The Scripts Desk
Shabba's Placement Exclusions script: cleaning junk placements across PMax and Demand Gen
A free, open-source Google Ads Script that scans placements across Display, Performance Max and Demand Gen and semi-automates excluding the junk, from a maintainer still shipping in 2026.
Negative-keyword and search-term n-gram tools cover text and Shopping campaigns well, but they have nothing to say about where an ad actually showed. Placement waste is the newer problem: Performance Max and Demand Gen both spend across Google's automated inventory, and both will happily serve on games apps, low-quality sites and irrelevant YouTube channels if nobody is watching. Shabba's Placement Exclusions script is built for exactly that gap. It scans placements across Display, PMax and Demand Gen, then semi-automates the exclusion step, so you review a shortlist and approve it rather than trawling the placement report by hand.
Why it made the cut
Placement waste on the newer inventory types is a live 2025-2026 pain point precisely because PMax and Demand Gen hide the placement report a click or two further away than a standard Display campaign, and Google's own account-level exclusion lists are a blunt instrument. Shabba's script targets that specific gap rather than duplicating the keyword and search-term tools already on this shelf, and it is listed as new for 2026 on Shabba's own free scripts index, with sibling scripts on the same site carrying dated version history that shows the collection is actively maintained, not abandoned.
How to use it
- Open the source page below and copy the script.
- Paste it into Tools, Bulk actions, Scripts in Google Ads, and authorise it.
- Run a preview first: the script reports candidate placements, it does not exclude anything on its own.
- Review the shortlist and apply exclusions manually, or per the script's own configuration options, then schedule it to run on a cadence that suits your review rhythm.
Worth knowing
The script is fully free and open source under GPL-3.0, with the code hosted on GitHub, so you can read exactly what it does before you run it in a live account. We link to the source rather than reproducing the code here, so you always pull the current version and the credit stays with the person maintaining it.