layer-based-business-structuring

Structures a business, service concept, or business model into distinct functional layers (OSI-model-like) from infrastructure to brand, in order to decide which layers to build in-house and which to

SKILL.md

GitHub

Layer-Based Business Structuring

Purpose

Structures a business, service concept, or business model into distinct functional layers — analogous to the telecommunications OSI model — from the physical/infrastructure layer up to the brand layer, or, from a commerce perspective, into layers such as logistics, contracts, payments, assortment, and marketing. The core idea of a layer model is to separate distinct functional units into their own, independent layers, so that each layer can be decided on its own merits: build it in-house, or source it through a partner.

Anchored in research

  • The OSI model (telecommunications, ISO/IEC 7498-1) — the original inspiration for the layer principle: each layer handles its own, bounded task and can be swapped out and tested independently without other layers needing to know anything about its internal implementation.
  • Hagel, J. & Singer, M. (1999), "Unbundling the Corporation," Harvard Business Review — three underlying businesses that most companies bundle into a single organization: infrastructure management (optimized for scale), product innovation (optimized for speed), and customer relationship management (optimized for scope). These three can't be optimized simultaneously — which is why separating them into their own layers is often worthwhile.
  • Baldwin, C. & Clark, K. (2000), Design Rules: The Power of Modularity — modularity theory: splitting a system into modules/layers with clear interfaces allows each part to be developed, replaced, or outsourced independently, without having to redesign the whole system.

Method

  1. Pick a layer perspective that fits the situation. Examples: a technical/functional stack (infrastructure → operations → product/ service → customer interface → brand), or a commerce stack (logistics → contracts → payments → assortment → marketing → customer experience). The layer perspective always has to be adapted to context — there's no single correct way to divide it up.
  2. List every functional part of the business or concept and assign each one to the layer it belongs in.
  3. Define each layer's interface to its neighboring layers — what it takes in and what it produces — so the layers stay genuinely swappable (as in the OSI model: each layer provides a service to the layer above it without that layer needing to know how it's implemented internally).
  4. Classify each layer using the Hagel & Singer lens. Ask which of the three underlying businesses the layer belongs to — infrastructure management (wins on scale and low unit cost), product innovation (wins on speed to market and creative talent), or customer relationship management (wins on scope and reach) — and note that these three logics pull against each other inside a single organization. A layer that is a genuine source of competitive advantage under its logic is a build candidate; a layer that is mature or commoditized, or that is optimized under a different logic than the rest of your business, is a partner or buy candidate.
  5. Make a build/partner/buy call layer by layer, using step 4's classification as the primary input (the same decision logic applies outside AI contexts too — see ../../../ai-strategy-and-governance/skills/build-vs-buy-vs-partner-ai/SKILL.md).
  6. Check the modularity of the interfaces you've drawn, per Baldwin & Clark: can a layer actually be replaced without redesigning its neighbors? A layer boundary that leaks implementation detail into the next layer isn't a real module yet — tighten the interface before treating the build/partner decision as final.
  7. Visualize the layer model as a stack (e.g. bottom to top) and mark each layer's build/partner decision and its rationale.
  8. Check the whole: does the sum of the layers produce a coherent customer experience, or does friction at the interfaces show through to the customer (e.g. a slow handoff when two layers sit with different providers)?

What this skill does NOT do

  • Doesn't hand you a ready-made layer breakdown for every situation — the layer perspective always has to be chosen and adapted to context.
  • Doesn't make the build/partner/buy decision for you — it structures the decision criteria layer by layer to support a human decision.
  • Doesn't replace in-depth supplier or partner due diligence for an individual layer's outsourcing decision.

Refinement notes

Areas to keep deepening with real practice:

  • your own standard layer breakdowns that you reuse repeatedly in specific industries/situations
  • concrete templates (into ../../references/, e.g. layer-stack templates)
  • reference cases / your own examples of a successful or failed layer split
  • what this skill deliberately does not do (guardrails, common mistakes) — add to the list above

Once this section is filled in and validated in practice, update the maturity field in skills_index.json to draft, validated, or canonical (see ../../../meta/maturity_levels.md). Don't add new fields to the frontmatter — name and description are the only ones allowed (see ../../../meta/frontmatter_schema.md).

Continue from here

  • Next in this pack: ../value-chain-mapping/SKILL.md — structures the business according to Porter's value chain model into primary and support activities; a second, more traditional way to structure the same business.
  • Related skill in another pack: ../../../ai-strategy-and-governance/skills/build-vs-buy-vs-partner-ai/SKILL.md
  • This pack's shared guardrails: ../../CLAUDE.md

References

  • ../../references/ — the pack's shared background material
  • ../../CLAUDE.md — the pack's shared guardrails