Test summary report, Payroll 4.2 release, Quenwick Systems

Test report template with cases by area, a pass rate chart and open defects

A test report has to tell the people who decide on a release what was tested, what passed, what is still broken and whether the agreed exit criteria are met. This four page report for a payroll module release counts 378 cases by area, charts the pass rate, lists every open defect with a decision, and ends with sign off from QA and the product owner.

Create a document with OneCraft4 A4 pages, editable, then download as a PDF

The document, page by page

Every page as it renders and as it prints, with nothing summarised. Read the wording before you reuse it.

Quenwick Systems · Test summary report · Payroll 4.2Page 1 of 4
Quenwick Systems · Test summary report · Payroll 4.2Page 2 of 4
Quenwick Systems · Test summary report · Payroll 4.2Page 3 of 4
Quenwick Systems · Test summary report · Payroll 4.2Page 4 of 4
Test summary report: Payroll 4.2
Release
Payroll module 4.2, build 4.2.0-rc3
Test cycle
Monday 21 September to Friday 2 October 2026
Planned release
Thursday 15 October 2026
Prepared by
Anouk Verhaegen, QA Lead
Report date
Monday 5 October 2026
376 of 378
Cases executed
96.0%
Pass rate
11
Defects raised
8
Still open

Recommendation. Release on Thursday 15 October 2026 on two conditions: the fixes for PAY-412, a High severity rounding error on part period pay, and PAY-417, a bank file limit, must pass retest by Monday 12 October. Every other exit criterion is met or has an agreed exception, and the six other open defects are Medium or Low with a later fix or deferral agreed by the product owner.

1. Scope

Release 4.2 adds pay for part periods, a new leave accrual engine and a revised bank payment file. Testing covered those three changes, the areas they touch, and a regression pack for existing payroll functions. Load testing and the mobile payslip app were out of scope; load testing is covered in a separate report.

Item
Detail
Environment
Test environment TEST-2, a copy of production data from 14 September, with names and bank details masked
Test data
1,200 masked employees across 6 pay frequencies and 14 leave policies
Methods
Scripted functional cases, a regression pack, and two days of exploratory testing on part period pay
Testers
Four QA analysts and one payroll specialist from the customer advisory group
2. Test cases by area
Area
Planned
Executed
Passed
Failed
Blocked
Pay calculation
86
86
81
4
1
Leave accrual
42
42
40
2
0
Tax and deductions
58
58
55
3
0
Retirement contributions
24
24
24
0
0
Bank payment file
18
18
16
1
1
Payslips
30
30
29
1
0
Regression pack
120
118
116
2
0
Total
378
376
361
13
2

Two regression cases were not run because the old payment file format they test is retired in this release; both are marked not applicable. Two cases were blocked by an expired bank simulator certificate on 29 September and remain blocked, which is the one exception to the exit criteria below. Blocked cases count as executed but not passed.

3. Pass rate by area
Pass rate by area, percent of executed cases
Pay calcLeaveTaxRetirementBank filePayslipsRegression0255075100Percent passed

The overall pass rate is 361 of 376 executed cases, 96.0 percent, above the 95 percent exit criterion. The bank payment file is the weakest area at 88.9 percent, from one failure and one blocked case out of 18, which is why it carries a retest before release.

4. Defects

The 13 failed cases raised 11 defects, because three failures in pay calculation share one cause. No Critical defect was found. Three defects are fixed and retested; eight remain open.

Severity
Raised
Fixed
Open
Deferred
Critical
0
0
0
0
High
2
1
1
0
Medium
5
2
3
1
Low
4
0
4
3
Total
11
3
8
4
Open defects
ID
Defect
Severity
Decision
PAY-412
Part period pay rounds per day instead of on the total, up to 4 cents out
High
Fix and retest by 12 Oct
PAY-405
Leave balance shows yesterday's accrual until the page is refreshed
Medium
Fix in 4.2.1
PAY-417
Bank file total line rejected when a batch holds more than 999 payments
Medium
Fix before release
PAY-409
Deduction report sorts by ID, not by name
Medium
Deferred to 4.3
PAY-398
Payslip footer wraps on A5 paper
Low
Fix in 4.2.1
PAY-401
Tooltip typo on the leave policy screen
Low
Deferred to 4.3
PAY-414
Export button stays active while an export runs
Low
Deferred to 4.3
PAY-420
Help link opens the 4.1 guide
Low
Deferred to 4.3
5. Exit criteria
Criterion
Target
Actual
Met
Pass rate of executed cases
95%
96.0%
Yes
Open Critical defects
0
0
Yes
Open High defects
0
1
No, PAY-412 retest
Regression pack pass rate
98%
98.3%
Yes
Applicable cases executed
100%
374 of 376
Exception, 2 blocked

The product owner accepts the release subject to PAY-412 and PAY-417 passing retest and the two blocked bank file cases being run once the simulator certificate is renewed. If either retest fails, the release moves to Thursday 22 October 2026 and this report is reissued.

