agentleFS
fourteenwmfourteenwm/ppc-ai-skills/portfolio-pacing-rules/SKILL.md

portfolio-pacing-rules

Skill72 starsChanged 4 months agoMITView on GitHub

Portfolio-specific pacing thresholds, budget tolerances, and brand caps. Auto-invoke when analyzing pacing variance, checking if accounts are on pace, investigating underspending/overspending, or discussing portfolio-specific rules (Portfolio A ±5%, Portfolio B ±8%, Portfolio C). Also loads when user mentions specific portfolios by name.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    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
  1. Distinctive:

    No shared budgets (need granular control for tight pacing)

    line 75
  2. Distinctive:

    No brand campaign cap

    line 91
  3. Distinctive:

    Only show ROAS if specifically testing that metric

    line 150
  4. Distinctive:

    Algorithm safety: Avoid shocking smart bidding with large changes

    line 182

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Portfolio Pacing Rules Skill

Purpose: Provides portfolio-specific pacing thresholds, budget tolerances, brand caps, and performance targets that auto-load when analyzing account performance.

Type: Domain knowledge skill (auto-invoked)


What This Skill Deliberately Does NOT Do

  • Pull the data. It defines the thresholds; the pacing numbers come from your own tracking sheet and the Google Ads API (see Data Sources Referenced).
  • Calculate pacing. Variance is auto-calculated in your Pacing Dashboard — this skill sets and interprets the tolerance bands, it doesn't recompute the number.
  • Size the budget change. It holds the conservative 5-10% ceiling; the actual increase math is budget-recommendation-calculator, which consumes the tolerance defined here.
  • Select or investigate accounts. portfolio-health-prioritization picks which accounts to act on and underspending-investigation diagnoses them; this skill provides the thresholds both rely on.
  • Execute changes. It is protocol-only — no mutations happen here.

Quick Reference: Pacing Thresholds by Portfolio

Portfolio A

  • Budget Pacing: ±5% monthly tolerance (95%-105%)
  • Brand Campaign Cap: 15% maximum of total account spend
  • Primary KPI: ROAS (Conversion Value / Cost)
  • Warning Zone: >±5% (investigate)
  • Critical Zone: >±10% (immediate action)

Portfolio B

  • Budget Pacing: ±8% monthly tolerance (92%-108%)
  • Brand Campaign Cap: None (no restriction)
  • Primary KPI: Cost per Conversion (testing ROAS on some accounts)
  • Warning Zone: >±8% (investigate)
  • Critical Zone: >±10% (immediate action)

Portfolio C

  • Budget Pacing: ±8% monthly tolerance (92%-108%)
  • Brand Campaign Cap: None (no restriction)
  • Primary KPI: Cost per Conversion
  • Warning Zone: >±8% (investigate)
  • Critical Zone: >±10% (immediate action)

Universal Pacing Rules (All Portfolios)

Critical Thresholds

  • Critical: >±15% variance = Immediate investigation required (KPI miss, client escalation)
  • High Priority: >±10% variance = Requires investigation
  • Warning: >±8% variance = Monitor closely
  • On Pace: Within portfolio tolerance = Acceptable

Month-Start Normalization (Days 1-5)

  • Early-month variance is expected (budget ramp-up period)
  • Reduce severity tier for pacing variance in first 5 days
  • Full threshold rules apply after Day 5
  • Note: "Early-month variance expected to normalize by Day 7"

Portfolio Constraints

Portfolio A Accounts

Strategic Context:

  • Budget pacing is non-negotiable (strict client KPI)
  • ROAS optimization happens WITHIN pacing constraint
  • Brand campaigns capped at 15% despite high efficiency (30-40x ROAS)
  • AI campaigns (Pmax/DGen) are workhorses (60-70% of spend)
  • Budget reallocation > budget cuts (maintain spend, improve efficiency)
  • No shared budgets (need granular control for tight pacing)

Account Examples:

  • Acme Plumbing (CID: [CUSTOMER_ID])
  • Best HVAC (CID: [CUSTOMER_ID])
  • City Dental (CID: [CUSTOMER_ID])
  • Account D (multiple accounts)

When to flag:

  • Pacing variance >±5%
  • Brand spend >15% of total
  • Any variance >±10% with <1 week left in month

Portfolio B Accounts

Strategic Context:

  • More relaxed pacing tolerance (±8% vs Portfolio A's ±5%)
  • No brand campaign cap
  • Primary metric: Cost per Conversion (some testing ROAS)
  • Shared budgets common (GEO + Bedroom campaigns)
  • Budget flexibility allows for performance optimization

Account Structure:

  • Original "Core 4": Brand, GEO, Bedroom, GDN Remarketing
  • Current: Added Pmax/Demand Gen, paused GDN Remarketing
  • Shared budgets: GEO + Bedroom campaigns use shared budget

When to flag:

  • Pacing variance >±8%
  • Any variance >±10% with <1 week left in month
  • Shared budget imbalance (one campaign taking >80%)

Portfolio C

Strategic Context:

