Alpha/Beta Framework
A framework from Matt MacInnis (CPO at Rippling) applying financial asset concepts to the evaluation of people and processes in product organisations.
The terms
In finance:
- Alpha = outperformance relative to a benchmark index. Positive alpha means the asset did better than the market.
- Beta = volatility relative to the index. High beta means the asset moves unpredictably; low beta means it tracks reliably.
The ideal financial asset: high alpha, low beta. In practice these are correlated — high-variance bets tend to be both the most upside-generating and the most unpredictable.
Applied to people
High-alpha people are creative, unconventional, and generate outsized upside — but introduce unpredictability into teams and processes. Dennis Rodman is MacInnis’s example: exceptional value, exactly one slot on a team.
Low-beta people are reliable, predictable, and consistent — ideal for functions where variance in output is costly.
The framework is useful when decoding intuition about a candidate: “Why didn’t that click?” Running down alpha-beta often surfaces the answer — the candidate is excellent but high-beta in a role that demands low-beta execution, or vice versa.
Applied to processes
Processes in an organisation exist for the sole purpose of lowering beta. They reduce variance in output. The cost of every process is that it also suppresses alpha: it removes the freedom that generates creative upside.
The consequent design question: in which parts of the product organisation do I want to lower beta (mature products, compliance-adjacent functions, infrastructure), and in which do I need to preserve alpha (zero-to-one product work, experimental features, new market entry)?
MacInnis’s example at Rippling: payroll demands very low beta (users depend on perfect execution); the PQL (Product Quality List) implements low-beta standards there. New product areas require high-beta tolerance.
Related
- Power Law and Entropy — companion framework from the same episode
- Matt MacInnis on Extraordinary Efforts, Power Law and Entropy, and Building Rippling — source episode
- Matt MacInnis — speaker page