Retrieval
Finding the relevant information a system will use to answer.
Technical Term
Full explanation
Retrieval is the process an AI system uses to find relevant information from available sources before generating a response.
Retrieval typically works by matching a query against an index of content, using a mix of keyword matching, semantic similarity through Embeddings, and source-quality signals to select which documents to pull forward for the system to reason over. What is not retrieved cannot influence the answer, no matter how accurate or well-written it is.
This makes retrieval a gate rather than a ranking. A brand does not need to be ranked first to be recognized; it needs to clear the retrieval step at all. Everything Aingworth measures downstream, Citation Share, Answer Share, presumes retrieval already happened.
Why this matters
It is the step where a brand is either in the running or not. Everything downstream, citation, description, recommendation, only happens to sources that were retrieved.
Example
An AI system answering "who offers same-day alteration services in Leeds" retrieves a shortlist of local business pages before generating a recommendation. A business with no page describing that specific service is never part of the shortlist, regardless of its reputation.
Where this sits
Retrieval belongs to Retrieval, Evidence & Consensus, one of the seven layers of the Aingworth Glossary. Information exists, then it is retrieved, then sources are selected, then they corroborate one another, then consensus emerges. This is where Aingworth's Consensus Over Rankings thesis lives.
Related terms
Ready to see where your brand stands?
Start with the free diagnostic to see which condition of the Recognition Triangle is leaking, or read the frameworks these terms belong to.