QA Lead
Name:
Anouk Verhaegen
Date:
Product Owner, Payroll
Name:
Desmond Achebe
Date:

Section by section

What each section is for, so you can keep the ones you need and drop the rest.

Release details and recommendation
Build, cycle dates, planned release, author, four figures and the conditional go decision.
1. Scope
What changed, what was in and out of scope, environment, data, methods and testers.
2. Test cases by area
Planned, executed, passed, failed and blocked for seven areas with totals, and the not run and blocked cases explained.
3. Pass rate by area
A bar chart of pass rate by area and the weakest area named.
4. Defects
Raised, fixed, open and deferred by severity, and the eight open defects with a decision each.
5. Exit criteria and sign off
Five criteria with target, actual and met, the conditions, and two signature blocks.

How to adapt this document

First, change the areas in the cases table to the features in your release, and keep the five count columns, because planned against executed is what shows whether testing finished. Second, if you test in sprints rather than a release cycle, make this a one page summary per sprint and keep the open defects table carried forward, so nothing is quietly forgotten between sprints. Third, for a client project, add a short section listing any acceptance tests the client ran themselves and their result, and add the client as a third signatory.

Keep the numbers honest

Test reports lose trust when the numbers are massaged, for example by dropping failed cases from the executed count or quietly lowering a severity before release. Show cases that were not applicable and say why, keep blocked separate from failed, and record each deferral with the person who agreed it. The international software testing standard and the ISTQB glossary give shared definitions for these terms.

What makes this document work

The recommendation comes with its conditions

The first paragraph says release on Thursday 15 October if two fixes, PAY-412 and PAY-417, pass retest by Monday 12 October. The last paragraph says what happens otherwise: the release moves to 22 October and the report is reissued. A go or no go without conditions is rarely honest.

Every count reconciles

Passed, failed and blocked add to executed in every row, and the totals read 361, 13 and 2 of 376. Thirteen failures become eleven defects because three share a cause, and the severity table adds to 11 raised, 3 fixed and 8 open.

The exit criteria show the one miss

Five criteria with target and actual: pass rate 96.0 percent against 95, regression 98.3 against 98, and one open High defect against zero, marked No. An honest exit table lets the product owner sign knowing exactly what is being accepted.

Questions people ask

What is the difference between a test report and a test plan?

A test plan is written before testing and says what will be tested, how, by whom and what counts as done. A test report is written after and says what actually happened against that plan: cases run, results, defects and whether the exit criteria were met. The report should point back to the plan's criteria rather than invent new ones.

What pass rate is good enough to release?

There is no universal number; agree it before testing starts, usually in the test plan. Many teams use 95 percent of executed cases with no open Critical or High defects, plus a higher bar such as 98 percent for regression. The pass rate alone is not enough: one High defect in pay calculation matters more than ten cosmetic ones.

How should defect severity be defined?

By the effect on users and the business, not by how hard the fix is. Critical stops the system or corrupts data with no workaround; High gives wrong results in an important function, such as pay out by cents; Medium has a workaround; Low is cosmetic. Write the definitions in the test plan so they are not argued during triage.

Who should sign off a test report?

The person accountable for testing, here the QA lead, confirming the results are accurate, and the person accountable for the release decision, here the product owner, accepting the remaining risk. For a regulated system or a customer contract, a client representative may also sign. Signing records who accepted which open defects.

What does a blocked test case mean?

A case that could not be run to a result because something outside the software under test got in the way, such as a broken environment or, here, an expired bank simulator certificate. Report blocked cases separately from failures, give the reason, and say when they will be run, because they are untested risk rather than known defects.

Build your own in about a minute

The button below opens the generator with this use case already described. Change the wording to match your own, generate, then edit anything you like.

Make one like this

Other document examples

Want the steps in the builder? Read Create a document with AI, then Send a document for signature. For everything this generator can do, see the document maker.

Sources

Written and checked by the OneCraft team. Last checked .