Budget Planning (Global)
A budget is not a fixed number — it is a decision tool: where you place bets and when you pull back. The leader must know total spend, ROI per channel, what to increase, and what to cut. Run 10-reverse-kpi-global first to derive the minimum budget from the revenue target before allocating anything.
Information gathering
Read .agents/product-marketing-context-global.md, the marketing plan, and the previous period's report if available. If missing, ask up to 4 questions:
- Total budget and planning period? How much, for a month, a quarter, or a year (USD)?
- Revenue target for the period? Needed to compute marketing as a percent of revenue and the minimum viable budget.
- Active channels and last period's ROI? Meta / TikTok / Google / email + SMS / affiliates / agency — which has the best ROAS or CPA?
- Any large campaign needing a reserve? Launch, Black Friday / Cyber Monday, or another seasonal peak.
Principles
- Budget follows performance. Do not stay rigid to the opening plan. Channels that work get more; channels that do not get cut.
- Always hold 10-15% in reserve. Never allocate everything on day one. The reserve covers unexpected opportunities and incidents.
- Every dollar has a measurable KPI. "Branding you cannot measure" is not an acceptable line item.
- Do not spread evenly. Concentrate on the best-performing channel. New channels get a minimum test allocation (10-15%) first.
- Plan backwards from revenue, not from channels. The budget is the output of a revenue equation, not an arbitrary number.
- Do not divide the annual budget evenly across quarters. Median Meta CPM in Q4 runs roughly 26% above Q1 (
references/benchmarks-global.md), so identical monthly spend buys materially less reach in Q4. Either front-load acquisition into cheap Q1 months or explicitly reserve a larger Q4 allocation for the same volume.
Workflow
1. Choose the budget-setting method
| Method |
How it works |
Strength |
Weakness |
| Percent of revenue |
Budget = X% of expected revenue |
Simple |
Ignores growth stage |
| Backwards from targets (recommended) |
Set revenue > derive CAC/CPL > derive spend |
Logical, evidence-based |
Needs historical CAC/CPL |
| Competitive benchmark |
Match category spend norms |
Avoids under-investment |
Ignores your own conditions |
Seven-step backwards calculation (full detail in 10-reverse-kpi-global):
1. Target revenue
2. / AOV = orders needed
3. / close rate (30-50%) = hot leads needed
4. / nurture CVR (50-60%) = warm leads needed
5. / lead capture CVR (20-40%) = total reach needed
6. total leads x CPL = minimum ad budget
7. + content + tools + agency = total marketing budget
Build 3 scenarios: Conservative (CVR below category median, CPL above), Base case (at median), Optimistic (post-optimization). Plan against Base; prepare cashflow against Conservative.
2. Allocate by bucket
| Bucket |
Suggested share |
Purpose |
| Paid ads (performance) |
50-65% |
Meta, TikTok, Google, retargeting |
| Content production |
10-15% |
Video, design, copywriting |
| Creators / UGC / affiliates |
5-10% |
Social proof, third-party reach |
| Agency / freelancers |
Per scope |
If used |
| Tools / software |
2-5% |
Ad tools, design, email/CRM, analytics |
| Campaign reserve |
10-15% |
Opportunities and incidents — NOT pre-allocated |
3. Allocate by channel
Starting split (adjust to your category and your own data):
| Channel |
Share of ad budget |
Objective |
Expected CPA / ROAS |
| Meta Ads (Facebook + Instagram) |
45-60% |
Lead / purchase |
|
| TikTok Ads |
15-20% |
Awareness / lead |
|
| Google Search + Shopping |
15-25% |
High-intent capture |
|
| Email + SMS (Klaviyo, Mailchimp, Attentive) |
5-10% |
Retention, nurture, retargeting |
|
| New channel (test) |
10-15% |
Validate |
|
For B2B, shift toward LinkedIn and Search — LinkedIn CPM runs $30-100+, so plan a higher CPL and a longer payback window. Allocation logic: channels with strong ROAS or CPA get a larger share; new channels get the minimum test allocation first; never divide evenly.
