The Monday report, from a real week.
Every Monday the team gets this in Slack, counted from the log — not written from memory. Below is one real week of the QA engineer at our first customer, a data-analytics company, with the customer's text taken out and the shape kept.
Where the real report links to a ticket, a plan or a screenshot, this page says so instead.
1. Work completed.
97
30
8
14
263 working turns in the team's channel; 599 actions in the board, the repositories, CI, the browser and the deploy tool. The week's largest piece: a thirty-case plan for a new feature, executed against the sandbox build, with a consolidated verdict per ticket — one comment, a checklist of what was verified and what remained, the way the QA lead asked for it.
2. Evidence.
Each of the 97 results carries what he actually ran: a screenshot or a log line, the build it ran against, the time. In the real report every line below is a link into the team's own tools.
- 97 recorded results → the QA plan, one case each, with the evidence attached.
- 8 deploys → the deployment records, with the build and the outcome.
- 2 measured findings → the two comments with the numbers (see 3).
3. Bugs detected.
10 defects before production, filed where the team's rule says they go. The two that mattered most came from the live API, not the interface: a revoked key kept authorising real calls long after the interface said it was gone, and an expired key did the same. He measured both windows in minutes, posted each as its own blocker with the numbers, corrected his own earlier wording when the measurement finished, and the task went back to the developer with a precise description of what to fix.
4. What waited for a yes.
103 changes ran after a teammate's yes — none without: deploys to the sandbox, status moves, comments on tickets, the defect tickets themselves. Under today's delegated trust level most of these would run and be reported; deploys and status moves would ask once per ticket. Nothing touched production.
5. What the team corrected.
Three rules in two days, none written by us:
- The QA lead found a fourth defect himself and sent a screenshot. Jarred recorded the miss against his own plan, named the case that would have caught it, and proposed a rule so it is added next time.
- Defects of a feature not yet in production go into the feature's task, not into separate bug tickets — he cancelled the two tickets he had opened that morning, moved the descriptions, and wrote the rule down. Confirmed the same evening; in force since.
- One consolidated result comment instead of three, with a checklist of what was verified and what remained — rewritten, approved, the default since.
6. What it would have cost with people. And what it cost.
≈ 38 hours
€900–1,200
$94
Two kinds of numbers on this page: the counts and the AI figure are logged; the hours and the cost range are our estimate, with the assumptions shown. Part of the work is avoided labour, part is work that would not otherwise have happened. The Installed program agrees the measure and the assumptions with you before the first week, so this line is one you signed, not one we chose.
7. What he still got wrong.
He is a junior with a very good memory, not a senior. In the same week he over-corrected a finding after a hint instead of asking, posted a blocker before a measurement had finished, and once answered a question about the wrong ticket after his working session rotated. Each is logged; each became a fix on our side. The report says so, because a team that hires him should know what the first month looks like.
This is what Monday looks like. Every Monday.
Counted from the log, sent to whoever pays the bill. A 30-minute call to pick the process and the number we'd measure.