Campaign Concept

Campaign Concept Skill

This skill turns a campaign objective into a complete campaign concept, and hands the result to cc-content as brief.md.

This skill follows the shared step sequence in ../_shared/skill-contract.md. Steps below add only campaign-concept-specific behavior.

Step 0–1: Recall learnings and load context

If .claude/learnings.md exists, read it silently. Apply relevant entries — in particular, campaign-type preferences, rejected frameworks, or discovered strategy constraints. Do not announce this step. If the file is absent, continue normally.

Read the ## Context files table in CLAUDE.md. Match the following needs against the Summary column semantically — never against a label or filename:

• Business goals/KPIs — campaign objectives, business targets, success metrics, growth stage, revenue targets
• Target audience — positioning, audience profile, customer segments, buyer personas, organization background (prefer a positioning output file if registered)

Note silently if present (do not ask about): competitive landscape, current marketing channels, growth stage, brand-specific strategy overrides, and a prior strategy performance review (strategy-performance-review output) if registered — factor validated/invalidated goal bets into the concept.

Step 2: Gating — Three-State Model

This skill gates on exactly two needs: business goals and target audience. For each, resolve one of three states:

1. Covered — present in the context table (Summary match) → use silently, no ask.
2. Absent-but-supplied — ask once (never twice in one invocation); user answers → use, non-degraded.
3. Absent-and-unanswered — asked once, no usable answer → that section is labelled ⚠ DEGRADED, never fabricated, and the rest of the campaign still generates.

Check business goals

Search the context table Summary column for semantic matches: business goals, KPIs, objectives, targets, growth stage, success metrics.

If covered, use silently.

If absent, ask once:

"I need to understand what this campaign should achieve. Briefly: What are your primary business goals or KPIs right now? (e.g. revenue growth, lead volume, retention, brand awareness) Is there a specific target or metric this campaign should move?"

If the user provides usable answers, proceed non-degraded. If the user declines or the answer is too vague to use, label the section ## Business Goals ⚠ DEGRADED in the output and continue.

Check target audience

Search the context table Summary column for semantic matches: target audience, positioning, organization background, customer segments, buyer profile.

Prefer a positioning output file's Summary if one is registered; else fall back to organization background.

If covered, use silently.

If absent, ask once:

"I need to understand who this campaign is for. Briefly: Who is the target audience? (role, company type/size if B2B, or demographic if B2C) What stage are they typically at when they encounter you? (just learning you exist, actively comparing options, or already a customer)"

If the user provides usable answers, proceed non-degraded. If the user declines or the answer is too vague to use, label the section ## Target Audience ⚠ DEGRADED in the output and continue.

Step 3: Calibrate

If --type was passed in $ARGUMENTS, use it directly and skip the campaign-type question. Otherwise infer campaign type, funnel stage, and B2B/B2C from loaded context; where inference isn't confident, ask the missing question directly rather than guessing.

Present a one-line inference summary (e.g. "Type: product launch · Audience: B2B mid-market · Stage: TOFU") and wait for user confirmation or correction before drafting.

Step 4: Generate

Apply the three frameworks from ../_shared/campaign-frameworks.md per that file's facet mapping. State the mapping being applied before generating. Example:

"Structuring with SOSTAC; sequencing channels with RACE (retention emphasis); messaging with Message House."

Produce a campaign concept that includes:

• Campaign name — memorable, descriptive working name for this initiative.
• Campaign objective — the specific outcome this campaign drives, tied to the loaded business goals.
• Target audience summary — the specific segment(s) this campaign addresses, reflecting positioning or organization context if available.
• Key messages — at least two key messages, each corresponding to a Message House pillar grounded in loaded context (positioning output, organization background, competitive landscape if available).
• Channel mix — proposed channels sequenced by RACE stage (Reach, Act, Convert, Engage) appropriate to the campaign type and funnel stage.
• Creative brief — tone/voice notes if brand-voice context is loaded, primary call-to-action, constraints (budget, timeline, resource availability), and suggested content formats per channel.

Step 4a: Generate Success Metrics

Before moving to the quality check, generate an upfront Success Metrics section that commits to measurement before the campaign runs.

Use the resolved campaign type from Step 3 and the campaign's stated business goals to select 2–4 leading/lagging metric pairs from the following table. For each pair:

Output format: For each selected metric pair, emit:

Bet: <specific claim about what this campaign will achieve for this KPI> | Expected signal: <what success looks like numerically, e.g. "achieve 28% sign-up conversion by Q3"> | Source: campaign concept §Success Metrics

Include exactly 2–4 bets (one leading metric per pair, paired with its lagging outcome).

After the metric bets, add one sentence naming the attribution approach (e.g., "We will report last-touch attribution for all paid channels, acknowledging it under-credits awareness work and should be triangulated with a 'How did you hear?' field on high-intent forms") and one sentence naming the review cadence (e.g., "We will review leading KPIs weekly during the active campaign window and lagging KPIs monthly in steady state").

The Success Metrics section becomes part of the delimited output in Step 6; it will be recognized by /strategy-performance-review without modification.

Step 5: Internal quality check

Review the generated campaign concept for:

• Completeness: All six elements above are present (campaign name, objective, target audience, key messages, channel mix, creative brief).
• Extractability: The four cc-content-extractable elements (business goals, key messages, constraints, CTAs) are present in prose — this check is what Step 6.5 relies on before offering to save to brief.md.
• Framework alignment: SOSTAC structures the campaign; RACE sequences the channels; Message House anchors the messaging.
• Consistency: All sections reinforce the same campaign positioning and objective.
• Success Metrics present and in bet-shape: The Success Metrics section includes 2–4 bets in the format Bet: [claim] | Expected signal: [result] | Source: campaign concept §Success Metrics, plus sentences on attribution and review cadence.

Fix any gaps before proceeding.

Step 6: Delimited output

If Step 2 ended in a degraded state for one or both gated needs (business goals and/or target audience), prefix the output with:

⚠ DEGRADED OUTPUT — generated without: <comma-separated list of unmet needs>

For example, if target audience is degraded but business goals are not, prefix with: ⚠ DEGRADED OUTPUT — generated without: target audience

If both are degraded, prefix with: ⚠ DEGRADED OUTPUT — generated without: business goals, target audience

If neither need is degraded, omit the prefix.

Return the campaign concept inside clear delimiters:

─────────────────────────────────────────────────────────────────
## [Campaign Name]
─────────────────────────────────────────────────────────────────

[Generated campaign concept here]

─────────────────────────────────────────────────────────────────

Step 6.5: Offer to save as brief.md

After showing the delimited output:

1. Check if brief.md exists in the working directory (the project root — not under context/). Run ls brief.md from $PWD.
2. If it exists: read it, extract the campaign name (or first heading) and the file's last-modified timestamp, and show both to the user. Ask: overwrite / save under a different name / cancel.
3. If the user chooses save under a different name: ask for a filename (or propose one derived from the new campaign name), write to that filename instead, and do not touch the existing brief.md.
4. If the user chooses overwrite or if no brief.md exists: write directly to brief.md.

Do not register brief.md in the ## Context files table in CLAUDE.md — it is a handoff artifact, not persistent context.

Step 7: Feedback

Store learnings tagged [cc-concept:campaign-concept] in .claude/learnings.md. Examples:

[cc-concept:campaign-concept] client rejected the inferred "lead-generation" type for a customer-only campaign; retention fits their base better — 2026-07-18
[cc-concept:campaign-concept] client prefers email + LinkedIn as primary channels, treats paid social as a last resort — 2026-07-18

If the session revealed a campaign-type preference, framework override, or project constraint that future runs should apply, record it now.