Lesson 5 of 5
Bridging sprints to SIT, UAT and progress reporting
8 min
Sprints prove slices. Go-live needs proof of the whole. In SAP Activate, Realize closes with dedicated work for integration testing and final user acceptance, and the program reports progress in a way that shows whether the slices are adding up. This lesson connects the agile and phase-level views, and covers how to report agile progress truthfully.
What sprint testing cannot prove
- End-to-end processes across teams, such as order to cash flowing into revenue and intercompany postings for company code 1810.
- Interfaces with real partner systems, such as the dealer portal.
- Behavior with migrated data volumes and history from ECC.
- Business roles combined across processes, as real users will hold them.
- Regression: whether sprint 7 broke what sprint 2 delivered.
| Test level | Where and when | What it proves | Who accepts |
|---|---|---|---|
| Sprint story tests | Development or quality system, every sprint | A slice meets its criteria | Product owner |
| Integration testing (SIT) | Quality system, once build is largely complete | End-to-end processes across teams and interfaces work | Test lead against exit criteria |
| Sprint UAT and show-and-tell | During sprints | Business sees and shapes the slices early | Key users give feedback |
| Final business acceptance | Quality system with migrated data, end of Realize | The business accepts the integrated solution | Process owners such as Dan and Maria |
- Sprints 1 to 6Build sprintsStories built, tested and accepted by product owners; regression suite grows
- Sprint 7Integration sprintSIT cycles across all teams and interfaces in QAS client 100
- Sprint 8Firm-up and acceptanceFinal business acceptance on migrated data; fixes and retests
- Realize gateDecisionExit criteria checked; process owners sign; move to Deploy
Keeping sprints and SIT connected
Three habits stop the closing sprints from becoming a surprise. First, every story's test cases join the SIT and regression pool as soon as the story is Done, so SIT plans are assembled, not invented. Second, a regression subset runs at the end of each sprint in the quality system, catching breakage early. Third, the test workstream holds the SIT entry criteria from the start of Realize and reports against them every sprint, so nobody discovers in sprint 7 that interfaces or data loads are not ready.
Reporting agile progress
Agile metrics can mislead as easily as any other. SAP Cloud ALM offers reports such as the burn-up chart, user story traceability, requirement traceability and test execution analysis. Grace's weekly report combines them so each number is checked by another:
| Measure | Source in Cloud ALM | What it reveals |
|---|---|---|
| Stories Done versus scope | Burn-up chart | Progress and scope growth separately |
| Velocity per team | Completed story points per sprint | Whether forecasts are realistic |
| Requirements with passed tests | Requirement traceability | Whether Done work is proven against business needs |
| Regression pass rate | Test execution analysis | Whether new sprints break earlier ones |
| SIT entry criteria status | Test plans and the RAID log | Whether the closing sprints can start on time |
Contoso's agile progress report, built from SAP Cloud ALM analytics.
What would you do?
By sprint 6, Contoso's burn-up shows 85 percent of stories Done, and the PMO slide is green. Priya's requirement traceability view shows only 55 percent of requirements have passed tests linked. What should Grace report?
- Sprint tests prove slices; SIT and final acceptance on migrated data prove the whole.
- SAP Activate Realize closes with sprints for integration testing and final acceptance.
- Feed every Done story's tests into the SIT and regression pool; run regression each sprint.
- Report completion and test evidence together; high completion with low evidence is a warning.
Check your understanding · not graded
Lesson check
1. Why keep SIT even when every sprint tests its stories?
2. Where does final business acceptance happen in SAP's agile Activate content?
3. Which Cloud ALM report shows whether requirements are proven by tests?
4. What does high story completion with low test evidence indicate?