00-marketing-plan-global
About 00-marketing-plan-global
A master skill for global marketing plans, combining competitor research, customer insights, and reverse KPI work before assembling a seven-section plan. It is intended for teams planning around a product, audience, market, and budget.
Key features
- 01Calls competitor, insight, and reverse KPI skills
- 02Assembles plans with a seven-section template
- 03Covers SAVE, content, and channel planning
- 04Structures budgets, KPIs, and risk matrices
Use cases
- 01Build a global plan around a product and audience
- 02Plan content distribution, channel mix, and budget
- 03Translate revenue targets into marketing KPIs
MCP support
The README mentions pulling ad data through connected MCP servers as part of an audit workflow; it does not describe this skill as exposing MCP tools.
Install
Copy the command and run it in a terminal.
npx skills add https:/ / github.com/ minhnv0807/ ai-business-skills --skill 00-marketing-plan-globalAfter installing, type/00-marketing-plan-globalin Claude Code to confirm it loaded.
Details
From the project READMEGlobal Marketing Plan
Master skill — calls 08-competitor-research-global, 09-customer-insight-global, 10-reverse-kpi-global before producing the final output.
For newcomers
If this is your first global marketing plan, follow this order:
- Step 0 — Read
.agents/product-marketing-context-global.md(set once per product) - Information gathering — Answer the 4 questions below
- Run sub-skills — 08 (competitors), 09 (insights), 10 (reverse KPIs)
- Assemble plan — Use the 7-section template
- Cross-check — Run the quality checklist
If .agents/product-marketing-context-global.md does not exist, ask the user to run product-marketing-context-global first to lock down product, ICP, region, currency, and brand voice.
Step 0 — Read context file
Before producing any output, read .agents/product-marketing-context-global.md to load:
- Product / service description, USP, pricing tier
- Target region (US / EU / SEA / LATAM / APAC / GLOBAL)
- Reporting currency (USD / EUR / GBP / SGD / etc.)
- Brand voice & tone-of-voice rules
- Existing channels and audience data
If the file is missing, ask the user to create it first. Do not guess context.
Information gathering
Ask the user up to 4 questions before starting:
- Product / service? Short description, core USP, price tier (entry / mid / premium).
- Target audience? Age range, location, occupation, behaviour, biggest pain point.
- Goal & budget? Target revenue per month, marketing budget (ads + content + people), campaign duration.
- Stage? Pre-launch / Launch / Growth / Mature? Existing channels with traction? Available data (followers, email list, customer database)?
If region was not captured in context file, also ask: Primary market? (US / EU / SEA / LATAM / APAC / GLOBAL multi-region).
Section 1 — Overall strategy
1.1 Situation summary
| Item | Detail |
|---|---|
| Product / service | [name + USP] |
| Market | [region + size] |
| Stage | Pre-launch / Launch / Growth / Mature |
| Plan horizon | [months] |
| Total budget | [amount + currency] |
1.2 SWOT analysis
| Positive | Negative | |
|---|---|---|
| Internal | Strengths: [list] | Weaknesses: [list] |
| External | Opportunities: [list] | Threats: [list] |
1.3 Competitive moat analysis
Call skill
08-competitor-research-globalfor the full data set.
| Competitor | Strong channel | Exploitable weakness | Their moat | Your moat |
|---|---|---|---|---|
| [Competitor 1] | ||||
| [Competitor 2] | ||||
| [Competitor 3] |
Moat types to consider:
- Brand trust (recognition + credibility)
- Content depth (long-tail expertise competitors will not replicate)
- Community lock-in (loyal customer community)
- Data advantage (better customer understanding)
- Distribution (more channels, better partners)
- Price / unit economics (sustainable lower CAC or higher LTV)
1.4 Customer insight
Call skill
09-customer-insight-globalto ground the insight in evidence.
| Element | Description |
|---|---|
| Biggest pain | [specific, observable] |
| Hidden desire | [what they truly want but rarely say] |
| Purchase barrier | [why they have not bought yet] |
| Buying trigger | [the situation that flips them to action] |
| Trusted source | [people / channels that influence them] |
Section 2 — SAVE Framework
Replaces the traditional 4P. More appropriate for services and digital products.