4. Split by funnel and by test vs scale
Standard performance split inside each channel:
| Purpose |
Share |
Note |
| Testing (new creative + audiences) |
30% |
Always keep testing — creative fatigue arrives fast |
| Scale (confirmed winners) |
50% |
Only scale campaigns with winning data |
| Retargeting |
15% |
Warm audience — cheapest CPA in the funnel |
| Lookalike / similar audiences |
5% |
Expansion from converter seeds |
By funnel: TOFU (reach/awareness), MOFU (lead/nurture), BOFU (conversion/retargeting). The ratio depends on stage: a launch skews TOFU, a mature program skews BOFU and retention.
5. Allocate by campaign and month
| Campaign |
Month |
Budget (USD) |
Objective |
KPI |
| Always-on (retargeting + nurture) |
Full period |
|
|
|
| Campaign A (launch or seasonal) |
|
|
|
|
| Testing budget |
Full period |
|
New channels or angles |
|
| Reserve |
— |
10-15% |
|
|
Seasonality overlay before locking monthly numbers:
| Period |
Media cost vs Q1 |
Planning implication |
| January |
~-21% below Q1 average |
Cheapest acquisition window — stock budget here if cashflow allows |
| Q2-Q3 |
Rising |
Normal operating cadence, build audiences for Q4 |
| October |
~+10% |
Start of the expensive window |
| November |
~+27% (peak) |
Reserve extra budget or accept lower volume |
| Q4 average |
~+26% above Q1 |
Do not plan Q4 at Q1 unit costs |
In US election years, political spend displaces commercial inventory and adds roughly 20-40% to CPM in the last 30 days before the vote — build that into Q4 plans for US-targeted campaigns.
6. Scale-up and stop-loss rules
Increase budget when (BOTH conditions hold):
- ROAS above 1.5x target for 7+ consecutive days, OR CPA at or below 70% of target with sufficient volume
- Audience not yet saturated (frequency below 2.5)
- Increment: +20-30% per step, never a sudden doubling
Cut or pause when (stop-loss):
| Condition |
Action |
| ROAS below break-even for 14 days |
Cut the channel or campaign, move budget to what works |
| CPA above 2x target after 7 days of testing |
Pause, replace creative or audience before restarting |
| Frequency above 4 |
Audience saturation — stop scaling, refresh creative |
| Spend above 110% of the monthly plan |
Cap daily budgets, review the allocation |
Weekly CPA review rule (expressed relative to your own target, since absolute CPA varies 3-5x between regions and industries):
| CPA vs target |
Frequency |
Verdict |
| <= 70% of target |
< 2.0 |
WIN — scale +20-30% |
| 70-100% of target |
< 2.5 |
Acceptable — monitor |
| 100-130% of target |
2.0-2.5 |
Optimize — replace creative |
| > 130% of target |
> 2.5 |
PAUSE and replace |
Set the target itself using references/benchmarks-global.md for your region and industry, not a global average. Never increase budget while CPA is deteriorating — fix creative or tracking first.
7. Tracker and review cadence
| Month |
Planned budget |
Actual spend |
Variance |
ROAS / ROI |
Note |
| Month 1 |
|
|
|
|
|
| Month 2 |
|
|
|
|
|
| Month 3 |
|
|
|
|
|
| Quarter total |
|
|
|
|
|
| Cadence |
Work |
Output |
| Weekly |
Review CPA and ROAS per channel, apply the step-6 decision rules |
Immediate adjustment when a threshold trips |
| Monthly |
Review the full allocation, reconcile plan vs actual |
Rebalance for next month |
| Quarterly |
Review overall allocation strategy and LTV:CAC |
Strategy adjustment |
Unit economics health — LTV:CAC:
| Ratio |
Meaning |
Action |
| < 1:1 |
Losing money on every customer |
Stop scaling, fix product or price |
| 1:1 to 3:1 |
Break-even to viable |
Optimize conversion, raise AOV |
| > 3:1 |
Healthy |
Begin scaling |
| > 5:1 |
Strong |
Increase ad budget aggressively |
Output structure
File name: budget-plan-[brand-name]-[YYYYMMDD].md — contains: I. Budget overview (total, revenue target, marketing as % of revenue, period) · II. Bucket allocation · III. Channel allocation · IV. Test vs scale split · V. Campaign and monthly allocation with seasonality overlay · VI. Decision rules and stop-loss thresholds · VII. Plan-vs-actual tracker · VIII. Review cadence.
