Developer productivity: how DX measures it
Episode participants
A solo episode without invited guests.
What we discussed on the recording
I trace DX from Pull Panda’s PR analytics to developer surveys and show how Core 4 brings together DORA, SPACE and DevEx. I suggest reading throughput and quality alongside developer experience and where teams spend their effort.
I explain how the AI Measurement Framework connects uptake, outcomes and costs. Using a hypothetical return calculation, I show how the value of added engineering capacity depends on whether throughput gains last and how much review costs. I then apply CAFE(S) to an AGENTS.md instruction for running tests.
I use DX’s 2024–2026 data to show how PR throughput changed and how much the gains varied across companies. I explain how to read quarterly TrueThroughput, which weights PRs by complexity, and what surveys can tell us about workflow friction and time saved with AI.
Correction to the recording (42:37): DX’s Q3 2026 report measures time to first review in minutes. The lowest and highest groups have medians of 40 and 82 minutes. The group comparison covers a single quarter. Timestamps follow the Podster audio.
Links to episodes mentioned
- Episode 12. Measuring Developer Productivity With DX Core 4 — mentioned at ≈14:10 (Podster)
- Episode 30. Developer Productivity: Google’s Human-Centered View — mentioned at ≈17:05 (Podster)
- Episode 11. Measuring AI Code Assistants and Agents — mentioned at ≈20:04 (Podster)