Context

NASA initiated two heavy-lift launch systems with comparable mission goals: the Space Launch System (SLS) for the Artemis program and Starship as a commercial, multi-purpose platform.

SLS follows a government-driven, specification-based model shaped by technical, political, and industrial constraints. Starship follows a privately funded, iteration-driven model aimed at reducing cost and increasing flight rate across multiple mission types, including lunar landing (NASA HLS contract), satellite deployment, and long-term Mars missions. The Starship approach shows many aspects of Product Velocity (see other SpaceX case studies).

The divergence reflects different system goals. SLS optimizes for mission assurance, political control, and reuse of existing infrastructure. Starship optimizes for cost per flight and iteration speed.

Value Thinking

AlnAlign Starship aligns development with a single dominant economic outcome: reducing marginal cost per launch through full reusability. This drives architectural decisions across propulsion, structure, and operations. SLS aligns with a broader stakeholder set, including political and industrial constraints, which shape architecture beyond technical optimization. PrioPrioritize SpaceX prioritizes rapid experimentation and flight frequency to reduce uncertainty and converge on viable solutions. NASA prioritizes reliability and stakeholder satisfaction through extensive upfront definition and validation. These priorities reflect different risk strategies: market risk versus mission risk.

Accelerate

SpaceX builds and flies integrated prototypes in rapid succession. CstConstruct Teams introduce changes between flights instead of stabilizing a baseline. This keeps cycle time short at system level. VrfVerify Each flight tests the full stack under real conditions, including ascent, staging, and recovery attempts. Integration happens in the flight itself, not in staged environments. MonMonitor Extensive telemetry captures trajectory, loads, thermal behavior, engine performance, and control response. Engineers analyze this data after each flight to identify failure modes and update models. EvoEvolve Failures are treated as experimental outcomes. VrfVerify Insights from telemetry feed directly into the next design, enabling fast convergence through repeated build–test–learn cycles.

Outcome

Starship achieves higher iteration speed and explores more radical design space at lower cost. SLS achieves high assurance with limited iteration. The comparison shows how system-level value priorities determine development strategy, cost structure, and innovation potential.