Related skills
10-reverse-kpi-global: derive the minimum budget from the revenue target — run before allocating.
00-marketing-plan-global: the budget plan is the detailed version of the plan's budget section.
54-media-plan-global: turn the ad budget into a concrete campaign and ad-set level media plan.
21-ads-audit-global: audit a channel that repeatedly trips the stop-loss thresholds.
07-marketing-report-global: budget reconciliation inside the monthly report.
Quality checklist
1---
2name: 61-budget-planning-global
3description: "Use when a marketing BUDGET has to be planned or managed in USD — allocation by bucket, channel, and month, test versus scale split, scale-up and stop-loss thresholds, reserve, a plan-versus-actual tracker, and a review cadence. Trigger on 'budget planning', 'marketing budget', 'split the budget across channels', 'stop loss threshold', 'quarterly budget', 'we keep overspending and nobody notices'. Also use when a number has been approved and nobody has divided it up. Not for — the revenue-backward math that produces the number, see `10-reverse-kpi-global`; paid media allocation only, see `54-media-plan-global`; paying an external vendor, see `67-agency-vendor-brief-global`."
4metadata:
5 version: 1.0.1
6 category: strategy
7license: MIT
8triggers:
9 - "budget planning"
10 - "marketing budget"
11 - "budget allocation"
12 - "how much should we spend on ads"
13 - "campaign budget"
14 - "stop loss threshold"
15 - "quarterly budget"
16 - "media budget split"
17output: "File .md — budget plan: overview, allocation by bucket / channel / campaign / month, test vs scale split, adjustment rules and stop-loss thresholds, tracker, review cadence"
18related:
19 - product-marketing-context-global
20 - 10-reverse-kpi-global
21 - 00-marketing-plan-global
22 - 54-media-plan-global
23 - 55-scaling-ads-global
24 - 21-ads-audit-global
25 - 07-marketing-report-global
26---
27
28# Budget Planning (Global)
29
30> A budget is not a fixed number — it is a decision tool: where you place bets and when you pull back. The leader must know total spend, ROI per channel, what to increase, and what to cut. Run `10-reverse-kpi-global` first to derive the minimum budget from the revenue target before allocating anything.
31
32## Information gathering
33
34Read `.agents/product-marketing-context-global.md`, the marketing plan, and the previous period's report if available. If missing, ask up to 4 questions:
35
361. **Total budget and planning period?** How much, for a month, a quarter, or a year (USD)?
372. **Revenue target for the period?** Needed to compute marketing as a percent of revenue and the minimum viable budget.
383. **Active channels and last period's ROI?** Meta / TikTok / Google / email + SMS / affiliates / agency — which has the best ROAS or CPA?
394. **Any large campaign needing a reserve?** Launch, Black Friday / Cyber Monday, or another seasonal peak.
40
41## Principles
42
431. **Budget follows performance.** Do not stay rigid to the opening plan. Channels that work get more; channels that do not get cut.
442. **Always hold 10-15% in reserve.** Never allocate everything on day one. The reserve covers unexpected opportunities and incidents.
453. **Every dollar has a measurable KPI.** "Branding you cannot measure" is not an acceptable line item.
464. **Do not spread evenly.** Concentrate on the best-performing channel. New channels get a minimum test allocation (10-15%) first.
475. **Plan backwards from revenue, not from channels.** The budget is the output of a revenue equation, not an arbitrary number.
486. **Do not divide the annual budget evenly across quarters.** Median Meta CPM in Q4 runs roughly 26% above Q1 (`references/benchmarks-global.md`), so identical monthly spend buys materially less reach in Q4. Either front-load acquisition into cheap Q1 months or explicitly reserve a larger Q4 allocation for the same volume.
