bmc-canvas-diagnostic-reading

Reads a finished BMC canvas as a diagnostic instrument using seven research-based rules (Hook Rule, value-cost consistency, segment specificity, evidence grade, defensive canvas, missing block, gravit

SKILL.md

GitHub

BMC Canvas Diagnostic Reading

Purpose

Provides a systematic way to read a finished (or nearly finished) BMC canvas as a diagnostic instrument — not just a checklist of filled blocks, but six questions about whether the canvas is internally consistent, honest about evidence, and ready for the hypothesis-testing phase. Use this ONCE the canvas (or a variant) is ready for assessment — not mid-session while it's still being built (for that, see ../bmc-session-facilitation-design/SKILL.md and ../bmc-canvas-clarity-and-iteration/SKILL.md).

Anchored in research

Based on the owner's private research layer (an April 2026 expert interview) (30_domain_packs/bmc/diagnostics/bmc_diagnostic_rules.md, canvas_quality_rubric.md, antipatterns/counterexamples.md — all status: template, source: research_layer), a synthesis drawing on van der Linden, Jeffries, Williams, and general facilitation practice. [EXPERT INPUT] sections are not yet filled in. DR-07 (added later) is additionally grounded in Osterwalder & Pigneur's The Invincible Company (2020) "Gravity Creators" pattern — see ../../references/bmc-resilience-heuristics-research.md for why this became an added rule on an existing skill rather than a new standalone one.

Note: This is different from ../bmc-canvas-clarity-and-iteration/SKILL.md's owner-validated "clarity > depth" readiness criterion. That criterion is deliberately simpler and more opinionated ("does the canvas tell one thing clearly?"). This skill's six rules are more precise and analytical ("is the canvas internally consistent and evidence-based?"). Use both — they answer different questions.

Method

  1. DR-01 — Hook Rule. Every element on the right side needs a matching hook on the left — and vice versa. If customer relationships is defined as "dedicated personal service," key resources needs senior customer-facing staff and key activities needs relationship management. Trace the connections between related elements — orphan elements (without hooks) are assumption gaps. This mirrors a core consistency check in Osterwalder and Pigneur's original BMC methodology: the nine blocks aren't independent fields, they're a connected system, and a block that doesn't visibly connect to anything else is a warning sign.
  2. DR-02 — Value-cost consistency. If the value proposition promises premium quality, uniqueness, or a high-touch service, the cost structure has to reflect that. A "best quality" + "cost leadership" combination is internally contradictory. A contradiction doesn't mean the business is wrong — it means the team hasn't resolved a fundamental strategic tension. Ask: "can we deliver this value at this cost?"
  3. DR-03 — Segment specificity test. A segment is too broad when: it's defined only by demographics ("people aged 25-45"); it could include any competitor's any customer ("SMEs"); it lumps several distinct buying logics into one group; it includes both the user and the payer without distinguishing them. A segment is specific enough when a team member could walk outside and identify five specific people who fit it, and explain why those five share the same job, pain, and gain — a test in the spirit of Jobs-to-be-Done thinking (van der Linden, Jeffries) applied directly to segment definition.
  4. DR-04 — Evidence grade check. Read every block and classify it: proven (tested with real customers/data), weakly proven (an internal assumption or secondhand information), assumed (an untested belief). The blocks with the most red are the testing priority. A canvas that's entirely green either describes a genuinely mature business model or hasn't been honest about what's actually known.
  5. DR-05 — Defensive canvas signal. When a team fills in the canvas quickly, confidently, and without debate, two things are possible: (1) the model is genuinely mature and well understood (rare), or (2) the team is projecting certainty instead of examining it (common). Signals: no block was debated or changed during the session; every element is positive (no tensions or trade-offs named); the team refers to the canvas as "our story" rather than "our assumptions"; customer segments are presented as facts, not hypotheses.
  6. DR-06 — A missing block tells you something. When a block is left empty or marked "TBD," this is diagnostic information, not a formatting error. An empty revenue streams block means the team hasn't resolved how value gets captured. An empty key partners block usually means dependencies haven't been thought through yet. The most diagnostic empty block: customer relationships — teams often skip it by conflating it with channels, which reveals that the nature of the interaction (transactional vs. relational, high-touch vs. automated) hasn't been thought through yet.
  7. DR-07 — Gravity/lock-in check. Ask directly: once a customer is won, is there anything about this model that makes leaving meaningfully harder or less attractive than arriving was — a switching cost, a data-portability friction, an ecosystem effect, accumulated personal investment (the Apple iPod/iTunes pattern)? Or is the model a wide-open door, where a competitor one feature ahead can take the customer with no friction at all? A canvas can pass every other rule and still be fragile for this reason alone — a model with no gravity is vulnerable to exactly the kind of disruption that looks irrelevant right up until it isn't (the canonical case: the iPod's absence of any real gravity once Spotify removed the need to "own" a music library at all). This check corresponds to two patterns already in this pack's library — operating.chain.lock_in and experience.relationships.switching_costs in ../../references/bmc-innovation-pattern-library.md — use ../bmc-innovation-pattern-matching/SKILL.md to find or strengthen a gravity mechanism if this check comes back negative, rather than inventing one ad hoc.
  8. Use the four-dimension rubric for a final score (each 1-5): segment specificity, value proposition quality, internal consistency, evidence honesty. Interpretation: 16-20 = strong canvas, ready for the testing phase. 11-15 = workable, needs fixes before testing. 6-10 = significant rework needed. Below 6 = start over. (DR-07 is a pass/fail gravity check, not part of this four-dimension score — a canvas can score well here and still fail DR-07, and that combination is itself a useful, specific finding to report: "well-built but structurally easy to leave.")
  9. Watch for two counterexamples that look like good work but aren't: (C-01) a specific but still wrong segment — specificity doesn't guarantee validity if the segment is too small, internally heterogeneous, or limited to the founder's own network; test: can the team name three competitors chasing the same group? (C-02) an honest but still incomplete canvas — honesty about uncertainty doesn't mean the RIGHT uncertainties have been identified; test: when asked "what would kill this business model?", do the risks named match anything visible on the canvas?

