lp-offer — Startup Offer Design (MARKET-06)

Produces a comprehensive offer artifact for startups. Defines ICP, pain/promise mapping, offer structure, positioning, pricing hypothesis, and objection handling in one document.

Invocation

/lp-offer --business <BIZ>

Required:

• --business <BIZ> — business identifier (e.g., BRIK, SEG, INT)

Business resolution pre-flight: If --business is absent or the directory docs/business-os/strategy/<BIZ>/ does not exist, apply _shared/business-resolution.md before any other step.

Operating Mode

RESEARCH + DESIGN + DOCUMENT

This skill:

• Researches competitor offerings, pricing, and customer reviews
• Designs a structured offer hypothesis with 6 sections
• Documents the offer artifact at docs/business-os/startup-baselines/<BIZ>/offer.md
• Does NOT make final pricing decisions (produces hypothesis to validate)
• Does NOT skip sections (all 6 MUST be present)

Inputs

Required:

• lp-readiness output (offer clarity gaps from RG-01 gate)
• Business context from docs/business-os/strategy/<BIZ>/ or user-provided description
• What the business does, what it sells, who it targets

Optional (enhances quality):

• Competitor websites, pricing pages, product listings
• Customer reviews (competitor products/services)
• Market research docs from docs/business-os/market-research/
• Existing offer drafts or positioning statements
• docs/business-os/strategy/<BIZ>/<YYYY-MM-DD>-brand-identity-dossier.user.md — enhances Positioning One-Pager (Section 4) voice and personality

Workflow

Stage 1: Load Context (RESEARCH)

1. Read lp-readiness output to understand offer clarity gaps
2. Load business context from strategy docs
3. Identify 3-5 direct competitors
4. Dispatch ONE subagent per competitor in a SINGLE message (parallel). Each subagent receives .claude/skills/lp-offer/competitor-research-brief.md as its brief. Subagents run read-only Mode A (analysis phase only — no file writes). Protocol: _shared/subagent-dispatch-contract.md.
   • Hard limit: 200 words max per subagent output
   • If a subagent returns > 200 words: orchestrator truncates to structured schema fields, discards prose before synthesis
   • If a subagent returns status: fail: quarantine result, continue with remaining, flag gap in Evidence Register
5. Merge all structured extracts into the Evidence Register. Then proceed to Stage 2.

Stage 2: Build Offer Artifact (DESIGN)

Create artifact with 6 mandatory sections:

Section 1: ICP Segmentation

Define 1-2 Ideal Customer Profiles (ICPs). Each ICP includes:

• Demographics: Age, location, income, job title, company size (B2B)
• Psychographics: Values, fears, aspirations, buying style
• Jobs-to-be-done: What functional/emotional/social job does the offer solve?
• Buying triggers: What event/pain/threshold prompts them to buy NOW?

Specificity test: If "everyone" qualifies, the ICP is too broad. Narrow it.

Section 2: Pain/Promise Mapping

For each ICP, map top 3 pains to specific promises:

Review mining guidance:

• Search competitor reviews for: "I wish...", "frustrating", "difficult", "doesn't work", "missing"
• Extract verbatim customer language (use their words, not marketing speak)
• Note patterns: if 5+ reviews mention same pain, it's a top priority

Section 3: Offer Structure

Define what the customer receives:

• Core offer: The main product/service (be specific: "12-week coaching program" not "coaching")
• Bundles: Any package variants (basic/pro/enterprise)
• Included deliverables: What tangible/intangible items are included?
• Exclusions: What's NOT included (set boundaries)
• Guarantees: Money-back, satisfaction, performance guarantees
• Risk reversals: Free trial, sample, audit, assessment, no-commit first session

Concreteness test: Could a competitor easily copy this structure? If not specific enough, add detail.

Section 4: Positioning One-Pager

Positioning statement using Geoffrey Moore's template:

For [target customer segment]
Who [have this specific need/pain]
[Product/service name] is [category]
That [key differentiated benefit]
Unlike [primary alternative/competitor]
Because [reason to believe / unique capability]

Additional elements:

• Category creation: Are we defining a new category or claiming position in existing one?
• Competitive frame: What do customers compare us to? (Not always direct competitors)
• Key message: One-sentence elevator pitch derived from positioning statement

Section 5: Pricing/Packaging Hypothesis

THIS IS A HYPOTHESIS TO VALIDATE, NOT A FINAL PRICE.

Pricing analysis:

• Proposed price point(s): Primary offer price + any tier variants
• Pricing model: One-time, subscription, usage-based, value-based, freemium
• Competitor price comparison: List 3-5 competitors with their pricing (create comparison table)
• Anchor pricing: What higher-priced alternative sets the reference point?
• Price positioning: Premium, mid-market, value/budget?
• Rationale: Why this price? (Cost-plus, value-based, competitive parity, penetration, skimming)

Confidence level: State LOW/MEDIUM/HIGH confidence and what evidence is missing.

