Field notes · 26 November 2025
What we ask before touching your metrics
An assessment is only as useful as the questions asked on day one. Here is the intake we use in Busan and remote kickoffs.
Before we read a single chart, we ask which streams carry customer-facing consequences if they stall. Vanity pipelines can wait; checkout events and fulfillment notices cannot.
Next we ask who owns producer changes versus consumer changes. Split ownership is common and often explains why lag grows after a “small” schema tweak.
We also ask for the last three incidents that felt related to throughput or failures, even if they were never formally written up. Informal notes in chat are enough to start a timeline.
Access matters. Read-only metrics plus a technical counterpart who can answer within a business day beats wide access with nobody available to explain naming quirks.
If your team is preparing for a Throughput & Failure Assessment, gather those answers first. The observation window then starts cleaner and the findings arrive sooner.