Playbooks
I'm about to…
Pick the moment you are in. Each playbook walks you through a short sequence of models, with the one question each model asks you.
Design and build
Shaping a system before and while it is built.
Starting a new system
Greenfield, guided by Fowler, Brooks, Beck, and Nygard.
Architecture Is About the Important Stuff → Monolith First → Conceptual Integrity…6 steps
Reviewing a solution design
Spend review time where the irreversible risks live.
C4 Model (Zoom Levels for Audiences) → Everything Is a Trade-off → Coupling and Cohesion…9 steps
Deciding whether to split into services
Microservices solve organisational problems at a technical price. Make sure you have the problem first.
Microservice Premium → Monolith First → Bounded Context…7 steps
Integrating systems or designing an API
Every interface you publish becomes a promise. Decide which promises you can keep.
Information Hiding → Hyrum's Law → Postel's Law…6 steps
Building a feature with AI
Start simple, measure everything, and never combine the dangerous capabilities.
The Jagged Frontier → Workflows Before Agents → Context vs Behaviour…7 steps
Choosing a data store
One of the most consequential, least reversible choices you will make.
One-Way vs Two-Way Doors → CAP and PACELC → Data Gravity…6 steps
Designing for failure
Decide how the system fails before production decides for you.
Everything Fails, All the Time → Single Point of Failure → Circuit Breaker…8 steps
Deciding what to build
Make sure the architecture serves a real need before you design it in detail.
Jobs to Be Done → Outcomes Over Output → Opportunity Solution Tree…7 steps
Run and scale
Keeping it fast and standing in production.
Change
Evolving or replacing what already exists.
Migrating a legacy system
Replace the plane while it is flying — without dropping passengers.
Chesterton's Fence → The Map Is Not the Territory → Strangler Fig…6 steps
Making the case to pay down technical debt
Nobody funds a clean-up. They fund faster delivery and lower risk.
Technical Debt Quadrant → Technical Debt as Interest → Design Stamina Hypothesis…6 steps
Rationalising the application portfolio
Decide what to keep, fix, move or retire across hundreds of applications.
Capability Mapping → TIME Portfolio Triage → Total Cost of Ownership…7 steps
Integrating after a merger
Combine two companies' systems without stalling both businesses.
Operating Model (Standardisation and Integration) → Capability Mapping → Bounded Context…7 steps
Strategy and influence
Choices that span teams, vendors, and the business.
Evaluating a vendor or platform
Buy the capability, not the brochure.
Core vs Context → Wardley Mapping → Build, Buy, or Partner…7 steps
Designing team boundaries
Shape the organisation and the architecture together.
Conway's Law → Bounded Context → Inverse Conway Maneuver…5 steps
Setting an architecture strategy
Decide where to invest, where to standardise, and where to let things change fast.
Operating Model (Standardisation and Integration) → Capability Mapping → Wardley Mapping…7 steps
Presenting to executives
Translate architecture into the language of value, risk, and time.
The Architect Elevator → Relativity (Stakeholder Frames) → The Pyramid Principle…7 steps
Getting a decision agreed
Win support for an architecture decision across people you don't manage.
Power–Interest Grid → Influence Without Authority → Ubiquitous Language…8 steps
Setting principles and running governance
Keep a large estate coherent without making the architecture board a bottleneck.
Guardrails, Not Gates → Why Is More Important Than How → Fitness Functions…7 steps
Responding to new regulation
Turn a regulation such as DORA, NIS2 or the EU AI Act into architecture work, not a paper exercise.
Capability Mapping → Power–Interest Grid → Exit Strategy (Lock-in as a Design Input)…7 steps
Defining a target state and roadmap
Turn a strategy into a target architecture and a sequence of steps people can fund.
Operating Model (Standardisation and Integration) → Capability Mapping → Three Horizons…7 steps
Your first 90 days as an architect
Understand a new organisation before you try to change it.
Circle of Competence → Capability Mapping → The Map Is Not the Territory…8 steps
Making a hard decision
Decide well under uncertainty, at the right speed, with the right people.
Decision Rights (Who Has the D?) → Cynefin → One-Way vs Two-Way Doors…8 steps
AI transformation
Rethinking the enterprise so AI brings real speed and value, in four stages.
AI transformation 1: Before you start
Decide whether you are adding AI to today's organisation or redesigning the work around it.
The Dynamo Lesson (Productivity Paradox) → Amara's Law → The Jagged Frontier…8 steps
AI transformation 2: Making the enterprise AI-ready
Fix the fragmented knowledge, duplicated data and opaque integrations that stop AI delivering.
Capability Mapping → TIME Portfolio Triage → Data as a Product (Data Mesh)…8 steps
AI transformation 3: From assistants to agents
Raise AI's autonomy step by step, with evidence, safeguards and people who can still take over.
Workflows Before Agents → Evals as Fitness Functions → The Autonomy Slider…7 steps
AI transformation 4: Organising for AI
Reshape teams, incentives, sourcing and bets for an enterprise where machines do much of the work.
The Nature of the Firm (Transaction Costs) → Operating Model (Standardisation and Integration) → Inverse Conway Maneuver…8 steps