Assess Strategic Environment

Assess Strategic Environment

For a defined industry, diagnose the competitive environment type using BCG's Strategy Palette. Output: strategic-environment.md naming the environment in one causal sentence, evidencing the three diagnostic dimensions, flagging ambidexterity where present, and emitting a routing matrix that downstream sub-skills consume.

The discipline: Wrong environment diagnosis = wrong toolkit = useless output. A Classical analysis of an Adaptive industry produces a five-year plan that won't survive contact with reality. An Adaptive analysis of a Classical industry produces overcautious wait-and-see recommendations. This skill runs FIRST in the orchestrator chain.

Iron rules:

• Every dimension assessment carries V/C/A/I-tagged evidence — see ../_shared/provenance-tagging.md.
• The classification named in one causal sentence ("Industrial robotics is Adaptive because trajectory is unforecastable beyond 18 months [C: ...], no single player can shape the standards landscape [C: ...], and demand is not in survival-mode [V: ...].").
• Three diagnostic dimensions independently assessed: predictability, malleability, harshness — none skipped, none collapsed.
• Routing matrix included — names Phase 1+2 sub-skills with heavy / light / skip designation per the diagnosed environment.
• 2026 terminology only — see ../_shared/2026-terminology.md.
• Ambidexterity flagged when the industry genuinely spans two environments (different layers, geos, or customer segments).

Process

1. Intake

Confirm: industry slug, geographic scope (if relevant — the same industry can sit in different environments in different geos), focal layer (optional but recommended for multi-layer industries — see Five Forces focal-layer discipline). Read references/strategy-palette.md for the framework refresher and references/diagnostic-questions.md for the three-question discipline with per-environment anchors.

2. Assess predictability

Can the trajectory be reliably forecast over a 3-5 year horizon? Rate High / Medium / Low. Evidence anchors (see references/diagnostic-questions.md):

• Demand volatility (5yr coefficient of variation)
• Technology stability vs disruption cadence
• Regulatory cliffs / known mandates
• Number of viable strategic scenarios on a 5yr view

Write 2-3 sentences with ≥1 V/C/A/I-tagged evidence claim. Never assert High predictability for a market with active AI-native disruption.

3. Assess malleability

Can a single player (or coalition) materially shape the industry's structure or rules? Rate High / Medium / Low. Evidence anchors:

• Existence of platform dynamics / network effects
• Concentration — is there an actor with >25% share or de-facto standards power?
• Regulatory openness — is the rulebook negotiable?
• Active shaping moves visible (consortia, standards bodies, ecosystem investments)

Write 2-3 sentences with ≥1 V/C/A/I-tagged evidence claim. Default to Low unless there is named evidence of shaping power — Visionary and Shaping classifications require positive evidence.

4. Assess harshness

Is the industry in a survival-mode constrained environment (capital scarcity, regulatory cliff, demand collapse, cost-curve collapse)? Rate High / Medium / Low. Evidence anchors:

• Sustained negative growth (-3pp or worse, 2+ years)
• Capital markets closed to the sector (no funding rounds, debt repricings, bankruptcy waves)
• Existential regulatory event (ban, mandated wind-down)
• Cost-base structurally above price (margins below cost-of-capital across the sector)

Write 2-3 sentences with ≥1 V/C/A/I-tagged evidence claim. Harshness is rare — Renewal should be called only with named evidence of at least one of these triggers.

5. Classify the environment

Apply the decision rule from references/strategy-palette.md:

If the dimensions don't map cleanly, see references/diagnostic-questions.md "Edge cases" — typically the resolution is to declare ambidexterity (step 6) or restate the focal layer.

Write the classification in ONE causal sentence naming all three dimensions: "Industry X is [Environment] because [predictability evidence], [malleability evidence], and [harshness evidence]."

6. Ambidexterity check

Does the industry genuinely span TWO environments at different layers, geos, or customer segments? If yes, declare it: "Industry X exhibits ambidexterity — the regulated incumbent layer is Classical [evidence]; the agentic-commerce-native layer is Shaping [evidence]." Default is single-environment classification; only flag ambidexterity with named layer/segment evidence. See references/diagnostic-questions.md "Ambidexterity discipline."

Orchestrator-mode signal: when invoked by analyze-industry, ambidexterity in the output triggers the orchestrator's Ambidexterity Checkpoint (a hard stop where the user chooses single-focus / dual analysis / re-scope). Do NOT presume which choice the user wants — write the ambidexterity declaration clearly enough that the orchestrator's checkpoint prompt can render it directly. See analyze-industry/references/gate-prompts.md "Ambidexterity Checkpoint" for the prompt the orchestrator will use.

7. Direction of travel

Is the environment classification stable, or is the industry transitioning? Examples: Classical → Adaptive (AI disruption breaking long-run predictability); Adaptive → Renewal (capital tightening, demand collapse); Visionary → Shaping (first-mover advantage decaying as ecosystem opens). Write one sentence: "Stable" OR "Transitioning from [X] toward [Y] because [evidence]."

