Direct recommendation queries evaluate whether an AI model actively endorses a brand by prompting it in a binary yes-or-no format across repeated samples to compute a statistical Query Relevance Score.

flowchart TD
    Prompt["Input Prompt: Recommend Brand for Entity?"] --> Sample1["LLM Pass 1"]
    Prompt --> Sample2["LLM Pass 2"]
    Prompt --> SampleN["LLM Pass N"]
    Sample1 --> Aggregator["Tally Yes vs No"]
    Sample2 --> Aggregator
    SampleN --> Aggregator
    Aggregator --> Output["Calculate QRS Score %"]