Meta Ads Scale Campaign

[Didoo AI] Guides Meta Ads campaign scaling — increases budget, expands audiences, replicates to new geos. Use when a testing campaign has exited Learning Ph...

Security Scan

Instructions and requested credentials align with a Meta Ads scaling helper; nothing in the skill asks for unrelated secrets or to install code, but the skill assumes a Meta access token whose scope is not enforced and it references other skills not declared as dependencies.

Purpose & Capability

Name/description (scaling Meta Ads campaigns) matches the required environment variables (META_ACCESS_TOKEN, META_AD_ACCOUNT_ID) and the runbook-style guidance. The requested items are what you'd expect for querying Meta Ads data.

Instruction Scope

SKILL.md is an instruction-only runbook focused on human/agent actions in Ads Manager and high-level API usage. It does not attempt to read unrelated files or system state. It references other skills (meta-ads-analysis, meta-ads-recommendation, meta-ads-publisher, meta-ads-healthcheck) but those dependencies are not declared in metadata — this is a scope/coordination note rather than a direct security issue.

Install Mechanism

No install spec and no code files — lowest risk for disk write or arbitrary code execution.

Credentials

Only two environment variables are required, which is proportional. However the skill assumes a META_ACCESS_TOKEN with ads_read (read-only) scope but does not enforce or verify the token's actual scopes; if a user supplies a token with broader scopes (e.g., ads_management) it could be used to modify assets. The skill's instructions mostly describe manual or other-skill actions rather than automated writes, but the risk depends on the token you provide.

Persistence & Privilege

always is false, no install, no requests to modify other skills or system-wide settings. The skill does not request permanent presence or elevated platform privileges.

Assessment

This is an instruction-only Meta Ads scaling runbook and appears coherent with its stated purpose. Before using/installing: (1) Provide the minimum-scoped access token possible — prefer a token/app with ads_read-only if you only want read access; avoid providing long-lived tokens with ads_management unless necessary. (2) Verify the token scope yourself (Meta App settings) because the skill doesn't validate scopes. (3) Note the skill references other skills (analysis, publisher, healthcheck) that are not declared — ensure those tools exist and are trustworthy if you expect automated actions. (4) Because the skill contains action recommendations (budget changes, duplication), test any automated flows in a non-production account first and review paused duplicates before activation. (5) If you need the agent to actually perform changes via API, prefer issuing credentials limited to a dedicated ad account and rotate credentials after testing.

Required Credentials

Credential	Where to Get	Used For	OAuth Scope
META_ACCESS_TOKEN	Meta Developer Console → Graph API Explorer → Generate Token	Fetching campaign status and insights	ads_read (read-only)
META_AD_ACCOUNT_ID	Ads Manager URL: adsmanager.facebook.com/act_XXXXXXXXX	Identifying which account to query	—

When to Use

Loaded when a testing campaign has proven results and you want to scale it — increase spend, expand audience, or replicate to new geos. Not for campaigns still in testing phase.

---

Prerequisites

Before scaling, confirm:

1. Testing phase is complete — campaign has generated ~50+ results in the past 7 days and exited Learning Phase (not just 50+ lifetime results)
2. Cost per result is at or below target CPL/CPA
3. Data is stable — metrics are consistent across at least 3–5 days

If any of the above are not confirmed, do not scale — return to meta-ads-analysis + meta-ads-recommendation to fix problems first.

---

Step 1: Confirm Scale Readiness

Check the following signals:

Must have:

• Learning Phase has exited — need ~50+ results in the past 7 days (per adset for ABO, or per campaign for CBO). A campaign with 50+ lifetime results but only 2 results this week has NOT exited Learning Phase.
• CPL/CPA at or below target (or acceptable for the business)
• Delivery is consistent — not frequently limited by budget or audience

Should have:

• Frequency below 3 (if above, scale will hit fatigue faster)
• CTR stable and acceptable (not declining trend)
• LPV rate healthy (> 70% for e-commerce, > 50% for lead gen)

If these signals are not confirmed, return to meta-ads-analysis + meta-ads-recommendation to fix problems first.

