Agent Skills: Crossing the Chasm (Moore) — the v1.0 Spine

Crossing the Chasm (Moore) — the v1.0 Spine

Position in workflow: v0.9 Feedback Loop Setup → v1.0 Crossing the Chasm (Moore) → all v1.0 work

Execution Mode

Default is deep (this is a major strategic decision; quick mode is for hypothesis pre-work only). See .claude/rules/08-skill-execution-modes.md.

| Mode | What this skill produces | |------|--------------------------| | quick | Stage assessment only (ADO-STAGE-*); beachhead candidate hypothesis | | standard | Stage assessment + beachhead segment + top 3 whole-product gaps + reference-account candidate list | | deep (default) | Full stage assessment with evidence + sharpened beachhead with in/not-in criteria + complete whole-product gap analysis + reference-account cultivation plan + chasm-crossing risk register |

Framework: Moore's Technology Adoption Lifecycle

From Crossing the Chasm (Geoffrey Moore, 1991, updated 2014). The bestselling tech-strategy book of the 1990s, and still the canonical model for understanding why early traction doesn't predict mass adoption.

The lifecycle

| Stage | % of market | Buyer mindset | What they buy | |-------|-------------|---------------|----------------| | Innovators (2.5%) | Tinkerers; technology enthusiasts | Want to try new things; tolerate incomplete products | Vision, technical depth, access | | Early Adopters (13.5%) | Visionaries | Want strategic advantage from non-mainstream tech | Bold vision + first-mover ROI | | — THE CHASM — | — | — | — | | Early Majority (34%) | Pragmatists | Want reliable productivity gains from proven solutions | Whole product + segment-specific references | | Late Majority (34%) | Conservatives | Want safe, mature defaults | Market leadership, low risk | | Laggards (16%) | Skeptics | Resist change | (Generally not worth targeting) |

The chasm

The gap between Early Adopters and Early Majority is the chasm. Most products die here. The reason: visionary buyers (who got you to early traction) actively want the cutting-edge, while pragmatist buyers want to be the second penguin off the iceberg — they need to see peers in their segment succeeding first.

The chasm-crossing playbook

1. Pick a beachhead segment — A single, narrowly-defined sub-segment of the early majority. Not "small businesses." Specific: "freight-forwarding firms with 50-200 employees in the US Pacific Northwest using ERP X."
2. Build the whole product for that beachhead — Pragmatists don't buy your core product; they buy your core + integrations + reference customers + support + training + everything else they need to deploy. Identify and close the gaps.
3. Cultivate references in segment — A reference from outside the beachhead is worthless to a beachhead buyer. Three references inside the beachhead is the unlock.
4. Concentrate, then expand — Don't try to cross the chasm broadly. Win the beachhead, then use it as a reference base to win adjacent segments ("bowling alley" stage).

Consumes

• CFD-* customer evidence (all stages, especially v0.9 post-launch feedback) — Composition of paying customers; stage signal
• PER-* personas (sharpened by v0.9 Positioning) — Beachhead is a sharper variant of an existing PER-
• GTM-* positioning (from v0.9 Positioning) — Best-fit segment is the starting point for beachhead selection; beachhead is usually tighter
• GTM-* offer card (from v0.9 Offer Construction) — Offer must match beachhead expectations (pragmatists need different guarantees than visionaries)
• FEA-* features (from v0.3) — Whole-product gap analysis cross-references current feature set
• KPI-* baseline metrics (from v0.3 + v0.9) — Stage assessment uses retention shape, NPS by segment, conversion rate
• BR-PRICING-* (from v0.3 + v0.9) — Pragmatist buyers expect different pricing/contract terms than visionaries

Produces

• ADO-STAGE-* entries in SoT/SoT.ADOPTION.md — Current adoption-stage assessment with evidence
• ADO-BEACHHEAD-* entries — Beachhead segment definition with strict in/not-in criteria
• ADO-WHOLE-* entries — Whole-product gaps (one per gap, with owner and close date)
• ADO-REF-* entries — Reference-account targets (one per candidate, with consent + placement status)
• CFD-* gaps surfaced — When stage evidence is thin or beachhead selection lacks data, log as research gaps

Execution

Step 1: Assess current adoption stage

Audit current paying-customer composition. For each customer, classify by stage indicators:

| Indicator | Innovator/Early Adopter | Early Majority | |-----------|------------------------|-----------------| | Sales motion | Direct founder relationship | Asked for demo / pricing / case studies | | Setup tolerance | "I'll figure it out" | "Show me the integration with X" | | Reference need | None | Asked "who else in my industry uses this?" | | Renewal motivation | "We're betting on this" | "It saves us $X/month" | | Procurement | Single buyer | Procurement / SOC2 / contracts review |

Stage signal:

• Innovators / Early Adopters: >70% of paid customers show visionary indicators
• At the chasm: Mixed composition; new inbound asking pragmatist questions ("references", "integrations", "SOC2"); flat MRR despite growing leads
• Bowling Alley (post-chasm): >50% of new customers in one identifiable segment
• Tornado: Rapid acquisition across multiple segments
• Main Street: Stable growth, focus on expansion and retention

Deliverable: One ADO-STAGE-* entry with evidence, stage classification, and confidence score.

Step 2: Pick the beachhead segment [standard+]

Generate 3–5 candidate beachhead segments. Score each:

| Criterion | Question | Weight | |-----------|----------|--------| | Pragmatist density | Is this segment mostly pragmatists (already buying mature solutions)? | High | | Whole-product proximity | How close is our current product to the segment's whole-product expectation? | High | | Reference accessibility | Can we get 3 references in this segment within 6 months? | High | | Compelling reason to buy | Is there an urgent, segment-specific pain we solve uniquely? | High | | Adjacency value | If we win this segment, what adjacent segments unlock? | Medium | | Competitive intensity | How saturated is this segment with established competitors? | Medium | | Market size | Is this segment large enough to support a beachhead? | Low (most beachheads are small; that's fine) |

Pick the top-scoring segment. Write strict in-segment and not in-segment criteria.

