Skip to main content

Scenario

Google Ads shows you who you compete against in every auction — impression share, overlap rate, outranking share, position above rate. But this data sits in a table nobody reads. You pull auction insights for your top campaigns, identify the biggest competitive threats, then send each competitor to Mave for deep research and battle card generation. The result is AI-generated battle cards backed by both auction data and web research.

Architecture

Code

Example Output

Error Handling

Display and Video campaigns don’t populate auction_insights. The report only covers Search and Shopping campaigns.
Auction insights report at the domain level, not per-ad. A competitor with multiple products under one domain shows as a single entry.
Generating 5 battle cards sequentially takes 30–60s. The code includes 1s delays. For 10+ competitors, parallelize with asyncio.gather (Python) or Promise.all (JS) with a concurrency limit.

All Google Ads jobs

View all 7 Google Ads integration jobs

Mave Agent

Full reference for POST /api/v1/mave/chat