8. Emit routing matrix

Read references/routing-matrix.md. Produce a table mapping the diagnosed environment to weighting designations (heavy / light / skip) for each Phase 1+2 sub-skill: size-market, map-five-forces, map-value-chain-profit-pools, map-competitive-arena, analyze-trajectory, assess-moat-sources, analyze-demand. map-five-forces is never skip — Reeves's contention is that structural analysis is required in every environment.

9. Append structured next_skills YAML block

Because this skill runs FIRST, next_skills: names the Phase 1+2 sub-skills the orchestrator should dispatch next, ordered by weight (heavy-first):

---
next_skills:
  - size-market    # always next (calibrates dollar magnitudes)
  - map-five-forces    # always required (industry structure)
  - map-value-chain-profit-pools    # always required (profit location)
  - [heavy-weight Phase 2 skill per environment]
  - [heavy-weight Phase 2 skill per environment]
---

At minimum: the three Phase 1 skills plus the environment-specific heavy-weight skills from step 8.

10. Validate + write output

Run python scripts/validate_environment.py --output-path <path> — checks: classification = one of 5 named environments; 3 diagnostic dimensions each present with a tagged sentence; routing matrix present with all 7 sub-skills enumerated and weights assigned; tag coverage ≥3; next_skills: YAML block present with ≥3 entries; direction-of-travel statement present; ambidexterity declared OR explicitly noted as absent. Write to working/strategic-environment.md (orchestrator mode) or standalone/assess-strategic-environment-YYYY-MM-DD.md (standalone mode).

HTML on request (standalone only): markdown is the default and the only format the validator and the orchestrator consume. If the user explicitly asks for an HTML version of a standalone run, then after validation passes, also render the output via the html-output skill and review it per ../_shared/output-conventions.md § "HTML deliverables and quality review". Never produce HTML automatically.

Integrates with

• analyze-industry — orchestrator parent. This skill is step 2 of the orchestrator chain; its output routes all downstream sub-skill weighting.
• map-five-forces — always-run downstream. Five Forces consumes the focal-layer declaration from this skill's intake.
• assess-moat-sources (Phase 2) — gets heavy-weight in Visionary and Shaping environments.
• analyze-trajectory (Phase 2) — gets heavy-weight in Adaptive, Visionary, and (transitioning) Renewal environments.

Gotchas

• Symptom: classifies every industry as Classical. Cause: analyst trained pre-2010, defaulting to the familiar. Fix: validator requires the classification be one of the five names; Classical must be supported by High-predictability AND Low-malleability AND Low-harshness with tagged evidence on each.
• Symptom: declares Visionary for any company with a charismatic founder pitching a big vision. Cause: confusing firm-level ambition with industry-level malleability. Fix: Visionary requires industry-level evidence that a single player has shaped (or could shape) the standards / rules — not founder rhetoric. Default malleability to Low unless evidence is named.
• Symptom: declares all three dimensions as "Medium" with no classification. Cause: drafter unwilling to take a position. Fix: the framework requires binary-ish calls on each dimension; "Medium" across the board fails validation. Resolve to a high/low side per dimension, or declare ambidexterity.
• Symptom: assesses predictability only by gut-feel ("seems uncertain"). Cause: skipped the evidence anchors. Fix: references/diagnostic-questions.md lists 4 anchors per dimension; each rating must cite at least one with a tag.
• Symptom: routing matrix absent or generic ("run all the sub-skills"). Cause: drafter treated the routing as decorative. Fix: validator requires the matrix to enumerate all 7 sub-skills with heavy/light/skip designations; map-five-forces must never be skip.
• Symptom: classifies a multi-layer industry (e.g., semis) as one environment. Cause: ignored focal-layer discipline. Fix: either declare focal layer at intake OR declare ambidexterity in step 6 with per-layer classifications.
• Symptom: calls a sector Renewal because growth slowed last quarter. Cause: confusing cyclical slowdown with structural harshness. Fix: Renewal requires sustained (2+ year) evidence of one of the named triggers — capital closed, regulatory cliff, demand collapse, cost > price.
• Symptom: every direction-of-travel says "Stable" across a portfolio of analyses. Cause: drafter treating step 7 as box-ticking. Fix: in 2026, AI is reshaping predictability and malleability across most industries; real industries usually have at least one dimension in motion.

Rules

• Never default to Classical without testing all three dimensions against the named anchors.
• Never declare Visionary or Shaping without positive, tagged evidence of malleability.
• Never declare Renewal without tagged evidence of sustained harshness (one of the named triggers).
• Never produce an environment diagnosis without a routing matrix downstream sub-skills can consume.
• Every dimension assessment carries a V/C/A/I-tagged evidence claim.
• map-five-forces is never weighted "skip" in any environment.
• All file reads use encoding='utf-8'.

Old patterns

None yet — v1 (split out from analyze-industry/references/strategy-palette.md on 2026-05-18 as a standalone Phase 2 sub-skill).