The CTO Podcast · Insights & Strategies for Chief Technology Officers Navigating the C-Suite while Balancing Technical Strategy, Team Management, & Innovation

Greenest Programming Languages -- Stephan Schmidt // Amazing CTO

·51 min·3 clips
Compiled languages used 120 joules on average, while interpreted languages used 2,300 joules.
The episode opens light. A weekend RV story gives the conversation a familiar, slightly comic start before it turns back toward the work of leading engineering teams. Then it gets serious. The host and guest spend most of the middle section on metrics, asking what a CTO really gets from the time they allocate to a developer or a team. The answer is not simple. They argue that the first move is to decide what the company is trying to achieve, and only then choose metrics that help measure that outcome. That keeps the conversation grounded. It also avoids the trap of picking numbers in a vacuum and treating those numbers like the strategy. Outcome first, output second. Output metrics such as GitHub or Jira activity can still matter, but mostly as debugging tools when the business result goes missing. The talk stays practical. Team-level metrics come up as a way to evaluate senior technical leaders when their scope is broad enough to make that fair. The warning is familiar. Goodhart's law hangs over the whole discussion, because a metric that becomes the target can easily start pushing the wrong behavior. Nothing here is treated as enough on its own. They are careful to say that metrics cannot replace active performance management. Expectations still have to be set. Feedback still has to be given, and teams still need support when the numbers go sideways. The most surprising idea lands near the end. A support-ticket metric is floated as a possible signal, even to the point of tying unhappy customers and rising tickets to pay, which makes the point sharp even if it feels more like a provocation than a policy.

As heard by us

A practical take on outcomes, metrics, and the management tradeoffs behind measuring engineering work.

The conversation treats engineering metrics as a management problem, not a reporting exercise. After the RV opener, it gets into the awkward question of what a developer's time actually buys, and how to measure that without kidding yourself.

Read the full review in PlayNext →

Why you'd press play

A CTO conversation about metrics that turns into a real management problem.

Read the full recommendation in PlayNext →
Listen to the show on