What this skill does NOT do

  • Doesn't contain the owner's own take on the most common hook error, value-cost contradiction, or segment specificity test seen in their own practice — these remain open as [EXPERT INPUT] sections in the source files.
  • Doesn't replace ../bmc-canvas-clarity-and-iteration/SKILL.md's simpler, owner-validated readiness criterion — use both alongside each other, not one instead of the other.
  • Doesn't make the final proceed/return decision for you — it provides scoring and diagnostics; the decision to move to the testing phase is a human one.

Refinement notes

  • What's the most common hook error (DR-01) you find in practice?
  • What's the most common value-cost contradiction (DR-02) you see, and how do you raise it without triggering defensiveness?
  • What's your own segment specificity test (DR-03)?
  • How do you run the evidence grade check (DR-04) live in a session? Do clients push back on being marked red?
  • How do you tell genuine confidence apart from defensive certainty (DR-05)?
  • Which empty block (DR-06) tells you the most, in your experience?
  • How does your own quality assessment work — which dimension weighs the most?
  • What's the clearest real case where DR-07 caught a fragile model that otherwise looked strong on every other rule?

Continue from here

  • Previous skill in the same pack: ../bmc-canvas-clarity-and-iteration/SKILL.md — the owner's own, lighter-weight readiness criterion, use first.
  • Next skill in the same pack: ../bmc-tool-switching-decisions/SKILL.md — when the diagnostics reveal that the BMC is no longer the right tool.
  • This pack's shared guardrails: ../../CLAUDE.md

References

  • ../../references/bmc-source-material-notes.md — source material background
  • ../../references/bmc-resilience-heuristics-research.md — grounding and rationale for DR-07
  • ../../references/bmc-innovation-pattern-library.md — the operating.chain.lock_in / experience.relationships.switching_costs patterns DR-07 checks against
  • ../../CLAUDE.md — this pack's shared guardrails