Budget Allocation Optimizer
SKILL.md
HOW_TO_USE.md
sample_input.json
expected_output.json
skillsgoogle-adsSKILL.md
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
# Budget Allocation Optimizer
Get more conversions from the same budget by putting money where it works hardest.
## Core Philosophy
**Not all dollars are equal.** The last $100 spent in Campaign A might get you 1 conversion, while the first $100 added to Campaign B might get you 5.
**The Goal:** Find the allocation where every dollar is working equally hard across all campaigns.
**The Principle:** Marginal CPA analysis—move budget from campaigns where the NEXT dollar is expensive to campaigns where the NEXT dollar is cheap.
---
## Required Context
### Must Have
**1. Campaign Performance Data**
For each campaign:
- Campaign name
- Current daily/monthly budget
- Current spend
- Conversions
- CPA
**2. Time Period**
- At least 14 days of data (30 days preferred)
- Consistent period across all campaigns
### Strongly Recommended
**3. Targets & Constraints**
- Target CPA or ROAS
- Total budget (fixed or flexible)
- Minimum spend per campaign (if any)
- Maximum spend per campaign (if any)
**4. Budget Utilization**
- Is each campaign spending its full budget?
- Any campaigns limited by budget?
### Nice to Have
- Impression share data (especially lost IS due to budget)
- Historical performance at different spend levels
- Conversion value data (for ROAS optimization)
- Seasonality context
---
## Analysis Framework
### Step 1: Current State Assessment
**Campaign Efficiency Ranking:**
| Tier | CPA vs Target | Assessment |
|------|---------------|------------|
| ⭐ Star | <75% of target | Highly efficient, likely underfunded |
| ✅ Solid | 75-100% of target | Meeting goals, balanced |
| ⚠️ Borderline | 100-125% of target | Marginal, watch closely |
| ❌ Underperformer | >125% of target | Inefficient, likely overfunded |
**Budget Utilization Check:**
- Spending 100%+ of budget = Potentially constrained
- Spending <90% of budget = Demand issue or targeting too narrow
- Spending 90-100% = Healthy utilization
---
### Step 2: Marginal CPA Analysis
**The Key Question:** What happens to CPA as we increase/decrease spend?
**General Patterns:**
- **Stars at budget cap:** Marginal CPA likely still good—room to scale
- **Underperformers at budget cap:** Already inefficient at current spend
- **Low-spend campaigns:** Insufficient data to judge marginal efficiency
**Marginal CPA Estimation:**
When historical data is unavailable, use these heuristics:
| Current Efficiency | Budget Status | Estimated Marginal CPA |
|-------------------|---------------|------------------------|
| Star (<75% target) | Capped | ~1.1-1.2x current CPA |
| Star (<75% target) | Not capped | ~1.0x current CPA |
| Solid (75-100%) | Capped | ~1.2-1.3x current CPA |
| Solid (75-100%) | Not capped | ~1.1x current CPA |
| Borderline (100-125%) | Capped | ~1.3-1.5x current CPA |
| Underperformer (>125%) | Any | Already too high |
**The Reallocation Principle:**
Move budget from campaigns with HIGH marginal CPA to campaigns with LOW marginal CPA until marginal CPAs equalize.
---
### Step 3: Reallocation Recommendations
**Identify Donors (Reduce Budget):**
- Campaigns with CPA significantly above target
- Campaigns with declining efficiency trends
- Campaigns not spending full budget anyway
**Identify Recipients (Increase Budget):**
- Campaigns with CPA below target AND budget-capped
- Campaigns with strong efficiency AND room to grow
- High impression share loss due to budget
**Calculate Reallocation:**
```
Recommended shift = MIN(
Donor's reducible budget,
Recipient's absorbable budget,
Amount that keeps both at acceptable CPA
)
```
---
### Step 4: Impact Projection
**Conservative Estimate:**
- Assume recipient CPA increases 10-20% with more budget
- Assume donor CPA stays flat or improves slightly
- Net conversion gain = (new conversions from recipient) - (lost conversions from donor)
**Projection Formula:**
```
Additional conversions = Budget shifted / Recipient's projected CPA
Lost conversions = Budget shifted / Donor's current CPA
Net gain = Additional - Lost
If Net gain > 0, reallocation is beneficial
```
---
## Output Format
### Executive Summary
```
CURRENT ALLOCATION EFFICIENCY: 🟢/🟡/🔴
Total Budget: $XX,XXX/month
Blended CPA: $XX.XX
Optimization Opportunity: +XX conversions/month (estimated)
```
---
### 📊 Current State Analysis
**Campaign Efficiency Ranking:**
| Rank | Campaign | Spend | Conv | CPA | vs Target | Budget Status | Tier |
|------|----------|-------|------|-----|-----------|---------------|------|
| 1 | [Name] | $X,XXX | XX | $XX | -XX% | Capped | ⭐ |
| 2 | [Name] | $X,XXX | XX | $XX | -X% | OK | ✅ |
| 3 | [Name] | $X,XXX | XX | $XX | +XX% | Capped | ⚠️ |
| 4 | [Name] | $X,XXX | XX | $XX | +XX% | Under | ❌ |
**Key Observations:**
- [Observation about efficiency distribution]
- [Observation about budget utilization]
- [Observation about optimization opportunity]
---
### 💰 Reallocation Recommendations
**Summary:**
```