2.1 Solution (replaces Product)
| Question | Answer |
|---|---|
| What problem are customers facing? | [specific pain] |
| How does the product solve it? | [mechanism] |
| What outcome do they get? | [specific, measurable] |
| How fast do they see results? | [timeframe] |
2.2 Access (replaces Place)
| Channel | Funnel role | Priority |
|---|---|---|
| [Channel 1] | TOFU — attract | High / Medium / Low |
| [Channel 2] | MOFU — nurture | |
| [Channel 3] | BOFU — convert | |
| [Channel 4] | Retention — retain |
Region note: Channel mix differs per market. See
product-marketing-context-globalfor region-specific channel preferences. SEA leans on Zalo / LINE / Shopee Live, EU favours email + LinkedIn for B2B, US favours podcast + newsletter, LATAM favours WhatsApp + TikTok.
2.3 Value (replaces Price)
| Item | Detail |
|---|---|
| Product / service price | [tier + currency] |
| Customer value received | [quantified: saves X, gains Y] |
| Value-to-price ratio | [Nx] |
| Vs competitors | [cheaper / pricier — reason] |
| Pricing psychology | [anchor / bundle / tier / free trial] |
2.4 Education (replaces Promotion)
| Awareness stage | Education content | Goal |
|---|---|---|
| Unaware of problem | Pain-aware content, awareness pieces | Help them recognise the problem |
| Problem-aware, solution-unaware | How-to, comparison of approaches | Help them weigh the options |
| Solution-aware, undecided | Case study, testimonial, demo | Build trust in your brand |
| Decided, not yet acting | Offer, urgency, social proof | Get them to act now |
Section 3 — Content plan
3.1 Content pillars
| Pillar | Share | Goal | Examples |
|---|---|---|---|
| Education | 35% | Build trust, SEO | How-to, tips, explainers |
| Inspiration | 25% | Engagement, virality | Case study, before/after, story |
| Entertainment | 20% | Reach, follower growth | Trends, POV, behind-the-scenes |
| Selling | 15% | Conversion | Offer, deal, direct CTA |
| Community | 5% | Retention | Q&A, poll, user spotlight |
3.2 Funnel distribution
| Funnel stage | Content share | Primary angle | KPI |
|---|---|---|---|
| TOFU (Awareness) | 40% | Pain points, entertainment, education | View, reach, follower |
| MOFU (Consideration) | 35% | Proof, expertise, process | Engagement, save, click |
| BOFU (Conversion) | 15% | Offer, urgency, retarget | Message, lead, order |
| Retention | 10% | VIP value, referral | Repurchase, LTV |
3.3 Content repurposing matrix
One source asset = 7+ derivatives. See
01-content-calendar-globalfor detail.
| Source asset | Derivative | Channel |
|---|---|---|
| Long video (3–5 min) | Original | YouTube |
| 3 short clips (15–30s) | TikTok, Reels, Shorts | |
| 1 carousel (5–7 slides) | Instagram, LinkedIn | |
| 1 blog post (800–1500 words) | Website, SEO | |
| 1 email newsletter | Email list | |
| 3 quote cards | Story, Threads, X | |
| 1 audio clip | Podcast feed |
3.4 Source-type mix
| Type | Share | Description |
|---|---|---|
| FGC (Founder Generated Content) | 30% | Founder / expert content |
| Brand Content | 30% | Official brand assets |
| UGC (User Generated Content) | 25% | Reviews, testimonials from customers |
| EGC (Employee Generated Content) | 15% | Behind-the-scenes from team |
Section 4 — Channel system & budget
4.1 Channel deployment
| Channel | Funnel role | Monthly budget | Primary KPI |
|---|---|---|---|
| TikTok (organic + ads) | TOFU + BOFU | [amount] | View, message, CPMess |
| Meta — Facebook + Instagram (organic + ads) | MOFU + BOFU | [amount] | Reach, message, CPMess |
| Email (newsletter + flow) | MOFU + Retention | [amount] | Open rate, click rate |
| Website / SEO | MOFU | [amount] | Traffic, time on site |
| YouTube (long + Shorts) | TOFU + MOFU | [amount] | View duration, subscriber |
| Podcast / Newsletter sponsorship | TOFU | [amount] | CPM, brand lift |
| Influencer / UGC network | TOFU | [amount] | View, reach, content output |
| [Region-specific channel] |
Region note: Add LINE / Zalo / WhatsApp / WeChat per market (SEA, LATAM, China). See
product-marketing-context-globalfor region channel weights.
