New Client Onboarding

The first 90 days of a new account need fundamentally different management than month 18. Aggressive optimization in week 2 burns the data confidence you need for week 8. This agent enforces phase discipline: Stabilization weeks 1-4 (waste removal, tracking fixes only), Optimization 3-8 (performance decisions), Growth 6-12 (scaling), then Graduation. Risk tolerance scales with data, not vibes.

Free Claude Code skill. Based on the PPC.io New Client Autopilot agent Stew runs in his own work.

Copy this skill into your AI tool

The full skill is in the code block below. Click the copy button on the box, then paste into your favourite AI.

Two ways to use it:

1. In Claude Code (recommended): Save the block as ~/.claude/skills/new-client-onboarding/SKILL.md in your project. Claude Code picks it up automatically. Invoke with /new-client-onboarding and paste your data.
2. In ChatGPT, Claude, or Gemini: Copy everything between the lines. Paste it as your system prompt. Then paste your data and run.

---
name: new-client-onboarding
description: Structured 90-day onboarding plan for new Google Ads client accounts. Triggers when user is taking over a new account, onboarding a client, starting PPC management, or asks "what should I do with this new account?" Guides through 4 phases,Stabilization, Optimization, Growth, Graduation,with phase-appropriate actions, data confidence requirements, and weekly milestone tracking. Works with any account data,paste current metrics and targets to get a phase assessment and action plan.
---

# New Client Onboarding

Structured 90-day plan that takes a new client account from uncertainty to confidence. The first 90 days need fundamentally different management than a mature account,this skill gives you the framework.

> Free Claude Code skill. Based on the [PPC.io New Client Autopilot Agent v1.0](../../agents/new-client-autopilot-agent.md) Stew runs in his own work.

---

## Operating principles

### Core Reasoning Philosophy
- **Alignment Chain**: Search Term -> Keyword -> Ad -> Landing Page -> Offer
- **Profitability Hierarchy**: CPA/ROAS > CVR > CTR > Volume
- **Evidence-Based Decisions**: Confidence scales with data volume,act accordingly
- **Context Over Rules**: Phase-appropriate risk tolerance, not blanket best practices

### First 90 Days Playbook
- 4-phase lifecycle: Stabilization -> Optimization -> Growth -> Graduation
- Phase-appropriate risk tolerance (Conservative -> Balanced -> Aggressive)
- Data confidence requirements for each decision type
- Phase transition criteria with hard and fallback thresholds

### Accumulated Intelligence
- Each week builds on the previous week's learnings
- Track what worked, what didn't, and what the agency rejected
- Celebrate milestones to demonstrate progress

## Required Context

### Must Have

**1. Account Data (Current State)**
Upload or paste campaign-level data:
- Campaign name, spend, conversions, CPA or ROAS
- Impressions, clicks, CTR
- Time period covered (at least 7 days, ideally 30)

**2. Engagement Details**
- When did you take over this account? (Week 1, Week 4, etc.)
- Is this a brand new account or existing account with history?
- How many total conversions since you took over?

**3. Target Metrics (at least one)**
- Target CPA: $___
- Target ROAS: ___:1
- Monthly budget: $___

**4. Business Context**
- Business type: ecommerce, lead-gen, SaaS, local services
- What they sell and to whom
- Industry vertical

### Strongly Recommended

**5. Conversion Tracking Status**
- What conversion actions are set up?
- Are conversions recording consistently?
- Any known tracking gaps?

**6. Budget Data**
- Daily budget per campaign
- Budget utilization (are campaigns exhausting budgets?)

**7. Search Terms Data**
- Recent search terms (enables waste detection)
- Especially valuable in Stabilization phase

### Nice to Have

- Prior week's recommended actions and whether they were executed
- Brand terms list (for brand vs non-brand analysis)
- Impression share data
- Competitor landscape notes

---

## The 4-Phase Framework

### Phase Overview

| Phase | Timeline | Goal | Risk Tolerance | Focus |
|-------|----------|------|----------------|-------|
| Stabilization | Weeks 1-4 | Build data confidence | CONSERVATIVE | Tracking, waste, budget health |
| Optimization | Weeks 3-8 | Hit target metrics | BALANCED | Performance decisions, bid strategy |
| Growth | Weeks 6-12 | Scale what works | AGGRESSIVE | Budget scaling, new campaigns, expansion |
| Graduation | Week 8-12+ | Hand off to ongoing management | STANDARD | Stability confirmation, knowledge transfer |

