Context Window
The maximum amount of text an AI model can consider at once when generating a response, a hard limit on how much retrieved or provided information can inform an answer.
What it means
Every LLM has a finite context window, measured in tokens, that caps how much text (the conversation, retrieved documents, system instructions) can be processed together. When a shopping query triggers retrieval of product data, that data competes for space within the same window as everything else the model is considering. Bloated, poorly structured product content wastes context window space that could otherwise hold more relevant, decision-useful facts.
Why it matters for Shopify
Shopify merchants benefit from concise, well-structured product data, since it uses less of an AI system's limited context window per query, leaving more room for the facts that actually influence a recommendation.