From Inefficiency to Opportunity: Analysing How Generative AI Can Support Operational Workflows in a Media Agency

Sammanfattning

The rapid development of digital technologies and artificial intelligence (AI) is reshaping operational workflows in knowledge-intensive service organizations. Media agencies in particular face increasing pressure to improve efficiency across complex, multi-phase campaign processes characterized by manual task overload, system fragmentation, coordination overhead, and inadequate knowledge management. Despite growing interest in AI adoption, limited research addresses how generative AI can be systematically integrated into the specific operational workflows of media agencies.The purpose of this thesis is to examine and analyse how generative AI can support the improvement of operational workflows within media agency operations. The study is guided by three research questions: what are the current operational inefficiencies in the campaign workflow; how can generative AI support the analysis and improvement of these workflows; and what conditions are required to enable the integration of AI-supported tools.The study is conducted as a single in-depth case study at a large Swedish media agency, referred to throughout as the Agency. An abductive research approach is applied, combining semi-structured interviews with activation managers, client managers, and specialists, direct observation of operational work, a digital survey distributed to 35 employees, and analysis of internal process documents. Data was analysed using thematic analysis.The empirical findings reveal four interconnected structural categories of operational inefficiency: manual task overload, system fragmentation, coordination overhead, and inadequate knowledge management. The brief intake phase emerges as the most consequential root cause, generating downstream inefficiency across the entire campaign cycle. In response to these findings, the thesis develops the Artificial Intelligence Capability Scoring System (AICSS), a proposed six-pillar analytical framework for identifying, evaluating, and prioritizing AI integration opportunities in operational workflows.The analysis identifies three Tier 1 AI opportunities, AI-assisted brief structuring, automated Planner card generation, and automated report generation, as the highest-priority interventions based on their combination of high value, strong AI fit, and low implementation risk. The conditions assessment indicates that the primary barriers to AI integration appear to be organizational and process-related rather than narrowly technical, with standardization, AI literacy, clear ownership, and a pilot-first approach identified as critical prerequisites.The study concludes that generative AI holds genuine potential for improving operational workflows in media agency contexts, but that realizing this potential requires a structured, evidence-based approach grounded in an understanding of the operational environment and careful sequencing of interventions. The AICSS framework contributes a replicable analytical instrument for this purpose, though its generalizability beyond this single case should be treated as a proposition for future research rather than a demonstrated finding.