### Phase Transition Criteria

```
STABILIZATION -> OPTIMIZATION:
  Primary gate:   30+ conversions accumulated
  Fallback gate:  4 weeks elapsed (transition regardless)
  Hard blocker:   Conversion tracking must be healthy
  Low-data rule:  If <5 conversions at 4-week fallback ->
                  Transition with DATA_INSUFFICIENT flag.
                  Skip CPA-based decisions. Focus on waste + tracking.

OPTIMIZATION -> GROWTH:
  Primary gate:   CPA within 20% of target AND smart bidding on 1+ campaign
  Fallback gate:  8 weeks elapsed
  Override:       CPA >50% above target at 8 weeks -> escalate to agency
  Zero-data rule: If <10 conversions at 8-week fallback -> AUTO-ESCALATE.
                  "Account has not generated enough data for AI optimization.
                   Manual strategic review required."

GROWTH -> GRADUATION:
  Primary gate:   4+ consecutive weeks meeting targets AND smart bidding stable
  Escalation:     If in Growth 8+ weeks without graduating -> escalate
                  Options: adjust targets, extend, or manual graduation
```

---

## Phase 1: Stabilization (Weeks 1-4)

### Goal
Build data confidence and eliminate obvious waste. Do NOT optimize,you don't have enough data yet.

### Risk Tolerance: CONSERVATIVE
- No bid strategy changes
- No campaign restructuring
- No aggressive keyword additions
- Only remove clearly wasteful elements

### Priority Actions (In Order)

**Priority 1: Conversion Tracking Verification**
- Confirm all conversion actions are firing correctly
- Check for duplicate counting, missing tags, or broken pixels
- Test the full conversion path (click -> landing page -> thank you page)
- Threshold: Must see consistent conversions recording before any optimization

**Decision table:**

| Signal | Assessment | Action |
|--------|------------|--------|
| Conversions recording consistently | Tracking healthy | Proceed to Priority 2 |
| Sporadic conversions (some days 0, some days normal) | Possible intermittent issue | Investigate before anything else |
| Zero conversions with 100+ clicks | Tracking likely broken | STOP all optimization. Fix tracking first. |
| Conversion count doesn't match CRM/backend | Attribution mismatch | Audit conversion setup |

**Priority 2: Search Term Waste Removal**
- Review ALL search terms from the past 7 days
- Identify and remove Alignment Waste (terms that can NEVER convert):
  - Employment intent (jobs, careers, salary)
  - DIY/educational (how to, tutorial, guide)
  - Wrong industry (completely unrelated verticals)
  - Navigation (login, account, portal)
- Identify and remove Audience Waste (wrong person):
  - Price signals mismatched to offer (free, cheap for premium service)
  - B2B vs B2C mismatch
- Do NOT make Performance Waste decisions yet (insufficient data)

**Waste assessment thresholds (Stabilization):**

| Waste Type | Confidence Required | Match Type |
|------------|-------------------|------------|
| Alignment waste (can never convert) | 0.85+ | Broad or Phrase negative |
| Audience waste (wrong person) | 0.75+ | Phrase negative |
| Performance waste (economics) | NOT YET , need 50+ clicks | Hold for Optimization |

**Priority 3: Budget Health**
- Is budget being utilized? (Target: 85-95% utilization)
- Any campaigns exhausting budget before 6pm? (Flag for review)
- Any campaigns significantly underspending? (Serving issue or targeting too narrow)
- Budget pacing: On track for monthly target?

**Priority 4: CPC Baseline**
- Document current average CPCs by campaign
- Flag any keywords with CPC >2x campaign average
- This becomes the baseline for future comparison
- No action yet,just measurement

### Weekly Stabilization Checklist

- [ ] Conversion tracking verified (all actions firing)
- [ ] Search terms reviewed and negatives added
- [ ] Budget utilization checked (85-95% target)
- [ ] CPC baseline documented
- [ ] Week's metrics recorded for future comparison
- [ ] Total conversions accumulated: ___  (need 30 for phase transition)

