Innovation accounting is the lean-startup answer to a hard question: how do you measure progress when revenue is zero and the product is an experiment? Eric Ries’s framework replaces P&L-style evaluation with learning-based evaluation in three steps: establish a baseline with an MVP (where do the metrics stand today?), tune the engine through build-measure-learn iterations that try to move those metrics toward the business-model targets, and at defined checkpoints decide to pivot or persevere based on whether the curve is bending. Its currency is validated learning — hypotheses confirmed or killed by cohort behaviour, not by opinion.
In practice it disciplines two audiences. Teams stop reporting vanity totals and start reporting whether the model’s critical assumptions (activation, retention, willingness to pay) are converging on viability. Sponsors — VCs, corporate innovation boards — get a principled basis for continuing or stopping funding that does not depend on theatre. Tranches and stage gates in early-stage financing are, in effect, innovation accounting formalised.
Governance and reporting edges
The framework touches law where learning metrics become the basis for money decisions. Milestone-based investment documents should encode innovation-accounting metrics with contractual precision — defined events, defined windows, audit access — because “the cohort improved” is not a closing condition anyone can enforce. Inside corporates, innovation accounting feeds management reporting; where those numbers reach public disclosures or impairment tests on capitalised development costs, finance and legal need the same metric definitions the product team uses. Honest measurement is also the cheapest defence against the most expensive startup failure mode: scaling a model the data had already falsified.
If this is on your desk
Templates and checklists are free in the Founder Academy; for a specific situation, book a 30-minute intro call.
Founder AcademyBook an intro call