Diffusion of Innovation — The Bass Model
Takeaway
The Bass model explains adoption curves via innovators (external influence) and imitators (word-of-mouth), producing an S-shaped cumulative adoption over time.
The problem (before → after)
- Before: Adoption forecasting is ad hoc.
- After: A simple differential equation with two parameters (p, q) fits many product lifecycles.
Mental model first
Early adopters try new tech on their own; later adopters follow others. The more who adopt, the stronger the imitator effect—until saturation.
Just-in-time concepts
- Hazard of adoption: p + q F(t), where F is cumulative adoption fraction.
- Differential equation and analytic solution for N(t).
- Estimation from time-series sales data.
First-pass solution
Fit p and q by nonlinear regression on sales time series; forecast peak timing and saturation.
Iterative refinement
- Extensions: Marketing mix, pricing, and heterogeneity.
- Competition: Multi-product diffusion and substitution.
- Network-aware diffusion and contagion models.
Principles, not prescriptions
- Separate innovation push (p) from social pull (q).
- Validate forecasts against cohort and channel data.
Common pitfalls
- Overfitting with short histories; parameter instability.
- Ignoring supply constraints or channel effects.
Connections and contrasts
- See also: [/blog/network-effects], epidemic SIR models.
Quick checks
- What shapes the S-curve? — Interaction of p and q.
- What is peak sales timing? — When imitator effect starts to wane near saturation.
- How to extend? — Add covariates for marketing spend and price.
Further reading
- Bass (1969), subsequent empirical studies