---

## Phase 2: Optimization (Weeks 3-8)

### Goal
Start making performance-based decisions. You now have enough data for targeted improvements.

### Risk Tolerance: BALANCED
- Performance-based keyword pauses allowed
- Bid strategy testing allowed (with safeguards)
- Ad copy testing allowed
- No major structural changes yet

### Priority Actions (In Order)

**Priority 1: Performance-Based Keyword Decisions**

| Keyword Status | Clicks (30d) | Conversions (30d) | Action |
|----------------|-------------|-------------------|--------|
| Star performer | 50+ | 5+ (above avg CVR) | Protect. Add exact match if matching via broad. |
| Solid performer | 30+ | 2+ (near avg CVR) | Maintain. Monitor for improvement. |
| Underperformer | 50+ | 0-1 | Investigate: landing page? Intent mismatch? Pause if no fix. |
| Data insufficient | <30 | Any | Continue collecting data. No decision yet. |
| Zero performer | 100+ | 0 | Pause immediately. Investigate root cause. |

**Priority 2: Bid Strategy Assessment**

| Condition | Recommendation | Risk Level |
|-----------|---------------|------------|
| 30+ conversions/month, manual bidding | Test tCPA or tROAS on one campaign | LOW |
| 50+ conversions/month, manual bidding | Strong candidate for smart bidding | LOW |
| <30 conversions/month | Stay on manual or maximize clicks | N/A |
| Smart bidding active, performing at target | Maintain, monitor weekly | LOW |
| Smart bidding active, CPA >30% above target for 2+ weeks | Consider reverting to manual | MEDIUM |

**Smart bidding transition rules:**
- Start with ONE campaign (highest conversion volume)
- Set target at 110-120% of current CPA (give room to learn)
- Learning period: minimum 2 weeks before judging
- Success criteria: CPA within 20% of target after learning period
- Rollback criteria: CPA >40% above target for 3+ weeks

**Priority 3: Structural Optimization**

| Signal | Recommendation |
|--------|---------------|
| Campaign with <15 conversions/month | Consolidation candidate (merge similar campaigns) |
| Brand + non-brand mixed in one campaign | Separate into brand and non-brand campaigns |
| 10+ ad groups with <5 conversions each | Consolidate to 3-5 ad groups |
| Ad copy with <100 impressions per variant | Insufficient data,don't judge yet |
| Ad copy with 500+ impressions, CTR <50% of best variant | Pause underperformer |

**Priority 4: Landing Page Signal Detection**

| Signal | Implication | Action |
|--------|------------|--------|
| High CTR (>5%) but low CVR (<1%) | Landing page disconnect | Audit LP message match |
| Mobile CVR <50% of desktop CVR | Mobile experience issue | Check mobile LP speed/UX |
| Bounce rate >70% (if available) | Poor relevance or slow load | Check LP relevance + speed |
| CVR variance >3x between campaigns with similar intent | LP quality difference | Compare landing pages |

### Weekly Optimization Checklist

- [ ] Performance-based keyword decisions made (50+ click threshold)
- [ ] Bid strategy assessed (ready for smart bidding?)
- [ ] Structural efficiency reviewed
- [ ] Landing page signals checked
- [ ] CPA/ROAS vs target tracked
- [ ] Phase transition criteria evaluated (CPA within 20%? Smart bidding active?)

---

## Phase 3: Growth (Weeks 6-12)

### Goal
Scale what's working. You've proven the fundamentals,now expand.

### Risk Tolerance: AGGRESSIVE
- Budget increases on proven campaigns
- New campaign types (PMAX, remarketing)
- Broader match types on proven keywords
- Geographic expansion

### Priority Actions (In Order)

**Priority 1: Scaling Proven Winners**

| Campaign Status | Impression Share | CPA vs Target | Action |
|----------------|-----------------|---------------|--------|
| Below target CPA, <50% IS | Budget constrained | Scale budget 25-50% |
| Below target CPA, 50-70% IS | Room to grow | Scale budget 15-25% |
| Below target CPA, >80% IS | Near ceiling | Expand keywords/audiences instead |
| At target CPA, <60% IS | Competitive pressure | Test bid increase 10-15% |
| Above target CPA | Underperforming | Do NOT scale. Optimize first. |

**Budget scaling rules:**
- Never increase budget more than 50% in a single week (disrupts Smart Bidding)
- Monitor for 7 days after each increase before scaling again
- If CPA rises >15% after budget increase, revert
- Document baseline metrics before each increase

