The Agentic Shortlist Economy: concept, mechanism and research framework

AI assistants and agents built on large language models (LLMs) narrow the set of options before a person chooses. This paper proposes the concept of the Agentic Shortlist Economy (ASE): a market regime in which a brand's access to the consumer's decision increasingly depends on its inclusion in the shortlist generated by the agent and on how the agent describes it. The contribution has four parts. First, a definition and a five-step mechanism, from the formulation of a need to the market effect, with the two paths through which the agent forms the list (model memory and real-time search) and the shift from human trust to delegated trust. Second, a delineation of the Agentic Shortlist Economy (ASE) from the literature on consideration sets, recommendation agents, algorithmic consumers, Generative Engine Optimization (GEO) and platform economics. Third, five theoretical propositions and eleven testable hypotheses. Fourth, a measurement instrument, the Shortlist Inclusion Rate (SIR), defined as the share of relevant runs in which a brand appears in the agent's shortlist, complemented by the Brand Recognition Rate (BRR) and an accuracy code for brand descriptions. The instrument comes with a reproducible protocol across five assistants and a programme of five studies that starts with an audit of the Republic of Moldova market in Romanian, Russian and English. Central hypothesis: competition between brands is extending from capturing human attention to being included, and correctly represented, in agents' recommendations. The Agentic Shortlist Economy (ASE) is a proposed concept that has not yet been empirically validated.