---
id: "quote-agent-mandate"
type: "quote"
source_timestamps: ["\\\"§ Understand the prompt", "not just the agent.\\\""]
tags: ["prompt-engineering", "consumer-behavior"]
related: ["concept-prompt-driven-optimization"]
speaker: "Authors"
speakers: ["Jafar Sabbah", "Oguz A. Acar"]
sources: ["geo"]
sourceVaultSlug: "hbr-seg-geo"
originDay: 3
articleStem: "hbr-tier2-06-ai-shopping-agents"
sourceUrl: "https://hbr.org/2026/05/research-traditional-marketing-doesnt-work-on-ai-shopping-agents"
sourceTitle: "Research: Traditional Marketing Doesn’t Work on AI Shopping Agents"
---
# Agents arrive with user prompts

> "An AI shopping agent does not arrive with its own preferences. It arrives with the user's prompt."
> — [[entity-jafar-sabbah|Jafar Sabbah]] & [[entity-oguz-a-acar|Oguz A. Acar]]

**Why it matters:** The rationale for [[concept-prompt-driven-optimization|prompt-driven optimization]] and the action item [[action-analyze-user-prompts|analyze common user prompt structures]]. Because the agent's behavior is entirely bounded by the user's instruction, the prompt — not the shopper's mood — becomes the object of optimization.

**Related:** [[concept-prompt-driven-optimization]] · [[action-analyze-user-prompts]]


## Related across articles
- [[concept-prompt-driven-optimization]]
- [[claim-query-determines-competitive-set]]
- [[claim-search-queries-are-need-based]]