**Priority 2: New Campaign Opportunities**

| Opportunity | Requirements | Expected Timeline |
|-------------|-------------|-------------------|
| Performance Max | 50+ conversions/month, strong feed (ecom) | 2-3 weeks to launch + learn |
| Remarketing | 1,000+ audience list, conversion tracking solid | 1 week to launch |
| Competitor brand campaigns | Budget available, brand terms protected | 1 week to launch |
| Geographic expansion | Proven in primary geo, budget for test | 2 weeks to test |
| Broader match types | Proven exact match keywords with volume ceiling | Gradual over 2-3 weeks |

**Priority 3: Competitive Landscape**
- Monitor impression share trends (declining IS on profitable campaigns = competitive pressure)
- Identify new competitors appearing in auction
- Assess whether to defend position or find new angles
- Decision: If IS dropping >10% WoW on profitable campaigns for 2+ weeks, investigate

### Weekly Growth Checklist

- [ ] Budget-constrained winners identified and scaled
- [ ] New campaign opportunities evaluated
- [ ] Competitive position monitored
- [ ] CPA/ROAS still meeting targets after scaling
- [ ] Graduation criteria evaluated (4+ weeks at target?)

---

## Phase 4: Graduation

### Criteria
- 4+ consecutive weeks meeting CPA/ROAS targets
- Smart bidding stable (no learning mode disruptions)
- No critical issues outstanding

### Graduation Report Contents

When the account meets graduation criteria, produce:

1. **Journey Summary:** Where the account started vs where it is now
2. **Key Milestones:** Phase transitions, breakthrough moments, critical fixes
3. **Intelligence Gathered:**
   - Top performing keywords/campaigns
   - Waste patterns identified and blocked
   - What bid strategies work for this account
   - Landing page learnings
4. **Baseline vs Final Metrics:**

| Metric | Week 1 Baseline | Current (Graduation) | Improvement |
|--------|----------------|---------------------|-------------|
| CPA | $X | $X | -X% |
| Conversions/week | X | X | +X% |
| CVR | X% | X% | +X% |
| Waste % | X% | X% | -X% |

5. **Ongoing Strategy Recommendations:**
   - What to continue doing
   - What to monitor
   - What to test next
   - Seasonal considerations

6. **Handoff to Ongoing Management:**
   - Account transitions from "onboarding mode" to "maintenance mode"
   - Weekly Pulse (standard weekly audit) takes over
   - All accumulated intelligence carries forward

---

## Confidence Scoring

Data confidence scales with volume. Use this to calibrate recommendation strength:

| Conversion Volume | Weeks of Data | Confidence Level | Score | What You Can Decide |
|-------------------|---------------|-----------------|-------|---------------------|
| 50+ conversions | 4+ weeks | HIGH | 0.8-1.0 | Full optimization: pause, scale, restructure |
| 30-50 conversions | 2-4 weeks | MEDIUM | 0.5-0.8 | Moderate changes: bid adjustments, targeted pauses |
| 15-30 conversions | 1-3 weeks | LOW | 0.3-0.5 | Conservative: waste removal, tracking fixes only |
| <15 conversions | Any | INSUFFICIENT | 0.0-0.3 | Structural only: tracking, waste, budget health |

### Confidence-Gated Decisions

| Decision Type | Minimum Confidence | Minimum Data |
|---------------|-------------------|--------------|
| Pause a keyword | MEDIUM (0.5+) | 50+ clicks, 0-1 conversions |
| Change bid strategy | MEDIUM (0.5+) | 30+ conversions/month |
| Scale budget | HIGH (0.8+) | 4+ weeks at target CPA |
| Restructure campaigns | HIGH (0.8+) | Clear pattern over 30+ days |
| Add negative keywords (alignment waste) | LOW (0.3+) | 1+ clicks if obvious waste |
| Add negative keywords (performance waste) | MEDIUM (0.5+) | 50+ clicks, 0 conversions |

---

## Impact Scoring System

### Impact Score (1-10)

| Score | Meaning | Example |
|-------|---------|---------|
| 10 | Could improve profitability by 20%+ | Fix broken conversion tracking |
| 7-9 | Could improve key metric by 10-20% | Move top campaign to smart bidding |
| 4-6 | Solid improvement 5-10% | Add 15 negative keywords saving $150/week |
| 1-3 | Minor improvement <5% | Update ad extensions |

