Why Terraform's Single-Team Model Breaks as Organisations Scale

Managing Terraform infrastructure with a single engineer or team works well initially, but the model breaks down as more teams — networking, platform, application, and security — need to deploy independently. Shared state files cause pipeline bottlenecks, since Terraform's state locking blocks all teams when any one team runs a plan, even on unrelated resources. Shared credentials in a common CI pipeline also create security and compliance risks, as all teams' secrets become accessible to anyone who can trigger a run. Terraform currently lacks native per-team or per-workspace access controls, forcing organisations to manage separate backends per team, which in turn creates challenges around passing outputs across team boundaries. Common workarounds such as manual value handoffs or terraform_remote_state data sources each introduce their own reliability and coupling problems, highlighting a widely felt need for a scalable multi-root-module architecture.