PPC & Ads · from a Blog post, Apr 13, 2026 · Added
Morning Amazon PPC Risk Summary with n8n for an Agency Client
Catch keywords and campaigns burning ad spend every morning instead of once a week
“ACOS dropped from 34% to 16% in 8 weeks”
Per source Reported by an agency about its own client work, not independently verified.
Copy-paste prompt
Every morning at [7 AM], pull yesterday's Sponsored Products, Brands and Display data. Flag: keywords with ACOS above [target]% and [10]+ clicks; search terms with spend over $[x] and zero orders in the last [14] days; campaigns whose spend is [50]% above their 7-day average; and anything well off its 30-day baseline. Send me a short plain-English summary of the top 3 things that need attention, with the numbers behind each. Don't change bids or budgets.
The source didn't publish its exact prompt. We wrote this starter so you can try the same workflow. Replace anything in [brackets].
What they did
A client came to Andrii Nikulin's agency (Ping4Win) after one forgotten broad-match keyword spent €2,800 in 11 days with zero conversions. The agency built an n8n-based system that runs every morning: it scans every active campaign across ad types, flags keywords above the target ACOS, finds search terms eating budget with zero conversions, spots campaigns whose spend spiked overnight, compares yesterday with the 7-day and 30-day baseline, and writes a plain-English summary: "3 things need your attention today."
How it went
He says the client now spends 20 minutes a week on PPC decisions instead of 3 hours, and ACOS dropped from 34% to 16% in 8 weeks, "not because we're smarter. Because we stopped missing things."
We only report what the source says. Results are self-reported unless noted.
Worth knowing
An agency post; the client isn't named, and it doesn't name the AI model or how the system reads the ads data, or say whether any change is made automatically. Same agency as the stockout-alert case, different job and post.