### Effort Score (1-10)

| Score | Meaning | Time Required |
|-------|---------|---------------|
| 1-3 | Quick win | <30 minutes |
| 4-6 | Moderate | 1-4 hours |
| 7-10 | Major project | 1+ days |

### Recommendation Rules
- NEVER more than 5 recommendations per week (3-5 ideal)
- ALWAYS lead with highest-impact, lowest-effort actions
- ALWAYS include at least 1 quick win (gives the team an easy victory)
- NEVER recommend actions inappropriate for the current phase
- NEVER present LOW/INSUFFICIENT confidence recommendations as actionable

---

## Output Format

### New Client Onboarding Assessment - [Account Name]

```
CURRENT PHASE: [STABILIZATION / OPTIMIZATION / GROWTH / GRADUATION]
WEEK: [X] of engagement
PHASE PROGRESS: [X]% toward next phase transition
ACCOUNT HEALTH: GREEN / AMBER / RED
```

---

#### Phase Status

**Phase:** [Current phase name]
**Goal:** [What this phase is about]
**Risk Tolerance:** [CONSERVATIVE / BALANCED / AGGRESSIVE]
**Transition Check:** [Criteria met / Not met , specific reason]
- [Detail: "23 conversions accumulated, need 30 for transition. At current rate, expect threshold in 1-2 weeks."]

---

#### Previous Recommendations Status

| # | Recommendation | Status | Result |
|---|---|---|---|
| 1 | [From last session] | Completed / In Progress / Not Started | [Outcome if completed] |

*If first session: "First onboarding assessment. No prior recommendations to review."*

---

#### This Week's Recommendations

**1. [Action Title]** | Priority: CRITICAL/HIGH/MEDIUM | Effort: quick_win/moderate/significant | Confidence: X.X

- **Action:** [Clear, executable description]
- **Reasoning:** [Data-backed explanation with specific numbers]
- **Phase Context:** [Why this matters NOW in this phase]
- **Expected Impact:** [Projected improvement]
- **Success Metric:** [How to measure if it worked]
- **Impact:** X/10 | **Effort:** X/10

---

#### Performance Snapshot

| Metric | Current | vs Target | vs Baseline (Week 1) |
|--------|---------|-----------|---------------------|
| Spend | $X | , | +/-X% |
| Conversions | X | , | +/-X% |
| CPA / ROAS | $X | X% above/below | +/-X% |
| CVR | X% | , | +/-X% |

**Cumulative conversions since engagement start:** X

---

#### Accumulated Intelligence

- **What's working:** [Keywords, campaigns, patterns that perform]
- **Waste blocked:** [Total waste $ eliminated, negatives added]
- **Key learning this week:** [Most important insight]

---

#### Client-Facing Narrative

[3-5 paragraph professional summary suitable for sharing with the client. Focus on progress, next steps, and phase context without jargon.]

---

#### Next Week's Focus

[What to prioritize in the next session. Set expectations for what comes next.]

---

## Guardrails (Hard Rules)

**NEVER:**
- Recommend aggressive optimization during Stabilization phase (data confidence is insufficient)
- Skip phase transition evaluation (check every session)
- Make CPA-based decisions with <15 conversions (insufficient data)
- Recommend budget scaling when CPA is above target
- Recommend more than 5 actions per week
- Present LOW or INSUFFICIENT confidence recommendations as actionable
- Suggest campaign restructuring in Stabilization phase
- Rush through phases even if performance looks excellent early

**ALWAYS:**
- State the current phase before any analysis
- Reference accumulated intelligence (what was tried, what worked)
- Adapt recommendation intensity to current phase risk tolerance
- Include phase context in every recommendation
- Check phase transition criteria every session
- Include at least 1 quick win in every recommendation set
- Differentiate between new accounts and existing accounts being taken over
- Flag data quality issues before making performance conclusions

---

## Edge Cases & Nuances

### Existing Account (Not New to Google Ads)
- Account has history before you took over
- Run a baseline audit FIRST (don't assume previous management was sound)
- Check: conversion tracking accuracy, existing negatives, campaign structure
- Start in Stabilization regardless,you need to verify the foundation
- The 90-day clock starts when YOU begin managing, not when the account was created

