# Scaling Plan

> Create scaling strategy for winning campaigns including budget increments, multi-account distribution, and day-parting schedules. Use when you have proven winners and need a structured plan to increase spend while maintaining performance.

- Skill: `majiayu000/scaling-plan` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds add majiayu000/scaling-plan`
- Raw SKILL.md: https://api.skillmd.com/api/skills/majiayu000/scaling-plan/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Finance & Business
- Author: majiayu000 (https://skillmd.com/u/majiayu000)
- Updated: 2026-09-09
- Page: https://skillmd.com/skills/majiayu000/scaling-plan

---


# Scaling Plan

Create structured scaling roadmap for winning campaigns.

## Process

### Step 1: Validate Winner Criteria

**Confirm before scaling:**
- Winner doesn't lose more than 1x in 3 days (Jason K)
- Winner doesn't lose more than 2x in 7 days
- Consistent performance, not just good days
- CPA is below target with margin for variance
- Volume exists (not tapped out)

**Warning Signs to Address First:**
- Inconsistent daily performance
- Declining trend over time
- Audience saturation signals
- Creative fatigue indicators

### Step 2: Plan Budget Increment Strategy

**Conservative Approach (Recommended):**
- 20% increases every 2-3 days
- Wait for performance to stabilize
- Reduce if CPA rises >10%

**Aggressive Approach (Hot Offers):**
- Larger jumps when metrics are strong
- Risk of breaking campaigns
- Better to spread across accounts

**CBO Budget Limits:**
- $10K per CBO max (Casto)
- Beyond this, campaigns get unstable
- Better to replicate than push single CBO

### Step 3: Map Multi-Account Distribution

**Why Multiple Accounts:**
- De-risk against account issues
- Avoid single CBO instability
- Test different optimization approaches
- Parallel scaling

**Account Strategy:**
```
Account 1: Original testing + initial scale
Account 2: Replicate winners
Account 3: Replicate winners
Account 4: Backup/overflow
Account 5: New angle testing
```

**Replication Process:**
1. Export winning creatives
2. Replicate campaign structure
3. Start at moderate budget
4. Scale independently

### Step 4: Set Day-Parting Schedule (VSL)

**Recommended Schedule (Jason K):**
- Start: 4:00 AM Eastern
- In auction by: 6:00 AM Eastern
- Turn off: 1:00 PM Eastern
- Weekend exception: Run longer (until 8PM Pacific)

**Why Day-Part:**
- Afternoon dip in VSL conversions
- Team stops watching dead campaigns
- Focus resources on productive hours
- Weekend performance different

### Step 5: Output Scaling Roadmap

```
## SCALING PLAN: [Campaign/Offer Name]

### CURRENT STATE
- Daily spend: $[X]
- CPA: $[X] (Target: $[X])
- Account(s): [List]
- Top creatives: [List]

### VALIDATION CHECKLIST
- [ ] Winner criteria met (1 loss/3 days, 2 loss/7 days)
- [ ] Consistent 5+ days performance
- [ ] CPA has room for variance
- [ ] Volume potential confirmed
- [ ] No fatigue signals

---

### WEEK 1: INITIAL SCALE

**Day 1-2:**
- Account 1: Increase CBO from $X to $X (+20%)
- Monitor: CPA, volume, stability

**Day 3-4:**
- If stable: Increase to $X (+20%)
- If unstable: Hold or reduce 10%
- Launch Account 2 with replicated campaign at $X

**Day 5-7:**
- Account 1: Push to $X if stable
- Account 2: Scale to $X
- Launch Account 3 at $X

---

### WEEK 2: AGGRESSIVE SCALE

**Target Daily Spend:** $[X]

| Account | Mon | Tue | Wed | Thu | Fri | Sat | Sun |
|---------|-----|-----|-----|-----|-----|-----|-----|
| Acc 1   | $X  | $X  | $X  | $X  | $X  | $X  | $X  |
| Acc 2   | $X  | $X  | $X  | $X  | $X  | $X  | $X  |
| Acc 3   | $X  | $X  | $X  | $X  | $X  | $X  | $X  |
| TOTAL   | $X  | $X  | $X  | $X  | $X  | $X  | $X  |

---

### WEEK 3-4: MAINTAIN & OPTIMIZE

**Ongoing Actions:**
- Continue 20% increases where stable
- Add new creative batches weekly
- Monitor for fatigue signals
- Rotate underperformers out

---

### DAY-PARTING SCHEDULE

**Weekdays (Mon-Fri):**
| Time (Eastern) | Action |
|----------------|--------|
| 4:00 AM | Campaigns ON |
| 6:00 AM | In auction, monitoring begins |
| 1:00 PM | Campaigns OFF |

**Weekends (Sat-Sun):**
| Time (Eastern) | Action |
|----------------|--------|
| 4:00 AM | Campaigns ON |
| 8:00 PM Pacific | Evaluate/OFF |

**Weekend Campaigns:**
- Separate CBO for weekend only
- Higher budgets (2x daily)
- Best performers only

---

### CREATIVE REFRESH CADENCE

**Weekly:**
- 50-100 new variations
- Iterations of winners
- New hooks on proven angles

**Bi-Weekly:**
- New angle concepts
- Fresh UGC/footage
- Test new formats

---

### RISK MANAGEMENT

**If CPA Rises 10%+:**
1. Reduce budget 20%
2. Check for creative fatigue
3. Check for audience saturation
4. Review bid cap options

**If Account Goes Down:**
1. Shift spend to backup accounts
2. Have account provider ready
3. Document what triggered issue

**If Volume Drops:**
1. Expand targeting (2% → 10% LAL)
2. Test broad audiences
3. Open placements
4. New creative angles

---

### SUCCESS METRICS

| Week | Target Spend | Target CPA | Actual | Status |
|------|--------------|------------|--------|--------|
| 1    | $[X]         | $[X]       |        |        |
| 2    | $[X]         | $[X]       |        |        |
| 3    | $[X]         | $[X]       |        |        |
| 4    | $[X]         | $[X]       |        |        |
```

## Scaling Principles

**From Casto:**
- Spread across accounts (max $10K per CBO)
- Build automation tools for launching
- Layer bid/cost caps when margins thin
- Creative volume drives scale

**From Jason K:**
- 20% budget increases
- Scale in place, then replicate
- Weekend campaigns separate
- Day-part for VSL offers
- Goal: Large CBO with many winners

Source: Casto, Jason K (Meta-CastovsJasonK)