The Hang-Over Effect in Information-Intensive Service Systems

The literature on the business value of enterprise information systems is replete with stories of dramatic successes and colossal failures. Often some applications of a software system deliver benefits while its other applications do not. It is therefore not clear how these complex systems deliver benefits beyond the general argument of labor to capital substitution and why their impacts are unpredictable. We discuss a newly discovered factor that partly explains this unpredictability - the "Hang-over" effect. We report the results of a long-term empirical study that looked at the implementation of a newly developed Radiology Information System (RIS) at a regional network of radiology clinics. We show that the hang-over effect explains why the same system produced a beneficial impact on one modality (mammography) and a negligible impact on another (MRI), and that it also explains why an information system does not produce similar benefits across modalities or across customer classes.