Experience Curve Learning Time Calculator

The Henderson experience curve

Aircraft engineers had noticed the pattern since the 1930s, and in 1968 Bruce Henderson and the Boston Consulting Group turned it into a general rule. Every time the cumulative production of a good doubles, the real unit cost falls by a roughly constant percentage. The standard formula is C_n = C_1 · n^(-log₂(1/η)), where C_n is the cost of the n-th unit, C_1 the cost of the first, and η the learning rate. That rate usually sits between 70% and 90%; a rate of 80%, for instance, means each doubling shaves 20% off the cost.

Several things feed the curve at once: workers getting better at the job, process improvements, scale economies, automation, and product redesign. Boeing's move from the 707 (early 1950s) to the 747 (late 1960s) is a well-known case, where the accumulated experience of building airframes pushed unit production cost down by roughly 60%. You see the same logic in semiconductor pricing (the cost declines tied to Moore), photovoltaic modules (Swanson's law), and lithium-ion battery packs, which tracked an ~19% learning rate from 2010 to 2024.

Applications

Corporate strategy (BCG growth-share matrix, cost-leadership planning), pricing of new product launches, defense and aerospace procurement contracts (the U.S. DoD explicitly models learning curves), and forecasting renewable-energy LCOE trajectories.

FAQ

Learning curve or experience curve? The learning curve covers only labor efficiency; the experience curve also includes scale, technology and process gains across the whole value chain.

What happens when the curve flattens? Mature industries hit physical or material limits; further cost reductions then require disruptive innovation rather than incremental learning.

Is a lower η always better? Yes for cost — a 70% rate means costs fall faster — but it also signals fierce price competition. Strategic investments often hinge on the first mover capturing volume to slide down the curve first.