Validation plan: What experiments will test pricing hypothesis? (Landing page test, pre-sales, tiered offers, etc.)

Section 6: Objection Map + Risk Reversal

Top 5 objections with responses:

Sources for objections:

• Competitor reviews (what customers complain about)
• Sales conversations (if available)
• Price sensitivity ("too expensive")
• Credibility gaps ("never heard of you")
• Comparison anxiety ("how do I know you're better than X?")

Risk reversal mechanisms:

• Money-back guarantee (30/60/90 day)
• Free trial or freemium tier
• Performance guarantee ("results or refund")
• Satisfaction guarantee ("love it or leave")
• Pilot/sample/audit (B2B)
• No-contract/cancel-anytime (subscription)

Stage 3: Document Artifact (DOCUMENT)

1. Write complete offer artifact to docs/business-os/startup-baselines/<BIZ>/offer.md
2. Include metadata header:
   yamlCopy

   ---
   business: <BIZ>
   artifact: offer-design
   created: <date>
   status: hypothesis
   confidence: [low/medium/high]
   ---
   

3. Append evidence register at end: list all sources consulted (competitor URLs, review sites, research docs)
4. Self-audit against Quality Checks (see below)

Output Contract

Produces single file: docs/business-os/startup-baselines/<BIZ>/offer.md

Artifact registry: Canonical path defined in docs/business-os/startup-loop/artifact-registry.md (artifact ID: offer).

Structure:

1. Metadata frontmatter (YAML)
2. Executive summary (2-3 sentences: what is the offer, who is it for, what makes it different)
3. Section 1: ICP Segmentation (1-2 ICPs with all 4 elements)
4. Section 2: Pain/Promise Mapping (table format, 3 pains per ICP)
5. Section 3: Offer Structure (core offer, bundles, guarantees, risk reversals)
6. Section 4: Positioning One-Pager (Moore template filled, category/frame/message)
7. Section 5: Pricing/Packaging Hypothesis (price points, competitor comparison table, rationale, confidence, validation plan)
8. Section 6: Objection Map + Risk Reversal (table format, 5 objections)
9. Evidence Register (sources cited with URLs or file paths)

Downstream compatibility:

• lp-forecast consumes: ICP (for TAM/SAM estimation), pricing hypothesis (for revenue modeling), confidence level (for scenario ranges)
• lp-channels consumes: ICP (for channel-customer fit), positioning (for messaging), objections (for content strategy)
• lp-do-critique (DS-14 offer mode) consumes: entire artifact for critique

Quality Checks

Before finalizing, verify:

• QC-01: All 6 sections present and complete (ICP, pain/promise, structure, positioning, pricing, objections)
• QC-02: ICP is specific (not "everyone" or "small businesses")
• QC-03: Pricing hypothesis includes competitor comparison (≥3 competitors with prices)
• QC-04: At least 3 objections mapped with responses
• QC-05: At least 1 risk reversal mechanism defined
• QC-06: Confidence level stated explicitly (low/medium/high)
• QC-07: Evidence register cites ≥3 sources (competitor sites, reviews, or research docs)
• QC-08: Pain/promise mapping uses customer language from reviews/research (not marketing speak)
• QC-09: Positioning statement fills all 6 blanks in Moore template
• QC-10: Pricing rationale explains WHY this price (not just "seems right")

Red Flags

Invalid outputs that MUST be rejected:

• Missing any of the 6 sections
• ICP says "everyone", "anyone who needs X", "small to medium businesses" (too broad)
• No competitor pricing cited (makes pricing hypothesis ungrounded)
• Pricing shows final decision confidence ("this IS the price") instead of hypothesis to validate
• Pain/promise mapping uses generic pains ("wants better results") not specific customer language
• Positioning statement has blanks or says "TBD"
• <3 objections mapped
• No risk reversal mechanism (no guarantee, trial, or reversal)
• Evidence register empty or cites non-existent files
• Confidence level missing

Integration

Upstream (MARKET-01 + readiness context)

• Preceded by /lp-readiness --business <BIZ> (must pass RG-01 offer clarity gate)
• Consumes offer clarity gaps from readiness output
• Reads business context from docs/business-os/strategy/<BIZ>/

Downstream (S3, SELL-01)

• /lp-forecast --business <BIZ> (S3) reads ICP and pricing hypothesis for revenue modeling
• /lp-channels --business <BIZ> (SELL-01) reads ICP, positioning, and objections for channel selection
• lp-do-critique (DS-14 offer mode) critiques this artifact for logical gaps

Parallel Skills

• Can run after or parallel to /lp-market (market research baseline) if it exists
• Does NOT depend on /lp-forecast or /lp-channels (this skill is prerequisite for those)