  • Mid-tier accounts (single metro area)
  • Similar constraints to Portfolio B (±8% pacing tolerance)
  • 3 accounts total

Accounts:

  • Portfolio C - Multi-region (CID: [CUSTOMER_ID])
  • ProClean (CID: [CUSTOMER_ID])
  • Quick Fix (CID: [CUSTOMER_ID])

When to flag:

  • Pacing variance >±8%
  • Any variance >±10% with <1 week left in month

Performance Targets by Portfolio

Portfolio A (ROAS Focus)

Campaign ROAS Targets:

  • Brand Campaigns: 25-40x ROAS (elite, but capped at 15% spend)
  • Performance Max: 15-20x ROAS (workhorse, scalable)
  • Demand Gen: 15-20x ROAS (efficient, visual/discovery)
  • YouTube/Video: 10-15x ROAS (supporting role)
  • GEO/Keyword Campaigns: 5-10x ROAS (acceptable minimum)
  • Underperformers: <5x ROAS (candidates for cuts/optimization)
  • Money Losers: <1x ROAS (immediate action required)

Format for Portfolio A reports:

  • Show ROAS (not "conversion value per cost")
  • Example: "16.9x ROAS" (not "$16.90 per $1 spent")

Portfolio B (CPA Focus)

Cost per Conversion Targets:

  • Varies by property (based on lease value and market)
  • Primary metric: Target CPA (not ROAS)
  • Testing ROAS on select accounts as alternative metric
  • Campaign hierarchy: Brand (lowest CPA) → Pmax → GEO/Bedroom → Demand Gen

Format for Portfolio B reports:

  • Show Cost per Conversion (e.g., "$125 CPA")
  • Only show ROAS if specifically testing that metric

When to Use This Skill

Auto-Invoked When:

  • Checking if account is on pace
  • Analyzing pacing variance
  • Investigating underspending or overspending
  • Evaluating campaign performance vs. targets
  • User mentions "Portfolio A", "Portfolio B", or "Portfolio C"
  • User asks "what's the pacing threshold for [portfolio]"
  • Creating performance reports or summaries

Data Sources Referenced:

  • Your own budget/pacing tracking sheet — one pacing dashboard tab per portfolio (the Pacing Calculation section below shows what it needs to contain)
  • Google Ads API: Campaign performance data
  • Your own portfolio context files — strategic context and constraints

Budget Management Context (Smart Bidding)

Primary Control Lever: Monthly Budget

  • YOU CANNOT MANUALLY ADJUST BIDS with smart bidding (Max Conversions, Max Conversion Value)
  • Monthly budget is the primary lever for controlling spend
  • Budget increases are conservative: 5-10% monthly increases (not aggressive jumps)

Budget Philosophy Hierarchy:

  1. #1 Priority: Hit monthly spend targets (pacing within tolerance)
  2. #2 Priority: Maintain acceptable CPA/ROAS goals
  3. Algorithm safety: Avoid shocking smart bidding with large changes

Why Conservative Increases (5-10%)?

  • Smart bidding algorithms need 3-7 days to ramp up after budget changes
  • Large budget jumps risk CPA spikes during adjustment period
  • Multiple small increases > one large increase (safer, more controllable)
  • Client trust requires predictable, controlled spend changes

Pacing Calculation (Reference)

Data Source: Pacing Dashboard sheet (auto-calculated daily)

Formula:

Pacing Variance = % Through Month - % Spent

Where:
- % Through Month = (Current Day ÷ Days in Month) × 100
- % Spent = (MTD Spend ÷ Monthly Budget) × 100

Example:

  • Day 21 of 31 → % Through Month = 67.74%
  • MTD Spend: $513 / Monthly Budget: $1,000 → % Spent = 51.30%
  • Pacing Variance: 67.74% - 51.30% = +16.44% (UNDERSPENDING)

Interpretation:

  • Positive variance (+X%): UNDERSPENDING (spent less than expected by now)
  • Negative variance (-X%): OVERSPENDING (spent more than expected by now)
  • Zero variance (0%): PERFECT PACING (spent exactly as expected)

IMPORTANT: This calculation is already done in Pacing Dashboard sheet (Column E). You don't need to recalculate it manually - just reference the sheet value.


When to Load Other Skills

SkillWhen
budget-recommendation-calculatorWhen a pacing miss becomes a budget change — it sizes the increase within the tolerance defined here (it consumes this policy, it doesn't set it)
portfolio-health-prioritizationWhen deciding which accounts to look at first — its tiers are built on these ±5%/±8% thresholds
underspending-investigationWhen an account is outside tolerance and needs diagnosis — it uses these thresholds as its Step-1 gate
budget-guardian and shared-budget-updaterFor spend-pace variance context on a book — both reference this policy for what "normal" looks like

External pieces you supply yourself: your own portfolio context files (names, pacing rules, client KPIs), your accounts map (account name → CID → portfolio), and your dashboard/sheet config (tab schemas for the pacing data) — all referenced throughout this skill, all replaceable with your own equivalents.


Created: 2025-10-28 Last Updated: 2025-11-01 Status: Active

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.