Eric Ries reframes his original Lean Startup concept of Build-Measure-Learn as a general principle rather than a startup-specific ritual: in any high-uncertainty situation, value comes from the speed of the learning loop, judged relative to the normal pace of your own industry — not against some abstract standard. A team that iterates faster than the accepted convention in its field captures a competitive advantage even if its absolute cycle time looks slow by other industries' standards.
He applies this to the AI foundation-model race, crediting Anthropic's (and Claude Code's) shipping velocity as a "Build-Measure-Learn A+" — not just capability, but the rate at which the company learns and re-ships relative to competitors — as a source of competitive advantage distinct from the governance structures discussed under The New Governance (Compliance, Purpose, Coherence, Integrity).
Eric Ries extends Build-Measure-Learn beyond its original Lean Startup context: in any high-uncertainty situation, value comes from the learning each iteration generates, and loop velocity — not absolute project timescale — is the competitive advantage. Shipping speed should be judged relative to what's conventional within your own industry, not against a universal absolute standard: a "fast" cadence in enterprise software may be glacial in consumer social, and vice versa.
He cites Anthropic's iteration cadence on Claude Code as an example: the team prioritized shipping faster than competitors within the frontier-AI-lab convention specifically to compound learning and trust advantages over time, not because faster is inherently better in the abstract.
Apply: Judge shipping speed relative to your own industry's convention, not in absolute terms, and prioritize iterating faster than direct competitors to compound learning and trust advantages, as Anthropic did with Claude Code.