Adaptive Socio-Technical Systems with Architecture for Flow • Susanne Kaiser • GOTO 2024

Building adaptive socio-technical systems requires optimizing for fast flow of change and constant feedback while balancing “building the right thing” (effectiveness) with “building the thing right” (efficiency)

Use value chain mapping to identify important change streams and analyze how changes flow through the system to reveal potential blockers and inefficiencies

Domain-Driven Design helps partition complex systems into bounded contexts that serve as clear ownership boundaries and enable modular architecture

Team cognitive load must be optimized by limiting the size and complexity of systems teams are responsible for - exceeded cognitive load creates delivery bottlenecks

Platform teams should provide self-service capabilities that make cross-functional stream-aligned teams self-sufficient rather than dependent on others

Team Topologies provides clear team types (stream-aligned, platform, enabling) and interaction modes (collaboration, facilitating, x-as-a-service) to optimize team structures

Generic subdomains can leverage existing solutions (off-the-shelf, open source) while core domains that provide competitive advantage should be custom built

Teams should be aligned to customer-facing streams of changes rather than technical silos to reduce handovers and wait times

Conway’s Law means system design mirrors communication structure - team boundaries should align with software architecture boundaries

Evolution stages (Genesis → Commodity) require different mindsets (Explorer → Town Planner) and practices as components mature