49
50## Workflow
51
52### 1. Choose the budget-setting method
53
54| Method | How it works | Strength | Weakness |
55|--------|--------------|----------|----------|
56| Percent of revenue | Budget = X% of expected revenue | Simple | Ignores growth stage |
57| Backwards from targets (recommended) | Set revenue > derive CAC/CPL > derive spend | Logical, evidence-based | Needs historical CAC/CPL |
58| Competitive benchmark | Match category spend norms | Avoids under-investment | Ignores your own conditions |
59
60Seven-step backwards calculation (full detail in `10-reverse-kpi-global`):
61
62```
631. Target revenue
642. / AOV = orders needed
653. / close rate (30-50%) = hot leads needed
664. / nurture CVR (50-60%) = warm leads needed
675. / lead capture CVR (20-40%) = total reach needed
686. total leads x CPL = minimum ad budget
697. + content + tools + agency = total marketing budget
70```
71
72Build 3 scenarios: Conservative (CVR below category median, CPL above), Base case (at median), Optimistic (post-optimization). Plan against Base; prepare cashflow against Conservative.
73
74### 2. Allocate by bucket
75
76| Bucket | Suggested share | Purpose |
77|--------|-----------------|---------|
78| Paid ads (performance) | 50-65% | Meta, TikTok, Google, retargeting |
79| Content production | 10-15% | Video, design, copywriting |
80| Creators / UGC / affiliates | 5-10% | Social proof, third-party reach |
81| Agency / freelancers | Per scope | If used |
82| Tools / software | 2-5% | Ad tools, design, email/CRM, analytics |
83| Campaign reserve | 10-15% | Opportunities and incidents — NOT pre-allocated |
84
85### 3. Allocate by channel
86
87Starting split (adjust to your category and your own data):
88
89| Channel | Share of ad budget | Objective | Expected CPA / ROAS |
90|---------|--------------------|-----------|---------------------|
91| Meta Ads (Facebook + Instagram) | 45-60% | Lead / purchase | |
92| TikTok Ads | 15-20% | Awareness / lead | |
93| Google Search + Shopping | 15-25% | High-intent capture | |
94| Email + SMS (Klaviyo, Mailchimp, Attentive) | 5-10% | Retention, nurture, retargeting | |
95| New channel (test) | 10-15% | Validate | |
96
97For B2B, shift toward LinkedIn and Search — LinkedIn CPM runs $30-100+, so plan a higher CPL and a longer payback window. Allocation logic: channels with strong ROAS or CPA get a larger share; new channels get the minimum test allocation first; never divide evenly.
98
99### 4. Split by funnel and by test vs scale
100
101Standard performance split inside each channel:
102
103| Purpose | Share | Note |
104|---------|-------|------|
105| Testing (new creative + audiences) | 30% | Always keep testing — creative fatigue arrives fast |
106| Scale (confirmed winners) | 50% | Only scale campaigns with winning data |
107| Retargeting | 15% | Warm audience — cheapest CPA in the funnel |
108| Lookalike / similar audiences | 5% | Expansion from converter seeds |
109
110By funnel: TOFU (reach/awareness), MOFU (lead/nurture), BOFU (conversion/retargeting). The ratio depends on stage: a launch skews TOFU, a mature program skews BOFU and retention.
111
112### 5. Allocate by campaign and month
113
114| Campaign | Month | Budget (USD) | Objective | KPI |
115|----------|-------|--------------|-----------|-----|
116| Always-on (retargeting + nurture) | Full period | | | |
117| Campaign A (launch or seasonal) | | | | |
118| Testing budget | Full period | | New channels or angles | |
119| Reserve | — | 10-15% | | |
120
121Seasonality overlay before locking monthly numbers:
122
123| Period | Media cost vs Q1 | Planning implication |
124|--------|------------------|----------------------|
125| January | ~-21% below Q1 average | Cheapest acquisition window — stock budget here if cashflow allows |
126| Q2-Q3 | Rising | Normal operating cadence, build audiences for Q4 |
127| October | ~+10% | Start of the expensive window |
128| November | ~+27% (peak) | Reserve extra budget or accept lower volume |
129| Q4 average | ~+26% above Q1 | Do not plan Q4 at Q1 unit costs |
130
131In US election years, political spend displaces commercial inventory and adds roughly 20-40% to CPM in the last 30 days before the vote — build that into Q4 plans for US-targeted campaigns.
