Over several rounds of measured fixes, the pipeline cycle went from 11 to 16 minutes down to under 20 seconds.
A drayage logistics operator, dispatch and billing pipeline that had to finish inside a working shift.
Play both cycles
Both cycles on one clock
Before the fixes the cycle took 11 to 16 minutes. After the last round, 6.8 to 17.3 seconds, under 20 s. On one time axis the new cycle is a sliver at the start of the old one.
Both bars share one linear time axis from 0 to 16 minutes. Solid bars run to the fastest recorded cycle and the striped part to the slowest. The replay is compressed 120 times: 1 second on screen stands for 2 minutes of pipeline time.
| Value | Method | Recorded | Scope |
|---|---|---|---|
| 11 to 16 minutes | Timed end to end on the production pipeline before the change | 2026-09-24 | One full refresh cycle of a pipeline that runs every minute, dispatch and billing data |
| 6.8 to 17.3 seconds (under 20 s) | Project audit of the rebuilt cycle, two orders of magnitude tighter | 2026-09-24 | One full refresh cycle on the same sources, after the last round of fixes |
Dates are when each figure was recorded for this site; the work itself is dated in the setting above.
What this does not prove
The cut came in rounds: 11 to 16 minutes, then 8 minutes, 3 minutes 45 seconds, 49 to 57 seconds, and finally 6.8 to 17.3 seconds. The last round removed 30 seconds of metadata overhead from a cycle of about 50 seconds; no single fix accounts for the whole change. This is the cycle time as recorded by the project audits, not a percentile across every historical run, and it measures elapsed time, not the volume of work the cycle performs.