Command
Use list_projects({}) to confirm my supplied project host is visible. Use my chosen days value of 7, 30, or 90 for every time-windowed call. Read get_overview({host, days}) and get_brand_insights({host, days}). Compare them with my saved pre-campaign export, aligning metrics, engine coverage, and window lengths. Use list_prompts({host, limit: 100}) and get_prompt_runs({host, prompt_id, days}) to inspect recurring claims. Identify changes supported by comparable observations and mark unmatched samples. Provide engine-level detail only where returned. Explain possible confounders and draft a campaign review; never request custom overview dates or attribute changes solely to the campaign. Return evidence links, coverage limits, and explicit unknowns; never infer unsupported scores.What you need
- A visible host, a 7, 30, or 90 day window, and campaign context.
- A saved pre-campaign export with matching metrics and engine coverage.
Compare like with like
A before-and-after review needs a saved baseline because the overview accepts fixed rolling windows. Align measurement definitions and engine coverage first, then identify which differences reflect comparable observations and which remain uncertain because the available samples changed between reviews.
Explain shifts without claiming causality
Combine observed metric differences with examples of changing brand language. Record campaign timing alongside other plausible influences, and deliver a review that distinguishes evidence from interpretation so decision makers can evaluate the campaign without treating correlation as proven impact.