Skip to content

Concepts Overview

PRISM provides a multi-dimensional model for organizing metrics, ownership, and maturity tracking across your organization.

The PRISM Model

PRISM organizes metrics using four key dimensions:

Dimension Purpose Values
Domain Functional area with standards operations, security, quality
Layer Ownership boundary in stack code, infra, runtime
Stage Lifecycle phase design, build, test, runtime, response
Category Type of control prevention, detection, response, reliability, efficiency, quality

These dimensions work together to classify every metric and clarify accountability.

Organizational Model

Domains

Domains represent functional areas with overlay teams that define standards:

Domain Description Overlay Team
Operations Reliability, performance, efficiency SRE/Platform
Security AppSec, CloudSec, compliance Security Team
Quality Testing, code quality, defects QE Team

Layers

Layers represent the full value stream from ideation to support:

Layer Description Typical Owner
Requirements Product ideation, specs, design Product/Design
Code Application code, libraries, dependencies Stream-aligned teams
Infra Cloud resources, networking, platform Platform team
Runtime Running services, production workloads Stream-aligned + SRE
Adoption Product analytics, user engagement Product/Growth
Support Customer support, incident management Support/CS

Each layer can define golden signals (latency, traffic, errors, saturation) pointing to specific metrics.

Teams

Teams follow the Team Topologies model for clear accountability:

Type Role Owns
Stream-Aligned Build and run services Services, code/runtime metrics
Platform Provide infrastructure Infrastructure layer
Enabling Help adopt practices Cross-team capabilities
Overlay Define standards Domain standards

Services

Services are deployable units that connect teams, layers, and metrics:

  • Owned by a team
  • Deployed in a layer
  • Associated with metrics and SLOs

Metrics and SLOs

Metric Classification

Every metric belongs to:

  • A domain (what functional area)
  • A stage (when in lifecycle)
  • A category (what type of control)
  • Optionally a layer (where in stack)
  • Optionally a service (which system)
  • Optionally a quality vertical (ISO 25010 characteristic)

SLO Definition

Metrics can have machine-evaluable SLOs:

{
  "slo": {
    "target": ">=99.99%",
    "operator": "gte",
    "value": 99.99,
    "window": "30d"
  }
}

Maturity Roadmap

Goals

Goals represent strategic objectives with SLO-backed maturity levels:

  • Define 5-level maturity progression
  • Specify which SLOs must be met at each level
  • Track progress from Reactive to Optimizing

Phases

Phases organize work into time-bounded periods (quarters):

  • Set goal maturity targets (enter/exit levels)
  • Group initiatives into swimlanes
  • Track completion and SLO compliance

Connecting It All

The PRISM model creates a clear chain of accountability:

Team owns → Service has → Metrics with → SLOs required by → Goals tracked in → Phases

Example flow:

  1. Payments Team (stream-aligned) owns Payments API (service)
  2. Payments API has Availability metric (operations/runtime/reliability)
  3. Availability has SLO: >=99.99% over 30 days
  4. Reliability Goal requires this SLO at Level 4
  5. Q2 2026 Phase targets Reliability Goal from Level 3 to Level 4
  6. Security Team (overlay) defines security standards that apply to all services

Scoring

PRISM Score

A composite health score (0.0-1.0) combining:

  • Maturity scores (40% weight) - organizational capability
  • Performance scores (60% weight) - metric achievement
  • Awareness multiplier - customer communication effectiveness

Customer Awareness

Track customer awareness through four states:

State Weight Description
Unaware 0.0 Customer not aware
Aware (not acting) 0.25 Aware but not remediating
Remediating 0.5 Actively working on fix
Remediated 1.0 Issue resolved

Framework Mappings

Map metrics to industry standards:

  • DORA metrics
  • SRE practices
  • NIST Cybersecurity Framework
  • MITRE ATT&CK

Getting Started

  1. Define your organizational model
  2. Identify domains relevant to your org
  3. Map teams to Team Topologies types

  4. List services and their ownership

  5. Define metrics

  6. Classify by domain, stage, layer
  7. Set baselines and targets

  8. Define SLOs with operators

  9. Create goals

  10. Define maturity levels
  11. Specify SLO requirements per level

  12. Link initiatives to goals

  13. Plan phases

  14. Set quarterly goal targets
  15. Organize initiatives into swimlanes

  16. Track progress and SLO compliance

  17. Calculate and report

  18. Generate PRISM scores
  19. Create roadmap reports
  20. Track maturity progression