4.2 Budget split by stage
| Item | Launch (M1–3) | Growth (M4–6) | Mature (M7+) |
|---|---|---|---|
| Paid ads | 45–55% | 35–45% | 25–35% |
| Content production | 20–25% | 20–25% | 15–20% |
| UGC / influencer | 15–20% | 15–20% | 10–15% |
| Tools & platforms | 5–10% | 5–10% | 5–10% |
| Community & retention | 5% | 10–15% | 20–25% |
| Contingency | 5% | 5% | 5% |
| Total | 100% | 100% | 100% |
4.3 Detailed budget allocation
Example uses USD. Convert to local currency as defined in
product-marketing-context-global.
| Item | Monthly budget | % Total | Notes |
|---|---|---|---|
| Meta Ads | [amount] | [%] | |
| TikTok Ads | [amount] | [%] | |
| Google Ads (Search + YouTube) | [amount] | [%] | |
| Content (people + tools) | [amount] | [%] | |
| Influencer / UGC fees | [amount] | [%] | |
| Tools (analytics, CRM, design) | [amount] | [%] | |
| Contingency | [amount] | 5% | |
| Total | [amount] | 100% |
Section 5 — KPI & performance
5.1 Reverse KPI calculation
Call skill
10-reverse-kpi-globalfor an exact calculation.
Target revenue: [amount]
/ AOV: [amount]
= Orders needed: [count]
/ Booking->Customer [region benchmark]: [count]
= Bookings needed: [count]
/ Lead->Booking [region benchmark]: [count]
= Leads needed: [count]
/ Mess->Lead [region benchmark]: [count]
= Messages needed: [count]
x CPMess [region benchmark]: [amount]
= Required ad budget: [amount]
5.2 KPI table — 3 scenarios
| Metric | Pessimistic | Base | Optimistic |
|---|---|---|---|
| CPMess | +30% vs benchmark | Industry benchmark | -20% vs benchmark |
| Mess->Lead | -15% vs avg | Industry avg | +15% vs avg |
| Lead->Booking | -10% vs avg | Industry avg | +10% vs avg |
| Booking->Customer | -10% vs avg | Industry avg | +10% vs avg |
| Orders / month | [count] | [count] | [count] |
| Revenue / month | [amount] | [amount] | [amount] |
| ROAS | [Nx] | [Nx] | [Nx] |
| Required budget | [amount] | [amount] | [amount] |
Region benchmarks: see foundation skill
product-marketing-context-globalfor the regional benchmark file (US, EU, SEA, LATAM). Do not assume one global benchmark — CPMs vary 3–5x across regions.
5.3 KPI by channel
| Channel | Primary KPI | Month 1 target | Month 3 target | Month 6 target |
|---|---|---|---|---|
| TikTok organic | View / video, follower | |||
| TikTok ads | CPMess, ROAS | |||
| Meta ads | CPMess, CPL, ROAS | |||
| Google Search ads | CPL, ROAS | |||
| Open rate, click rate | ||||
| SEO | Organic traffic, ranking |
5.4 Business KPIs
| Metric | Formula | Target |
|---|---|---|
| ROAS | Revenue / ad spend | >3x |
| CAC | Total marketing spend / new customers | <30% AOV |
| LTV | AOV × purchases × duration | >3x CAC |
| Payback period | CAC / (AOV × margin) | <90 days |
| LTV:CAC | LTV / CAC | >3:1 |
Section 6 — Risk matrix
6.1 Risk table
| Risk | Probability | Impact | Severity | Mitigation plan |
|---|---|---|---|---|
| CPMess spikes during peak season | High | High | CRITICAL | Cut ad budget 30%, shift to organic + UGC |
| Creative fatigue (within 2 weeks) | High | Medium | HIGH | Prepare 3–5 new creatives per week, A/B test continuously |
| Competitor heavy discounting | Medium | High | HIGH | Lead with value + trust, do not race to the bottom |
| Algorithm change (TikTok / Meta) | Medium | High | HIGH | Diversify channels, no single-platform dependency |
| Key content creator leaves | Low | High | MEDIUM | Document SOPs, build backup roster |
| Negative review goes viral | Low | Very high | CRITICAL | Crisis protocol: respond within 2h, transparent, escalate |
| Budget gets cut | Medium | Medium | MEDIUM | Plan B with 50% budget — organic priority |
| Lack of customer data | Low | Medium | LOW | Capture data from day 1: forms, pixel, CRM |
6.2 Severity definitions
| Level | Definition | Action |
|---|---|---|
| CRITICAL | Direct revenue impact, lost customers | Resolve in 24h, escalate to stakeholder |
