TBM 421: Minimally Viable Consistency (Part 3)

John Cutler · rss · 2026-05-09 Reader 🖥️

Coordination Cost Learning Culture & Slack Organizational Legibility

Organizational consistency is not a single lever but a portfolio of three distinct strategies — sharp consistency (opinionated uniformity), flexible consistency (shared intent with local adaptation), and legible variety (named, visible differences) — and choosing the right mix at each point is the central design problem, one that AI can help with by translating between frames and scaling coaching, but only if the sensemaking friction it removes was overhead and not a learning mechanism.

practical · Newsletter essay — conceptual framework with concrete examples, positioned as Part 3 in a series. Genre: organizational design / product management thought-piece with an AI application coda. — Blends taxonomy-building (sharp/flexible/legible-variety) with prescriptive guidance and reflective questions. The AI section functions as an applied extension of the core framework.

Structure

Part	Role
Why consistency / why minimally viable consistency	Recaps the series premise: consistency reduces coordination costs but has diminishing returns and escalating costs, so you want the minimum that captures the benefits.
Three approaches: sharp, flexible, legible variety	Core contribution — introduces the taxonomy with definitions, use-criteria, and concrete examples for each strategy.
A basic portfolio	Shows the three strategies composed together in a single org, demonstrating they are complementary rather than competing.
Other frames (language agreement, CS analogy)	Re-describes the taxonomy through two alternative lenses (what words mean; schema/interface/tagged-union) to deepen understanding and reveal risks.
The AI opportunity (Jobs 1, 2, 4)	Applies the framework to AI adoption, identifying three jobs AI can do (translator, multi-frame enabler, consistent coach) with paired risks for each.
Questions to consider	Diagnostic prompts for the reader to apply the framework to their own organization.

Key terms

Propositions

explicitConsistency reduces coordination costs, enables cross-boundary flow, makes the system legible above the team level, supports faster feedback loops, and lowers cognitive load.conf: high
explicitConsistency has diminishing returns and escalating costs; each additional constraint adds governance burden and reduces local fit.conf: high
explicitNot all consistency works the same way — there are at least three structurally different strategies (sharp, flexible, legible variety).
refinesHow Fast-Growing Companies Can Make Better Decisions · p4 — Cutler shows the curated-options menu is only one of at least three structurally different consistency strategies (sharp, flexible, legible variety), qualifying the claim that it resolves the centralization trade-off.
explicitThe three strategies are complementary and should be composed as a portfolio, not selected as mutually exclusive alternatives.
explicitSharp consistency works when the value is in the sameness itself and the opinion embedded in the constraint matters; it is analogous to a strict schema.
explicitFlexible consistency works when the value is in shared purpose, not shared form; it is analogous to an interface or protocol.
explicitLegible variety works when work is structurally different across groups and people need to orient quickly; it is analogous to a tagged union.
explicitAI can serve as a translator between inconsistent local contexts, reducing the need to force premature consistency for legibility.conf: low
explicitAI can enable multiple concurrent operating frames (finance view, PM view, engineering view) without humans manually keeping them in sync.conf: low
explicitAI can scale coaching by providing consistent guidance adapted to local context, extending human coaches rather than replacing them.conf: low
explicitThe friction of manual translation and synchronization between frames is not just overhead — it is also a learning mechanism that builds organizational sensemaking.conf: high
explicitConsistency acts as a scaffold for learning: shared structure helps people new to a role build intuition faster and creates a common language for coaching.conf: high
explicitOver-consistency generates performative compliance — people fill in the fields but the data means nothing.conf: high

Arguments

Organizations should treat consistency as a portfolio of three strategies rather than a single dial

AI can reduce the cost of maintaining organizational consistency — but only if deployed with awareness of what friction serves learning versus what friction is pure overhead

Each consistency strategy has a characteristic failure mode that emerges when the strategy is over-applied or under-maintained

Narratives

→ Opening anchor that motivates the entire series and frames consistency as a design tradeoff, not a default good.
→ Concrete instantiation of sharp consistency that shows the opinion-as-feature characteristic.
→ Concrete instantiation of flexible consistency showing shared intent with local adaptation.
→ Concrete instantiation of legible variety showing named, navigable differences.
→ Makes the AI opportunity concrete while setting up the paired risk (losing the sensemaking that happens in messy direct conversations).

Questions addressed

✅ What are the structurally different ways an organization can implement consistency?
Three strategies: sharp consistency (opinionated uniformity), flexible consistency (shared intent with local form), and legible variety (named, visible differences).
⭕ How do you choose between enforcing sameness, enforcing shared intent, and naming variety?
Use-criteria are given (value in sameness vs. shared purpose vs. structural difference) but no systematic decision method is provided.
✅ What is the failure mode of each consistency strategy?
Sharp: enforced past relevance, causing workarounds. Flexible: erosion of shared meaning. Legible variety: complexity explosion as types accumulate.
⭕ Where and how can AI help with organizational consistency without destroying the learning that friction provides?
Three AI jobs identified (translator, multi-frame enabler, coach) with paired risks, but no heuristic for distinguishing productive friction from overhead.
⭕ How much consistency is enough, and how does context (enterprise vs. scaleup) change the answer?
Acknowledged as context-dependent but not operationalized beyond noting that more dependencies and regulation require more consistency.

Gaps

incompleteThe essay provides no method for diagnosing which strategy to apply where — the questions at the end are reflective prompts, not a decision framework. How do you determine whether a given consistency point should be sharp, flexible, or legible variety?
incompleteJob 3 is missing from the AI section (jumps from Job 2 to Job 4) — either an editing artifact or a deliberate omission, but it is unexplained.
incompleteThe taxonomy is presented as exhaustive (three strategies) but no argument is given for why these three are sufficient. Could there be a fourth strategy (e.g., emergent consistency that arises from shared tooling without explicit design)?
incompleteThe AI risk analysis identifies that friction can be a learning mechanism but offers no heuristic for distinguishing productive friction from pure overhead — the central practical question for anyone applying the framework to AI adoption.

Closes gaps elsewhere

How Fast-Growing Companies Can Make Better Decisions · g4 — Provides strategies for decision spaces that resist curated menus: flexible consistency (shared purpose/protocol, not shared form) and legible variety (making structurally different local work legible) cover novel or unbounded domains where no pre-curated option set fits.