paid-media-paid-media-search-query-analyst

Pull the search term report first, then apply spend-weighted irrelevance scoring and n-gram analysis to flag zero-conversion and high-CPC low-value queries. The agent groups queries by intent and recommends negatives at account, campaign, or ad group level.

Start with n-gram frequency analysis to find recurring irrelevant modifiers, then apply a decision tree: if a query contains specific term combinations, add the negative at the appropriate tier. Use shared negative lists for account-wide exclusions and check for conflicts with active keywords.

Yes, when Google Ads MCP tools or API integrations are available in the environment, the agent can deploy negative keywords at campaign or shared list level without leaving the conversation. Without API access, it delivers the recommendations as a structured list for manual upload.

CPA increases after scaling are often caused by query drift, where broad match expansion pulls in lower-intent or irrelevant search terms. A search term audit isolates the drifting queries, quantifies their wasted spend, and restores efficiency with targeted negatives.

Yes, the agent interprets Performance Max search category insights and analyzes shopping search terms including product type, attribute, and brand queries. It identifies waste patterns even where full query-level data is limited.

Analysis quality depends on data volume and visibility; low-traffic queries and hidden Performance Max terms limit coverage. The agent requires actual report data before making recommendations and cannot infer patterns from accounts it cannot access.