Abstract
This article presents evidence that--alongside the successes-- many information systems in developing countries can be categorized as failing either totally or partially. It then develops a new model that seeks to explain the high rates of failure. The model draws on contingency theory in order to advance the notion of design-actuality gaps: the match or mismatch between IS designs and local user actuality. This helps identify two high-risk archetypes that affect IS in developing countries: country context gaps and hard-soft gaps. The model is also of value in explaining the constraints that exist to local IS improvisations in developing countries. Overall, the article shows how model and theory help understand IS cases in developing countries, and equally, how those cases provide valuable data to help develop IS models and theories.