My Response To The Nonsense McKinsey Article On Developer Productivity • Dave Farley • GOTO 2025
Dave Farley critiques McKinsey’s article on measuring software developer productivity, arguing it’s fundamentally flawed and potentially harmful. His key points are:
-
Focus on the wrong metrics: The article focuses on individual developer productivity, which Farley considers a misguided approach akin to the “rockstar programmer” myth. He advocates for using the DORA metrics (Deployment Frequency, Lead Time for Changes, Change Failure Rate, Time to Restore Service), which measure team performance and predict successful outcomes based on evidence. Individual productivity is irrelevant compared to team contribution and effective workflow.
-
DORA metrics are superior: Farley highlights DORA metrics as evidence-based and predictive of success, unlike other metrics that lack such evidence. They focus on the quality and speed of software delivery, rather than individual output.
-
Misunderstanding of measurement: The article wrongly assumes that measuring business functions like sales is easy and accurate. Farley uses examples of sales teams prioritizing monetary value over profitability and customer satisfaction to demonstrate the flawed logic of focusing solely on easily measurable, but ultimately ineffective, metrics.
-
Proposed metrics are flawed: Farley criticizes specific metrics proposed by McKinsey, including lines of code, velocity, story points, and interruptions, arguing they are either irrelevant, easily gamed, or misinterpretations of existing frameworks like SPACE metrics. He also argues that focusing on “inner loop” vs “outer loop” activities (e.g., pipeline optimization) as a productivity measure is nonsensical, as improving the pipeline increases overall team productivity.
-
Harmful consequences: Farley concludes that the McKinsey report is harmful, potentially leading developers to game the system, experience learned helplessness, and discourage valuable activities like pipeline optimization. It promotes a narrow, inaccurate view of software development success.
In short, Farley argues that focusing on individual developer productivity is a mistake. Instead, organizations should focus on team performance and the overall efficiency and quality of software delivery, as measured by metrics like DORA. The McKinsey report, he contends, is based on flawed assumptions and proposes harmful metrics.