Produce Part 3 of the 12-Part engagement: the four strategic-spine documents across 61 steps — 3.1 Business & SBU Analysis, 3.2 Segmentation Framework, 3.3 Brand Positioning & Communications, 3.4 DMFlow — with --doc single-document runs, --view v2 re-runs, and a --combined executive stitch. Triggers on "/digital-marketing-pro:four-core-documents", "produce the four core documents", "run part 3 of the engagement", "generate the strategic spine", "re-run positioning as v2". Requires an initialised engagement with Part 1 complete; reads Stone facts, Part 2 research, and the brand profile, and feeds /digital-marketing-pro:growth-plan.
Produce Part 3 of the 12-Part engagement: the four strategic-spine documents across 61 steps — 3.1 Business & SBU Analysis, 3.2 Segmentation Framework, 3.3 Brand Positioning & Communications, 3.4 DMFlow — with --doc single-document runs, --view v2 re-runs, and a --combined executive stitch. Triggers on "/digital-marketing-pro:four-core-documents", "produce the four core documents", "run part 3 of the engagement", "generate the strategic spine", "re-run positioning as v2". Requires an initialised engagement with Part 1 complete; reads Stone facts, Part 2 research, and the brand profile, and feeds /digital-marketing-pro:growth-plan.
user-invocable
true
triggers
["produce the four core documents","run part 3 of the engagement","generate the strategic spine","business and SBU analysis","segmentation framework","brand positioning document","DMFlow document","re-run a core document as v2"]
allowed-tools
Read Write Edit Bash Glob Grep
engagement-part
3
view-preference
v2-primary
/digital-marketing-pro:four-core-documents — The Strategic Spine (61 Steps)
This skill produces Part 3 of the engagement methodology: the four documents that together define the brand at strategic depth. Every channel strategy, every creative brief, every piece of copy reads back to these.
Context efficiency
Heavy skill. Grep before Read any referenced file, then Read only matched ranges with offset + limit. List the brand's workspace at ~/.claude-marketing/brands/{slug}/ (or $CLAUDE_PLUGIN_DATA/digital-marketing-pro/brands/{slug}/ when that env var is set) before opening files. On re-invocation mid-session, skip files already in context.
Produces v2 versions of the specified documents. --doc takes a single id (3.1) or a comma-separated list ("3.1,3.3"). Used during Part 6 after the Decision Matrix has triggered re-runs. (The canonical flags are --doc, --view v2, and --combined — there is no --docs.)
Stitches all four canonical core documents (latest version of each) into a single executive-reference file with master TOC, master assumptions table, and master source index. Produced only when an executive audience needs a single-file read.
Production Order
The four documents are produced in sequence because each builds on the prior:
3.1 Business & SBU Analysis (18 steps) — foundational. Establishes the business reality.
3.2 Segmentation Framework (15 steps) — depends on 3.1's SBU and customer data.
(For v2 re-runs: same path but in v2/ instead of v1/)
Output structure:
---
document: 3.1-business-and-sbu-analysis
version: v1.0
engagement: {engagement-id}
brand: {brand-slug}
produced: {iso-timestamp}
view: v1 # or v2 for re-runs
---# 3.1 Business & SBU Analysis## Step 1: SBU Identification
[Content for step 1]
## Step 2: SBU Separation Rationale
[Content for step 2]
... (all 18 steps, in order)
## Sources
[Numbered list of every cited source: client docs, public sources, industry reports, Stone facts referenced]
## Open Questions
[Anything that could not be answered with available info]
## Change Log### v1.0 — {date}- Initial unbiased research version produced from Part 2 + Stone facts.
Every claim cites a source. No "we think" — only "the evidence shows" with cited source.
Every assumption is explicit. When estimating CAC, LTV, market size, etc., state the methodology and inputs.
Every recommendation flows from analysis. No conclusions the body does not support.
No generic statements. "Strong brand" is generic. "92% brand recall in target segment per [source]" is specific.
All four documents reference each other consistently. A persona defined in 3.2 must be the same persona referenced in 3.3 messaging variations and 3.4 channel selection.
Open questions are documented, not hidden.
Stone facts are facts; Opinion hypotheses are questions. Don't elevate Opinion to fact in the document.
Multi-Dimensional Decision Framework (see decision-framework.md) is used for any choice (channel selection, persona prioritisation, positioning trade-offs).
