Big deal and long tail: e‐journal usage and subscriptions

International audiencePurpose - This paper is aimed primarily at academic library managers and acquisition librarians. By analogy with Pareto's study of the relationship between clients and turnover, we will study subscriptions to e-journals and usage statistics. Our purpose is to evaluate the long tail of usage statistics and to compare it with the two modes of subscription, individually selected journals vs. packages (big deals). Design/methodology/approach - The paper exploits usage statistics and subscription data from a national usage study of an academic publisher. Data are from 2010. Findings - Usage statistics are partly shaped by the long tail effect. Individual subscriptions to journals are more selective than big deals, and tend towards a traditional retail model. Unlike subscriptions through packages, usage and individual subscriptions can be related by a similar inclination. But both types of subscriptions fail to predict the popularity of a journal in its usage. Research limitations/implications - The paper uses data from a national usage study and tries to identify global trends. Thus, it does not distinguish between customer categories, disciplines and activity domains. Practical implications - The paper considers the opportunity provided by big deal for acquisition policy. Ready-made big deals sometimes appear as an unbounded and excessive supply not suited to the users' true and sufficient needs, while selective acquisition policy cannot completely anticipate online usage behaviour. Originality/value - Only a few studies distinguish Pareto from long tail distributions in usage statistics, and there is little empirical evidence as to the impact of selected subscriptions vs. big deals on these statistics.Il s'agit d'une étude empirique sur le lien entre les abonnements aux revues électroniques et les statistiques d'usage, en comparant deux modes d'abonnement (titre par titre, "big deal") et en posant la question de la "longue traîne" des revues électroniques