Deliverable: One ADO-BEACHHEAD-* entry with in/not-in criteria, rationale, target (e.g., "10 closed-won in segment within 6 months"), confidence.

Step 3: Whole-product gap analysis

For the beachhead segment, list everything a pragmatist buyer in that segment expects to receive when they pay you:

• Core product (what you ship)
• Integrations (with their existing stack — specific)
• Compliance (SOC2, HIPAA, ISO — segment-specific)
• Support (response-time SLA, dedicated CSM if appropriate)
• Training / onboarding (videos, docs, certifications)
• References (segment-specific, named, willing to talk)
• Pricing / contract terms (annual contracts, MSAs, custom DPAs)
• Migration / data import
• Professional services (implementation help)
• Roadmap visibility

For each, score: Ship today? / Gap?

For each gap, create an ADO-WHOLE-* entry with severity (blocker / serious / nice-to-have), owner, target close date.

Deliverable: List of ADO-WHOLE-* entries. Blockers must close before sustained beachhead motion.

Step 4: Reference-account cultivation plan [standard+]

Identify 5–10 existing or near-term customers in the beachhead segment that could become public references. For each:

| Field | Notes | |-------|-------| | Customer | Name + segment fit confirmation | | Story strength | What outcome can they speak to publicly? (Quantified > qualitative) | | Relationship | Who owns the relationship internally? | | Consent path | What approval do they need internally to be public? | | Target placement | Logo on pricing page / quote on landing / blog case study / on-stage / podcast | | Confidence | 1–5 of "will become a reference within 90 days" |

Deliverable: 5–10 ADO-REF-* entries with cultivation plan.

Step 5: Chasm-crossing risk register [deep only]

For each major chasm risk, log a RISK-* entry:

• Reference-cold risk — Can't get 3 references in beachhead within 6 months
• Whole-product gap risk — Critical gap can't be closed in time
• Beachhead-too-small risk — Segment doesn't sustain the company economically even if won
• Competitor-incumbency risk — Pragmatist defaults are competitor X; switching cost too high
• Internal motion risk — Sales / support / product can't pivot from visionary motion to pragmatist motion

Each risk gets mitigation actions tied to ADO-WHOLE-, ADO-REF-, or BR-* updates.

Deliverable (deep): Risk register with mitigations.

Output Templates

See SoT/SoT.ADOPTION.md for the complete entry templates for ADO-STAGE-, ADO-BEACHHEAD-, ADO-WHOLE-, and ADO-REF-.

Anti-Patterns

| Pattern | Signal | Fix | |---------|--------|-----| | "We're in the chasm" without evidence | Stage assessment based on vibes | Audit actual paying-customer composition; require ADO-STAGE-* confidence ≥ 3/5 | | Beachhead too broad | "Small businesses" or "B2B SaaS" | Tighten until it disqualifies most buyers; segment of 100–10,000 prospects max | | Skipping whole-product gap | "Our product is great; we just need more marketing" | Pragmatists buy whole product, not core product; list every gap | | Reference from wrong segment | Touting a visionary customer to a pragmatist buyer | In-segment references only; cross-segment is worthless | | Crossing the chasm broadly | "Let's just scale paid ads" | Concentrate on beachhead; broad CAC will be 5× higher and worse-converting | | Confusing best-fit with beachhead | Treating Dunford best-fit as the beachhead | Beachhead is tighter than best-fit; usually one sub-segment of the best-fit | | Skipping stage assessment | Jumping to "we need to scale" without checking where we are | Step 1 is mandatory; without it, the rest is guesswork |

Quality Gates

Before proceeding to other v1.0 work:

• [ ] ADO-STAGE-* entry exists with evidence and confidence ≥ 3/5
• [ ] If stage is "at the chasm" or beyond: ADO-BEACHHEAD-* defined with strict in/not-in criteria
• [ ] At least 3 ADO-WHOLE-* gap entries identified (blockers + serious)
• [ ] At least 3 ADO-REF-* candidate accounts (standard+)
• [ ] RISK-* entries cover the top chasm risks (deep only)
• [ ] Beachhead does not contradict v0.9 Positioning's best-fit (or contradiction is explicit and rationalized)

Downstream Connections

| Consumer | What it uses | Example | |----------|--------------|---------| | Continuous Discovery (Torres) | Beachhead segment = discovery interview pool | Weekly interviews drawn from ADO-BEACHHEAD- | | Mom Test Interview | Beachhead-segment interview discipline | Validate ADO-STAGE- and ADO-WHOLE- via Mom Test | | Case Study Builder | Reference candidates = case study targets | ADO-REF-* graduates to case study | | Testimonial Collector | Reference candidates = testimonial targets | ADO-REF-* (lower-effort placement) | | v0.9 Re-runs | Beachhead may sharpen Positioning; whole-product gaps may shift Offer | Re-run Dunford with sharper segment | | Feedback Loop Setup | Pragmatist-shaped feedback signals chasm crossing | Inbound pattern shift → re-run stage assessment |

Detailed References

• Geoffrey Moore, Crossing the Chasm (1991, revised 2014) — canonical source
• Geoffrey Moore, Inside the Tornado (1995) — post-chasm bowling alley / tornado / main street
• Eric Ries, The Lean Startup (incompatible motion before chasm; useful contrast)
• wondelai's crossing-the-chasm skill (wondelai/skills)
• (No bundled references/ — read the book for depth)