Single-File Discipline
Each document is produced as a single file containing all its steps. If a document cannot be completed in one turn due to output limits:
Save the partial document so far
Continue the same file from where it left off in the next turn (NEVER start a new file)
The file is considered complete only when all steps are present
There is no hidden continuation state — the skill does not persist a cut-off pointer. The partial file already saved to disk is the only record. If output is interrupted mid-document, re-run the document: read the saved partial as context and continue appending the remaining steps to the same file. For a run interrupted at the engagement level, resume from the part's last checkpoint artifact via /digital-marketing-pro:resume. The saved file is always the source of truth.
After Production
Once all four documents are produced (or after each individually):
Update the Living Project Instruction File:
Strategic Facts section: extract positioning statement, primary persona summary, channel selections, unit economics
Version History section: add the new v1.0 entries
Mark Part 3 (or the specific sub-document) as completed in _engagement.json
Notify the user that Part 4 is the natural next step
v2 Re-runs
When invoked with --view v2, the skill:
Reads the v1 version of the requested doc
Reads the Part 5 Client Validation Document (the changes the client requested)
Reads the relevant subset of changes that apply to this doc
Produces v2 with a header section listing what changed vs v1:
Saves to v2/{doc-name}.md (the v1 file remains untouched)
Bumps the version in _engagement.json via engagement-state.py bump-version
Updates the LIF with the changed strategic facts
Examples
Example 1: First-time production of all four documents
User: Produce the four core documents for the Acme engagement.
Skill response:
1. Verify pre-conditions: brand profile ✓, engagement initialised ✓, Part 1 complete ✓, Part 2 in progress ✓
2. Begin 3.1 Business & SBU Analysis. Read Stone facts, Part 2 research, brand profile. Walk all 18 steps. Cite sources throughout. Save to v1/3.1-business-and-sbu-analysis.md.
3. Begin 3.2 Segmentation Framework. Read 3.1, Part 4.3 if available. Detect business model = B2B from profile, so include steps 10-13. Walk 15 steps. Save to v1/3.2-segmentation-framework.md.
4. Begin 3.3 Brand Positioning. Read 3.1, 3.2, Part 4.2 if available. Walk 19 steps. Save to v1/3.3-brand-positioning-and-communications.md.
5. Begin 3.4 DMFlow. Read 3.1, 3.2, 3.3. Apply 5 Digital Markets taxonomy. Walk 9 steps. Save to v1/3.4-dmflow.md.
6. Update LIF: positioning statement, primary persona, channel selections, blended CAC estimate.
7. Mark Part 3 completed.
8. Brief: "Part 3 complete. 4 documents, 61 steps. Total ~12,000 words across all four. Recommend reviewing 3.3 (positioning) carefully — the chosen positioning has notable trade-offs documented in step 6."
Example 2: v2 re-run after Part 5
User: Re-run 3.3 and 3.4 as v2 — client rejected the positioning and we need to redo the channel mix as a result.
Skill response:
1. Verify Part 5 Client Validation Document exists with positioning rejection captured ✓
2. Begin 3.3 v2. Read 3.3 v1 + Client Validation Document. Re-do positioning per client direction. Add v1→v2 changes header. Save to v2/3.3-brand-positioning-and-communications.md.
3. Begin 3.4 v2. Read 3.4 v1 + (now updated) 3.3 v2. Re-do channel mix to align with new positioning. Add v1→v2 changes header. Save to v2/3.4-dmflow.md.
4. Bump versions: engagement-state.py bump-version --doc 3.3 + --doc 3.4
5. Update LIF: new positioning statement, new channel selections, version history.
6. Brief: "v2 re-runs complete. 3.1 and 3.2 unchanged (still v1.0). 3.3 and 3.4 now at v2.0. Both views remain available. Operating decisions should reference v2 docs."
Example 3: Producing the Combined Core Document for executive review
User: Produce the Combined Core Document for the executive presentation next week.
Skill response:
1. Verify all four core documents exist (latest version of each) ✓
2. Read 3.1 v1.0, 3.2 v1.0, 3.3 v2.0, 3.4 v2.0
3. Synthesise into single document with:
- Master TOC at top
- Each document as a major section
- Master Assumptions Table (extract every "Assumption" call-out from all four docs)
- Master Source Index (de-duplicated list of every source cited)
4. Save to part-03-four-core-documents/3.C-combined-core-document.md
5. Brief: "3.C produced. 60+ pages. Includes master TOC + assumptions table + source index. Recommended exports: PDF for the executive deck, DOCX if they want to annotate."
Related skills
engagement-workflow — orchestrates the 12-Part flow that contains Part 3
client-validation-document — Part 5 deliverable that triggers v2 re-runs
Existing skills and agents consumed by this one: audience-intelligence, competitive-intel, market-intelligence, brand-guardian
