For [BRAND NAME], analyse the top 5 campaigns by spend over the last 30 days using BooleanMaths MCP.
ROAS DECOMPOSITION Pull campaign-level ads data. 
For each of the top 5 by spend, compute: CTR × CVR × AOV / CPI = ROAS (use BM pixel numbers for CVR and AOV, not platform-reported). 
Identify which lever is dragging ROAS for each campaign.

PRODUCTS Call get_channel_products per campaign. Identify top SKUs by quantity. 
Flag if accessories dominate (AOV drag) or a single hero product dominates (collab/influencer signal).

LANDING PAGES Pull get_web_analytics (top 30 pages) and get_web_analytics_by_source (groupBy: campaign). 
Match top products to their PDPs. Compute implied page CVR = BM orders / page sessions.

VERDICT PER CAMPAIGN High CTR + low CVR → LP problem Low CTR + low CVR → creative problem Low CTR + decent CVR → creative reach problem Good funnel + high CPI → bid/creative refresh Good funnel + low AOV → SKU mix issue

OUTPUT Render a widget: site-wide context, one card per campaign (levers + top products + LP data + insight + verdict tag), summary 2×2 grid. 
Follow with prioritized action list in text.