Total to reallocate: $X,XXX/month
From: [X] campaigns
To: [X] campaigns
Expected impact: +XX conversions/month
```
---
**Recommendation 1: [Donor] → [Recipient]**
| | Donor: [Campaign] | Recipient: [Campaign] |
|---|---|---|
| Current Budget | $X,XXX | $X,XXX |
| Current CPA | $XX | $XX |
| Recommended Change | -$XXX | +$XXX |
| New Budget | $X,XXX | $X,XXX |
| Projected CPA | $XX | $XX |
**Rationale:** [Why this reallocation makes sense]
**Expected Impact:**
- Conversions lost from donor: ~X
- Conversions gained from recipient: ~X
- Net gain: +X conversions
- Net CPA improvement: -$X.XX
---
**Recommendation 2: [Donor] → [Recipient]**
[Same format as above]
---
### 📈 Projected Outcome
**Before vs After:**
| Metric | Current | Projected | Change |
|--------|---------|-----------|--------|
| Total Budget | $XX,XXX | $XX,XXX | — |
| Total Conversions | XXX | XXX | +XX (+X%) |
| Blended CPA | $XX.XX | $XX.XX | -$X.XX (-X%) |
**Confidence Level:** High / Medium / Low
**Assumptions:**
- [Key assumption 1]
- [Key assumption 2]
---
### ⚠️ Risks & Considerations
**Scaling Risk:**
- [Campaign] may see CPA increase faster than projected if scaled aggressively
- Recommendation: Increase budget incrementally (20% per week)
**Reduction Risk:**
- [Campaign] contributes to brand visibility even if CPA is high
- Consider: Strategic value beyond direct conversions
**Data Limitations:**
- [Any data gaps that affect confidence]
---
### 🎯 Implementation Plan
**Week 1:**
- [ ] Reduce [Campaign A] budget by $XXX
- [ ] Increase [Campaign B] budget by $XXX
- [ ] Monitor daily for 3 days
**Week 2:**
- [ ] Evaluate performance
- [ ] If stable, continue with next reallocation
- [ ] If CPA spikes, pause and reassess
**Success Metrics:**
- Blended CPA should decrease within 2 weeks
- Conversion volume should increase within 2 weeks
- No single campaign CPA should exceed [threshold]
---
### 📋 Monitoring Checklist
After reallocation, check daily for first week:
- [ ] Recipient campaign spending new budget
- [ ] Recipient CPA within projected range
- [ ] Donor campaign stable or improved
- [ ] No impression share collapse
- [ ] Conversion volume trending up
---
## Special Scenarios
### Scenario 1: All Campaigns Efficient
**Situation:** Every campaign at or below target CPA
**Recommendation:**
- If budget-capped: Request budget increase
- If not capped: Expand targeting, add keywords
- Maintain current allocation if no expansion possible
### Scenario 2: All Campaigns Inefficient
**Situation:** Every campaign above target CPA
**Recommendation:**
- Don't reallocate (moving deck chairs on Titanic)
- Focus on fixing fundamentals first
- Consider: targeting, landing pages, offer, tracking
### Scenario 3: One Dominant Campaign
**Situation:** One campaign drives 80%+ of conversions efficiently
**Recommendation:**
- Prioritize scaling that campaign
- Reduce others to minimum viable spend
- Test new campaigns to diversify
### Scenario 4: New Campaigns with Limited Data
**Situation:** Some campaigns <14 days old
**Recommendation:**
- Exclude from reallocation analysis
- Let them mature before judging
- Note in limitations
---
## Calculation Reference
### Marginal CPA Estimation
When you don't have historical data at different spend levels:
**For budget-capped efficient campaigns:**
```
Projected CPA at +20% budget = Current CPA × 1.15
Projected CPA at +50% budget = Current CPA × 1.25
Projected CPA at +100% budget = Current CPA × 1.40
```
**For non-capped campaigns:**
```
Efficiency likely stable—CPA won't change much with budget increase
Focus on expanding targeting instead
```
### Net Conversion Calculation
```
Budget to move: $1,000
Donor campaign:
- Current CPA: $100
- Conversions lost: $1,000 / $100 = 10
Recipient campaign:
- Current CPA: $50
- Projected CPA at higher spend: $50 × 1.15 = $57.50
- Conversions gained: $1,000 / $57.50 = 17.4
Net gain: 17.4 - 10 = +7.4 conversions
```
---
## Limitations
**I can provide:**
- Efficiency-based allocation recommendations
- Impact projections based on heuristics
- Implementation framework
- Risk assessment
**I cannot provide:**
- Exact marginal CPA curves (need historical spend experiments)
- Incrementality analysis (need holdout tests)
- Cross-channel impact (need full funnel data)
- Guaranteed outcomes (projections are estimates)
**For better recommendations, also provide:**
- Historical performance at different budget levels
- Impression share data
- Conversion value data
- Competitive auction insights
---
## Quality Checklist
Before delivering recommendations:
- [ ] All campaigns ranked by efficiency
- [ ] Budget utilization status checked
- [ ] Donors clearly justified (why reduce)
- [ ] Recipients clearly justified (why increase)
- [ ] Impact projection includes assumptions
- [ ] Risks identified for each reallocation
- [ ] Implementation plan is gradual (not all at once)
- [ ] Monitoring checklist provided
- [ ] Limitations clearly stated
ReadyBudget Allocation Optimizer
MarkdownUTF-8Verified