58 January 2000/Vol. 43, No. 1 COMMUNICATIONS OF THE ACMCOMMUNICATIONS OF THE ACM January 2000/Vol. 43, No. 1 59ReexaminingORGANIZATIONALMEMORYIn this article we step back and investigatewhere memory exists currently within anorganizational setting, rather than focusingon potential technical enhancements. Inorder to accomplish this we study OMwithin a telephone helpline that answershuman-resource questions at a well-established Sili-con Valley company.We approach this problem using distributed cog-nition theory [4, 6] because its theoretical languagespans the diverse manifestations of OM—from pri-vate to public, small scale to large scale. Our goal isto find theoretical concepts that can help us framebetter designs.We begin with a synopsis of the OM literatureand the need for empirically based analyses of OM.We next provide a description of the field site andethnographic data collection. We then analyze a sin-gle helpline call, progressively describing the parts ofthe call, the surrounding work activity, and thememory used in the work activity. THE OM LITERATURE HOLDS MANY VARYING, ANDoccasionally competing, definitions. Intuitively, orga-nizations should be able to retrieve traces of theirpast activities, but the form of this memory isunclear in research literature. Early efforts assumeone could consider memory as though it were a sin-gle, monolithic repository of some sort for the entireorganization. This viewpoint proves largely unten-able. However, this technical reconsideration has notbeen matched with a theoretical reconsideration.Mark S. Ackerman and Christine A. HalversonReconceptualizing how an interpersonal memory—particularly oneincluding people and technology—may be defined.NORMAND COUSINEAUAfter nearly 10 years of research, “organizational memory” (OM) has become overworked and confused. Burdened by a practical wish to reuse organizational experience,researchers have often ignored critical functions of an organization’s memory in order to focuson only a few methods for augmenting memory. It is time for a reexamination.Indeed, little research rests on an empiricalexamination of OM within a context of use, eventhough there have been repeated calls for thesetypes of studies. For example, Walsh and Ungson,in their canonical article, note that: “a myriad ofunexamined conjectures has defined a concept thathas even served as a basis for prescriptive manage-ment advice” [11]. Indeed, Walsh and Ungson do notcite any empirical studies.Generally, other OM theory is at a grand scale,without empirical data. For example, Huber [5]argues that organizational learning and memorysupport would be useful, but he does not clearly dis-tinguish what constitutes OM. Stein and Zwass [8],while acknowledging the need for empirical studies,nonetheless rely on an organization’s high-levelmodel. Other studies have primarily focused on the tech-nology systems designed to replace human andpaper-based memory systems. Many of these studiesexamine memory systems in use, but the studies arelimited to particular systems, often prototypes (forexample, [1]). This lack of empirical examinations is unfortu-nate. The need for systematic work in order toexamine OM is pressing because as it is often con-ceived, OM lends itself to a number of theoreticalproblems. For example, organizations are hardly asingle, unified entity, as the metaphor implies. Nonetheless, we do not argue for abandoning orignoring OM as a concept. Despite the problems,there is something compelling about the idea. AsBannon and Kuutti [3] state: “That such a conceptis appealed to across a wide range of studies, even ifits definition is disputed, is testimony to the factthat even if people cannot agree on what exactly theterm means, there must be some set of issues … thatpeople feel are important and worth discussing.” However, OM as a theoretical concept mustresult initially from studies within an organizationalfield setting, that is, within a context of everydayuse. This allows the research community to recon-sider how an inter-personal memory—particularyone including people and technology—may bedefined. With such a basis, system construction canbe instituted upon empirically determined insightsinstead of building systems blindly. The followingstudy is one such empirical examination. Setting and Data CollectionWe observed a telephone helpline group (HLG) atCyberCorp, a computer company headquartered inSilicon Valley. HLG answers human resource ques-tions for CyberCorp, primarily about benefits andpersonnel policies for the company’s thousands ofemployees. We chose a telephone hotline because it is fast-paced and information intensive. Its many repetitivecalls require routine action by the HLG agents whomust begin forming their answers within 45 to 60seconds while simultaneously listening to callerinformation and elaboration. The routine doesmean many answers come directly from the hotlinemember’s memory. However, much of the repetitionis in what way the types of calls are processed and inhow types of information are accessed. Repetition isan advantage for analysis because of the opportunityto examine things more than once. Thus we wereable to observe how the helpline agents use theresources available to them.The field study took place over a period of 18months. A variety of data collection methods wereused, including direct observation, video, semi-structured interviews, and social network analyses.We observed at least 300 calls, video-taped approxi-mately 60, chose 10 for analysis in a manner similarto [4]. We describe one video-taped call here involv-ing Joan,1an experienced agent, who has beenemployed by CyberCorp for five years and workedin the HLG for one year. Joan, like the other agents, works in a cubicle thatis open to a central corridor (Figure 1). Like manycontrol room settings, the cubicles are close enoughto easily hear the activities of other agents. Joan’schair faces away from the corridor between the cubi-cles. This arrangement is important for accomplish-ing her work.Agents use a number of software applications dis-played on two monitors. One system is the CAll60 January 2000/Vol. 43, No. 1 COMMUNICATIONS OF THE ACMSharedsystemsJoanFigure 1. Joan's workplace. The left-hand portion shows how the cubicles are related to one another, as well as the shared terminals.The right is a larger picture of Joan's office.1All participants and their individual attributes have been disguised