google-ads-pmax-scaling
Skill15 starsChanged 38 days agoMITView on GitHub
Use when scaling Google Ads Performance Max campaigns: budget scaling rules, pause triggers, location/audience/asset expansion. Loads PMax evaluation first.
In this repository61 other files it gives its agents.
- access-and-permissionsSkill
- accounts-and-connectionsSkill
- ad-brief-generatorSkill
- affiliate-marketingSkill
- analyse-creativeSkill
- automate-with-agentsSkill
- brief-creativeSkill
- build-reportsSkill
- clean-wasted-spendSkill
- competitor-ad-intelligenceSkill
- creative-psychology-hooksSkill
- creative-researchSkill
- diagnose-performanceSkill
- email-marketingSkill
- ga4-source-of-truthSkill
- get-startedSkill
- gomarble-skills-creative-strategistsSkill
- gomarble-skills-ecommerce-brandsSkill
- gomarble-skills-google-ads-auditSkill
- gomarble-skills-meta-ads-auditSkill
- gomarble-skills-saas-companiesSkill
- google-ads-create-master-skillSkill
- google-ads-depth-of-analysisSkill
- google-ads-guardrailsSkill
- google-ads-keywordplannerSkill
- google-ads-pmax-evaluationSkill
- google-ads-search-analysisSkill
- google-ads-search-executionSkill
- google-ads-shoppingSkill
- google-ads-tool-fundamentalsSkill
- google-impression-shareSkill
- google-pmax-pulseSkill
- google-search-auditSkill
- google-search-term-auditSkill
- launch-campaignsSkill
- linkedin-adsSkill
- manage-campaignsSkill
- meta-ads-auditSkill
- meta-agent-operationsSkill
- meta-create-master-skillSkill
- meta-creative-analysisSkill
- meta-creative-fatigue-detectionSkill
- meta-creative-strategySkill
- meta-custom-event-interpretationSkill
- meta-daily-optimizationSkill
- meta-depth-of-analysisSkill
- meta-guardrailsSkill
- meta-performance-analysisSkill
- meta-tool-fundamentalsSkill
- microsoft-adsSkill
- organic-socialSkill
- own-creative-diagnosisSkill
- research-competitorsSkill
- search-console-master-skillSkill
- shift-budgetSkill
- shopify-order-disciplineSkill
- tiktok-adsSkill
- tiktok-create-master-skillSkill
- winning-ads-engineSkill
- winning-ads-orchestratorSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
12 rules. About 0 are common, 12 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈12
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 14
ALL must be true before scaling:
- Distinctive: line 33
Hard cap: Do not exceed 2–3× initial monthly budget until 90+ days of consistent profitability.
- Distinctive: line 41
Only after core locations profitable for 30+ days
- Distinctive: line 43
Create separate asset groups for new locations if targeting differs
- Distinctive: line 50
Start with Observation mode for new audience signals
- Distinctive: line 53
Do NOT force audience preferences without data
- Distinctive: line 73
Never scale before 14 days of profitability
- Distinctive: line 74
Never increase budget > 50% in one step
- Distinctive: line 75
Never scale beyond 2–3× initial budget before 90 days profitable
- Distinctive: line 76
Never continue scaling if CPA > 1.2× target OR ROAS < 0.8× target
- Distinctive: line 77
Never remove BEST or GOOD performing assets
- Distinctive: line 78
Never force device/location exclusions without checking for technical issues first
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Google Ads PMax — Scaling & Expansion
Rules for scaling PMax campaigns that have passed evaluation. Only load after confirming profitability via google-ads-pmax-evaluation.
Scaling Prerequisites
ALL must be true before scaling:
- Campaign profitable for 14–30 consecutive days
- ROAS ≥ target (default 300% if not specified)
- CPA ≤ target
- Performance trends stable or improving WoW
Budget Scaling
Increase daily budget by 20–50% every 14–30 days.
| Scenario | Action |
| Profitable 14+ days, trends stable | Increase 20–50% |
| Post-scaling CPA > 1.2× target | PAUSE scaling, optimize assets/negatives, reassess after 30 days |
| Post-scaling ROAS < 0.8× target | PAUSE scaling, optimize, reassess after 30 days |
| Consistently profitable 90+ days | Can exceed 2–3× initial budget cap |
Hard cap: Do not exceed 2–3× initial monthly budget until 90+ days of consistent profitability.
Tool: google_ads_propose_update_campaigns with the new daily budget_micros in account currency (not micros). Dry-run first; apply only after the user approves.
Location Expansion
- Only after core locations profitable for 30+ days
- Expand to similar/adjacent markets
- Create separate asset groups for new locations if targeting differs
- Monitor new locations independently for 30 days before further expansion
Audience Expansion
- Start with Observation mode for new audience signals
- Monitor performance for 14+ days
- Switch to Targeting mode only if performance holds
- Do NOT force audience preferences without data
Asset Expansion
Based on BEST-performing assets from evaluation:
| Asset Type | Expansion Rule |
| Images | Add 5–10 total, similar style to BEST. Different angles/contexts. |
| Headlines | Create variations of BEST. Same theme, different wording. Test different CTAs. |
| Descriptions | Variations of BEST. Same value props, different framing. |
| Videos | Test 30-second formats if shorter videos are BEST. |
| New asset groups | Create for high-performing audience segments identified in Insights. Use targeted messaging. |
Prohibitions
- Never scale before 14 days of profitability
- Never increase budget > 50% in one step
- Never scale beyond 2–3× initial budget before 90 days profitable
- Never continue scaling if CPA > 1.2× target OR ROAS < 0.8× target
- Never remove BEST or GOOD performing assets
- Never force device/location exclusions without checking for technical issues first
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

