AI-powered CRM capability model: Advancing marketing ambidexterity, profitability and competitive performance

Abstract

The rapid adoption of Artificial Intelligence (AI)–powered Customer Relationship Management (CRM) systems has exposed a critical gap: despite substantial investment, many organisations fail to derive meaningful business value from these technologies. Recent surveys show that while AI is a strategic priority for executives, only a fraction report significant returns, with adoption challenges particularly acute in customer-facing functions. This study addresses this gap by conceptualising and empirically examining AI-powered CRM as a higher-order organisational capability. Drawing on the microfoundations of dynamic capability theory, we adopt a three-stage research design. First, a systematic scoping review and in-depth interviews with industry experts identify the core dimensions and subdimensions of AI-powered CRM capability. Second, we operationalise and validate these dimensions within a nomological network. Third, a survey of 205 banking employees in Australia tests the influence of AI-powered CRM capability on marketing ambidexterity and, in turn, on organisational outcomes. The quantitative analysis confirms that AI-powered CRM capabilities positively shape marketing ambidexterity, which subsequently enhances profitability and competitive advantage. Theoretically, the findings advance CRM research by introducing a microfoundational capability model that integrates data management, multi-channel integration, and service offerings. Practically, the study provides actionable guidance for managers seeking to close the “value realisation gap” by cultivating AI-powered CRM systems as dynamic capabilities that balance exploration and exploitation in volatile markets.