---

Step 2: Define Scale Strategy

Strategy A: Increase Budget on Winner Campaign

When: CPL is stable and below target, delivery is near cap

• Increase budget by max 20% per adjustment
• Wait 2–3 days between adjustments to let Meta re-optimize
• If CPL starts rising after budget increase → stop, return to optimization

Strategy B: Replicate to New Audience

When: Current audience is getting fatigued (frequency > 3) or you want to test new segments

• Duplicate the winning campaign (see meta-ads-publisher → Duplicate / Clone Campaign)
• Modify one dimension: new audience, new geo, or new placement
• Keep the same winning creative — creative is proven, not the variable

Strategy C: Replicate to New Geo

When: Campaign is working in one market and you want to expand

• Duplicate campaign with same structure
• Change only the geo targeting
• Keep same audience parameters where possible
• New geo = new audience data → expect new learning phase

Strategy D: Expand Audience Within Campaign

When: Targeting is too narrow and limiting volume

• Add new interest layers or lookalike audiences
• Do NOT broad-target — define the expansion clearly
• Monitor closely after expansion — may change CPL

---

Step 3: Execute Scaling in Meta Ads Manager

For budget increases:

1. Go to the ad set in Meta Ads Manager
2. Change daily budget — increase by 20% max
3. Wait 2–3 days before next adjustment
4. Monitor CPL closely — if it rises > 15%, pause the increase

For campaign duplication:

1. Use meta-ads-publisher → Duplicate / Clone Campaign
2. Create PAUSED, review before activating
3. Confirm targeting/budget/geos are correct for new context
4. Activate and monitor

---

Step 4: Monitor Post-Scale

Post-scale, run meta-ads-healthcheck or meta-ads-analysis daily for 3–5 days:

Green signals:

• CPL stable or improving
• Frequency still below 3
• Spend increasing proportionally to budget increase
• Results volume growing

Red signals — stop and reassess:

• CPL rising > 15% after budget increase
• Frequency > 3 and climbing
• Spend not increasing despite budget increase (delivery problem)
• Learning phase re-triggered

---

Constraints

• Never scale a campaign still in Learning phase — wait for it to complete
• Never scale a campaign with CPL significantly above target — fix the problem first
• Budget increases: max 20% per adjustment, 2–3 days between changes
• When scaling to new geos: expect new learning phase — don't judge too early
• If 2+ scaling attempts fail: return to meta-ads-recommendation for structural diagnosis

---

Scale Failure — Quick Reference

Symptom	Most Likely Cause	Next Action
CPL rises > 15% after budget increase	Audience is saturated	Pause increase; rotate new creative or expand audience
Budget increased but spend stays flat	Delivery constrained — audience too narrow or bid too low	Check delivery preview; widen targeting or raise bid
CPL stable but frequency jumps above 4	Audience fatigue	Rotate new creative immediately; expand to lookalike
Campaign re-enters Learning Phase after changes	Too many simultaneous changes	Pause; wait for re-stabilization; reduce changes next time
All metrics flat — no response to budget increase	Bid ceiling reached	Switch to lowest cost bidding; or campaign has hit its natural ceiling
Performance drops day after budget increase	Over-delivered early in day (pacing issue)	Check 3-day average, not single-day
CPL acceptable but volume too low to scale	Audience too narrow	Replicate to new audience (Strategy B), don't just increase budget

When to abandon a scaling attempt:

• CPL rises > 30% despite 2 budget adjustments
• Delivery becomes erratic / spend unpredictable
• Frequency reaches 5+ across all adsets
• 2+ consecutive days with no spend increase despite budget increase

When to keep trying:

• CPL rises < 15% and stabilizes — this is normal as you scale
• Volume is growing proportionally to budget
• Frequency stays below 4