| HIGH | Significant performance drop, lost opportunity | Resolve in 48h, adjust plan |
| MEDIUM | Mild impact, recoverable | Resolve within 1 week, monitor |
| LOW | Negligible impact | Log and address opportunistically |
Section 7 — Implementation timeline
7.1 Roadmap
| Week | Theme | Detail | Output | Owner |
|---|---|---|---|---|
| Week 1 | Setup & research | Competitor research, customer insight, channel setup | Research report, channels live | [name] |
| Week 2 | Content & creative | Produce first content batch, brief UGC | 15–20 content pieces, 3–5 UGC briefs | [name] |
| Week 3 | Launch & test | Run paid test, post organically, start nurture | A/B test report, first 7 days of data | [name] |
| Week 4 | Optimise & scale | Cut bad creative, scale winners, drop weak channels | Month 1 report, Month 2 plan | [name] |
7.2 Milestones
| Month | Milestone | KPI check |
|---|---|---|
| Month 1 | Setup complete, test running, first data | CPMess, message count, reach |
| Month 2 | Found winning creative + audience | ROAS, CPL, conversion rate |
| Month 3 | Scaled main channel, funnel stable | Revenue, LTV:CAC, retention |
| Month 6 | System self-running, continuous optimisation | All KPIs stable, organic growing |
Skill chaining
This is a master skill — it calls the following sub-skills:
00-marketing-plan-global (MASTER)
|
|-- [1] 08-competitor-research-global -> Competitor analysis, gap discovery
|-- [2] 09-customer-insight-global -> Customer insight, pain points
|-- [3] 10-reverse-kpi-global -> Budget + KPIs from target revenue
|
|-- [4] Produce this plan
|
|-- [5] 01-content-calendar-global -> Detailed monthly content calendar
|-- [6] 02-campaign-brief-global -> First campaign brief
When the user requests a marketing plan:
- Ask the 4 information-gathering questions
- Run skills 08, 09, 10 (in parallel when possible)
- Assemble the plan in this template
- Suggest running 01 and 02 next if the user wants execution detail
Examples — diversified regions
Example A — US SaaS startup (Series A)
- Product: B2B project management SaaS, $49/seat/month
- Region: US, secondary CA + UK
- Stage: Growth (M9 since launch)
- Channels: SEO + content, LinkedIn ads, podcast sponsorship, lifecycle email
- Region note: B2B benchmark CAC payback <12 months, LTV:CAC >3:1
Example B — EU DTC skincare brand
- Product: Clean-formula serum, EUR 38 / 30ml
- Region: DE / FR / NL primary, expanding to UK
- Stage: Launch (3 months)
- Channels: Meta + TikTok ads, micro-influencers, Klaviyo email, Shopify
- Region note: GDPR compliance for email + ads tracking is non-negotiable
Example C — SEA fitness coaching app
- Product: Online fitness coaching subscription, SGD 29/month
- Region: Singapore primary, Malaysia + Thailand secondary
- Stage: Mature (M14)
- Channels: TikTok organic, Meta ads, LINE / WhatsApp nurture, Telegram community
- Region note: Mobile-first, conversational selling beats long-form sales pages
Quality checklist
Before delivering the plan, verify:
- All 7 sections present (Strategy, SAVE, Content, Channel, Performance, Risk, Timeline)
- SWOT grounded in real data, not assumption
- Competitive moat is specific — not generic
- SAVE Framework covers all 4 elements
- Content pillars sum to 100%
- Funnel distribution follows ratios (TOFU 40%, MOFU 35%, BOFU 15%, Retention 10%)
- Budget split by stage (Launch / Growth / Mature)
- 3-scenario KPI table calculated (Pessimistic / Base / Optimistic)
- Risk matrix has at least 5 risks with severity + mitigation
- Timeline has month-level milestones
- Region-specific benchmarks used (not blanket global numbers)
- Cross-references to related skills present
- All numbers measurable — no vague phrases like "increase a lot" or "much higher"
- Total budget sums to exactly 100%
- Currency consistent with
.agents/product-marketing-context-global.md