Decision Model Registry

This registry documents the methodology behind each architecture decision tool so model assumptions and scope stay transparent.

AI Cost Risk Model

v2.1

Estimates enterprise AI cost exposure by combining usage, pricing, and operational risk drivers before production scale.

Dimensions

  • request volume and traffic growth
  • input and output token mix
  • model pricing sensitivity
  • operational overhead assumptions

Tools Using This Model

AI Build vs Buy ROI Model

v1.2

Compares build and buy pathways using expected-value financial logic, implementation timelines, and risk-adjusted outcomes.

Dimensions

  • implementation timeline
  • benefit realization assumptions
  • execution and delivery risk
  • total cost of ownership

Tools Using This Model

Architecture Fitness Model

v1.0

Assesses whether architecture choices are proportionate to business need, team capability, and operating context.

Dimensions

  • system complexity versus need
  • delivery team capability
  • operational readiness
  • architecture maintainability

Tools Using This Model

Cloud Over-Architecture Model

v1.1

Detects over-engineering by comparing cloud component complexity against workload needs and team readiness.

Dimensions

  • component necessity fit
  • readiness-adjusted complexity
  • cost inflation risk
  • rationalization impact potential

Tools Using This Model

Integration Complexity Model

v1.0

Evaluates enterprise integration landscapes for fragility, coupling, and operational burden.

Dimensions

  • interface and system sprawl
  • integration pattern diversity
  • failure propagation exposure
  • governance and observability maturity

Vendor Lock-In Risk Model

v1.0

Quantifies long-term dependency risk by assessing portability, contractual constraints, and ecosystem concentration.

Dimensions

  • portability and exit complexity
  • proprietary service dependence
  • contract and commercial constraints
  • strategic optionality risk

Tools Using This Model

Agentforce Flex Credit Consumption Model

v1.0

Models Agentforce credit burn from first principles: token overflow inflation, billing model break-even analysis, and voice stack cost decomposition across action volume and deployment patterns.

Dimensions

  • token overflow rate and severity
  • effective actions per interaction
  • billing model selection (Flex vs. Conversations)
  • voice cost stack complexity
  • payment model and volume discount fit

Tools Using This Model

Architecture Brief Intelligence Pipeline

v1.0

7-step AI pipeline that ingests curated RSS feeds, applies binary relevance triage via CF Workers AI, deduplicates with semantic embeddings, and uses Vertex AI (Gemini) for deep architectural impact extraction and macro trend synthesis.

Dimensions

  • source relevance triage
  • semantic clustering and deduplication
  • architectural impact extraction
  • macro trend synthesis

Tools Using This Model