132
133### 6. Scale-up and stop-loss rules
134
135**Increase budget when (BOTH conditions hold):**
136
137- ROAS above 1.5x target for 7+ consecutive days, OR CPA at or below 70% of target with sufficient volume
138- Audience not yet saturated (frequency below 2.5)
139- Increment: +20-30% per step, never a sudden doubling
140
141**Cut or pause when (stop-loss):**
142
143| Condition | Action |
144|-----------|--------|
145| ROAS below break-even for 14 days | Cut the channel or campaign, move budget to what works |
146| CPA above 2x target after 7 days of testing | Pause, replace creative or audience before restarting |
147| Frequency above 4 | Audience saturation — stop scaling, refresh creative |
148| Spend above 110% of the monthly plan | Cap daily budgets, review the allocation |
149
150Weekly CPA review rule (expressed relative to your own target, since absolute CPA varies 3-5x between regions and industries):
151
152| CPA vs target | Frequency | Verdict |
153|---------------|-----------|---------|
154| <= 70% of target | < 2.0 | WIN — scale +20-30% |
155| 70-100% of target | < 2.5 | Acceptable — monitor |
156| 100-130% of target | 2.0-2.5 | Optimize — replace creative |
157| > 130% of target | > 2.5 | PAUSE and replace |
158
159Set the target itself using `references/benchmarks-global.md` for your region and industry, not a global average. **Never increase budget while CPA is deteriorating** — fix creative or tracking first.
160
161### 7. Tracker and review cadence
162
163| Month | Planned budget | Actual spend | Variance | ROAS / ROI | Note |
164|-------|----------------|--------------|----------|------------|------|
165| Month 1 | | | | | |
166| Month 2 | | | | | |
167| Month 3 | | | | | |
168| Quarter total | | | | | |
169
170| Cadence | Work | Output |
171|---------|------|--------|
172| Weekly | Review CPA and ROAS per channel, apply the step-6 decision rules | Immediate adjustment when a threshold trips |
173| Monthly | Review the full allocation, reconcile plan vs actual | Rebalance for next month |
174| Quarterly | Review overall allocation strategy and LTV:CAC | Strategy adjustment |
175
176Unit economics health — LTV:CAC:
177
178| Ratio | Meaning | Action |
179|-------|---------|--------|
180| < 1:1 | Losing money on every customer | Stop scaling, fix product or price |
181| 1:1 to 3:1 | Break-even to viable | Optimize conversion, raise AOV |
182| > 3:1 | Healthy | Begin scaling |
183| > 5:1 | Strong | Increase ad budget aggressively |
184
185## Output structure
186
187File name: `budget-plan-[brand-name]-[YYYYMMDD].md` — contains: I. Budget overview (total, revenue target, marketing as % of revenue, period) · II. Bucket allocation · III. Channel allocation · IV. Test vs scale split · V. Campaign and monthly allocation with seasonality overlay · VI. Decision rules and stop-loss thresholds · VII. Plan-vs-actual tracker · VIII. Review cadence.
188
189## Related skills
190
191- `10-reverse-kpi-global`: derive the minimum budget from the revenue target — run before allocating.
192- `00-marketing-plan-global`: the budget plan is the detailed version of the plan's budget section.
193- `54-media-plan-global`: turn the ad budget into a concrete campaign and ad-set level media plan.
194- `21-ads-audit-global`: audit a channel that repeatedly trips the stop-loss thresholds.
195- `07-marketing-report-global`: budget reconciliation inside the monthly report.
196
197## Quality checklist
198
199- [ ] Budget derived backwards from the revenue target, with 3 scenarios
200- [ ] All 3 allocation layers present: bucket > channel > campaign/month
201- [ ] Test vs scale split explicit (testing 30 / scale 50 / retargeting 15 / lookalike 5)
202- [ ] 10-15% reserve held back, not pre-allocated
203- [ ] Scale-up and stop-loss thresholds use concrete numbers (ROAS, CPA vs target, frequency, days)
204- [ ] New channels receive only a 10-15% test allocation
205- [ ] Every budget line has a measurable KPI attached
206- [ ] Monthly plan accounts for Q4 media inflation rather than dividing the year evenly
207- [ ] Plan-vs-actual tracker exists by month
208- [ ] Weekly, monthly, and quarterly review cadence with defined outputs
209- [ ] LTV:CAC checked before any scale decision