Week 7 of 8
SIT, UAT and the go/no-go call
Everything so far feeds this week: ranked processes, the customization inventory, validated imports and role tests. SIT proves processes work end to end across systems. UAT proves the business accepts them. The go/no-go meeting decides on the evidence.
When the evidence is not presented honestly, the meeting fails. Birmingham City Council’s auditors found that testing of its Oracle system was incomplete when it went live in April 2022, and that the steering committee was told there were no open defects except intermittent issues.
This week: write exit criteria with numbers, and present them straight
- Write SIT scenarios as end-to-end chains across systems, e.g. web order → credit check → pick and ship → invoice → payment applied → bank reconciliation.
- Make every expected result checkable: a document number, a posted amount, a status. Never "screen appears".
- Set numeric exit criteria, e.g. 95% of critical scenarios passed and zero open severity 1 defects (example: set your own).
- Get UAT signed per process by the process owner, with accepted defects and workarounds listed.
- Send the go/no-go checklist 48 hours ahead, with blocking criteria marked and evidence attached.
Number of the week
73.3%
test pass rate for Birmingham City Council’s replacement income system, against a 95% threshold, with 10 severe issues where zero were allowed. Go-live was delayed.
Source: The Register (September 24, 2025)Use this week
- SIT Cycle Plan & Entry/Exit Criteria →
End-to-end scenarios, sequence and testers per cycle, with live progress counts.
- UAT Sign-Off Form (Word) →
Results, accepted defects and signatures for each process owner.
- Go/No-Go Readiness Checklist →
Criteria with thresholds, blocking flags and an automatic recommendation.
Next week, the last of the playbook: the release calendar after go-live, and a regression practice that keeps up with it.
Get the playbook in your inbox, one week at a time
Subscribe free and every template in the library is yours.