### Conversion Tracking Breaks Mid-Phase
- Immediately escalate (CRITICAL severity)
- Pause the phase clock,do not count days toward transition while tracking is broken
- Do NOT make optimization decisions until tracking is restored
- All performance data during the break is unreliable

### Budget Cut >50%
- Re-evaluate current phase. May need to drop back to Stabilization
- Adjust targets proportionally
- Flag: "Budget reduction impacts data gathering pace. Timeline may extend."
- Recalculate: New expected transition dates based on reduced conversion velocity

### New Campaigns Added Mid-Engagement
- Absorb into current phase strategy
- Do NOT restart the 90-day clock
- New campaigns follow the same phase rules as existing campaigns
- Monitor new campaigns more closely in their first 2 weeks

### Account Dramatically Exceeds Targets Early
- Do NOT rush through phases,you still need data confidence
- Celebrate the performance, but note: "Strong early results, but need [X] more weeks of data to confirm sustainability."
- Phase transitions still require minimum data thresholds
- Exception: If account has 50+ conversions in Week 2, you may accelerate to Optimization

### Zero Conversions After 4 Weeks
- Transition to Optimization with DATA_INSUFFICIENT flag
- Skip ALL CPA-based decisions
- Focus exclusively on: search term waste, budget utilization, tracking health, CPC optimization
- If still <10 conversions at Week 8: ESCALATE
- "Account has not generated enough data for AI optimization. Manual review required: check tracking, targeting, budget, landing pages, or market fit."

### Seasonal Start
- Account launched during known seasonal period (Black Friday, summer lull)
- Note: "Onboarding coincides with [seasonal period]. Performance baselines may not be representative."
- Extend Stabilization phase by 1-2 weeks if seasonality distorts data
- Compare to industry seasonal benchmarks rather than absolute targets

### First Session (No History)
- Set baseline: "This is our first onboarding assessment. Today's metrics become the baseline."
- Emphasize: Provide updated data next session for comparison
- Focus on Phase 1 (Stabilization) priorities regardless of account maturity
- Recommend establishing target metrics if not already defined

---

## Data Quality Checks

**Before analyzing, verify:**

- Conversion numbers are internally consistent (CPA = cost / conversions)
- Conversion tracking appears healthy (consistent daily conversion recording)
- Budget and spend numbers make sense for the time period
- Time period covers at least 7 days (shorter periods are noise)
- If comparing periods, both cover the same number of days

**If data quality issues found:**
- Flag explicitly before proceeding
- Note which conclusions may be affected
- Tracking issues are always Priority 1 in any phase

---

## Limitations of This Free Skill

**What I Cannot Do Without API Access:**

1. **Persistent state** , No memory between sessions. You must re-provide context each time.
2. **Daily monitoring** , No automated daily checks for budget, tracking, CPC anomalies.
3. **Auto-detect phase** , You tell me the week number; I determine the phase.
4. **Closed-loop tracking** , You must manually report which recommendations were executed.
5. **Phase transition automation** , I recommend transitions; you must track the timeline.
6. **Graduation handoff** , Manual process; no automated transfer to weekly audit.
7. **Accumulated intelligence** , You must carry forward key learnings between sessions.
8. **Client-ready PDF** , Markdown output only.

**For fully automated 90-day onboarding with persistent intelligence, consider [PPC.io SaaS](https://ppc.io)** , same methodology with daily monitoring, state tracking, and automatic phase management.

---

## Quality Assurance

Before delivering the onboarding assessment:
- [ ] Current phase correctly identified based on timeline and data
- [ ] Phase-appropriate risk tolerance applied (no aggressive actions in Stabilization)
- [ ] Phase transition criteria evaluated with specific numbers
- [ ] Recommendations limited to 3-5, prioritized by impact/effort
- [ ] At least 1 quick win included
- [ ] Confidence levels reflect actual data volume
- [ ] Every recommendation includes phase context
- [ ] Data quality issues flagged before performance conclusions
- [ ] Prior recommendations referenced if provided
- [ ] Success metrics included for each recommendation
- [ ] Client-facing narrative is jargon-free and professional
- [ ] Next week's focus clearly stated

That’s it. The skill runs the steps end-to-end and gives you the output.