Three new substantive frameworks integrated from Founding Marketing
(Corey Haines book) into the marketing-plan skill. Bumps
marketing-plan from 1.0.0 to 1.1.0.
New reference files:
- references/budget-planning.md (168 lines) — two scientific methods
for setting the marketing budget: Method 1 Revenue-Based (5-40% of
ARR with conservative/standard/aggressive postures) and Method 2
Goal-Based (formula: [(New ARR / (ARPC * 12)) * CAC] / annual
retention rate). Includes the 3-3-2-2-2 VC growth path (3x in years
1-2, 2x in years 3-7 from $1M ARR), the 10-20% experimental buffer
rule, blended CAC calculation (must include salaries, content,
tools, retainers - not just paid spend), and the forecasting reality
check (no startup under $100M ARR reliably hits monthly forecasts).
- references/growth-patterns.md (148 lines) — the real shape of SaaS
growth. Phase identification ($0-10K, $10K-100K, $100K-1M, $1M+)
with binding constraints per phase. Linear vs step-function vs
layered S-curve growth patterns. Channel x Product x Market S-curve
layering principle (start the next curve before the current one
plateaus). Includes the "long slow SaaS ramp of death" framing and
the case against the exponential-growth myth.
- references/team-and-agency-model.md (278 lines) — strategy
in-house, execution outsourced principle. Three core marketing
functions (Growth = demand engine, Product = story engine, Content
= trust engine). pi-shaped marketer framework (vs T-shaped) with
high-leverage skill-set combinations. Title progression (Manager
-> Lead -> Director -> VP -> Chief) — don't inflate the first hire.
Agency selection framework (Individual contractors / Small niche /
Small generalist). Three-stage scaling model (Early / Growth /
Scale) with internal-vs-external ratio guidance.
Enhanced existing files to wire in the new frameworks:
- SKILL.md (244 -> 270 lines) — new sections "Setting the budget
scientifically," "Growth patterns," "Team and agency model" with
pointers to the new reference docs. Expanded "What every plan must
customize" from 7 items to 9 (added unit economics + phase of
growth). Description still 949 chars (under 1024).
- references/funding-stage-unlocks.md — Related docs pointer block.
- references/measurement-framework.md — Related docs pointer block,
new "Anchoring against the VC growth path" section, new
"Forecasting reality check" section.
- references/methodology.md — Intake 5 (Team) now probes for
pi-shaped vs T-shaped vs tactical-only marketing owner. Intake 6
(Budget) now probes for blended CAC, ARPC, retention rate.
- references/plan-template.md — Section 3 adds Phase of SaaS Growth.
Section 8 adds required unit economics table. Section 10 adds
budget-method framing block at top + per-quarter S-curve position
field. Section 11 adds RACI table mapping the three core functions
to internal-owner and external-executor.
Total new content: ~600 lines across three new reference docs plus
~120 lines of integration edits. SKILL.md remains well under the
500-line convention (270).
Validates: 43/43 skills pass validate-skills.sh.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Sweep missed one heading in the research.md template embedded in
methodology.md — "## Audit-marketing scores" was a section the
planner uses to capture rubric output in research.md. Renamed to
"## Current-state rubric scores" and pointed at the embedded rubric.
Surfaced during /review on the fix branch.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
The marketing-plan skill was authored in an environment where the
internal `audit-marketing` and `positioning` skills (cf-skills repo,
not publicly installable) were available. Public users of this skill
would see 27 references to a skill they cannot install, plus an
explicit private-repo path.
Sweep across 7 files:
- `SKILL.md` frontmatter description rewritten to drop "see audit-marketing"
and "see positioning"; replaced positioning reference with
product-marketing (its public-skill equivalent for foundational
positioning/ICP/voice capture)
- `SKILL.md` body: "Cross-references" and "Related skills" sections
reorganized around skills that actually live in this repo
- `methodology.md` Step 1.5 rewritten — instead of "offer to run
/audit-marketing," it now describes scoring from materials or
ingesting a separately scored audit
- `current-state-rubric.md` intro rewritten — the rubric is now the
source of truth for Section 3, not a derivative of audit-marketing.
Removed cf-skills repo path. Removed the "Path A vs Path B" framing.
- `plan-template.md`, `ops-stack-mapping.md`, `client-types.md`,
`example-quietude.md` updated to remove `audit-marketing` and
`positioning` from skill lists
The rubric content itself (17 sections with scoring guides) was already
embedded in current-state-rubric.md, so external users have everything
they need to execute Section 3 of the plan.
validate-skills.sh: 43/43 passing. Description still 949 chars.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Earlier sweep (f8ac011) missed three real-brand instances and several
real partner/venue names. Surfaced during PR #331 review.
- `olo` (original brand name surviving as a repo name) ->
`quietude-app` to match the other quietude-* repos in the example
- `EHA` (real venue) -> `Aurora`
- `Moonrise`, `Alchemy Springs` (real partner venues) -> `Lumen`, `Stillwater`
- `€257K` (suspiciously precise contract value tied to anonymized partner) ->
`~€250K` so the teaching value of "B2B install size" stays while the
identifying precision drops
Funnel numbers, $430K/install economics, budget shape, and structural
lessons preserved per the example file disclaimer.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Replaces all client-identifying details in the canonical example and
6 other reference files. Numbers and structural lessons preserved so
the example retains its teaching value.
Anonymizations applied:
- Client: Olo to Quietude (incl. all olo.app/space/audio/center domains)
- Team: Markus to Alex, Catarina to Sam, Corey Haines to Casey Reed,
Shariq to Devon
- Product: Ona (AI) to Mira
- Repos: olo-* and Olo-Space/* to quietude-* and Quietude-Inc/*
- Endorser/investor specifics: Bryan Johnson to longevity-influencer
(generic), Mark Pincus + OpenAI angels to consumer-tech angels,
Andre Hurd/Daybreakers to partner-event-business
- Journal: Psychophysiology study to peer-reviewed psychophysiology study
- Substack: Coffee with Catarina to Sams newsletter
Other changes:
- Renamed references/example-olo.md to references/example-quietude.md
- Updated disclaimer at top of example to explain anonymization
- Updated VERSIONS.md changelog and plan-template.md reference
- Fixed an earlier blockquote that broke the AARRR table layout
Verified zero remaining identifying refs across all marketing-plan files.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Avoids name collision with Claude Codes built-in Plan agent and matches
existing repo naming patterns (marketing-ideas, audit-marketing).
- Renames skills/plan/ to skills/marketing-plan/
- Updates name field and slash-command references in SKILL.md
- Updates canonical example reference in example-olo.md
- Updates VERSIONS.md row and changelog entry
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Corey Haines