groups of characteristics. We hypothesise that a user will be more satisfied when the TS service provider recognises concepts such as what they personally value, their level of expertise and how best to manage emotions.