1. Docs
  2. Use cases
  3. Use cases
  4. How do I prove every requirement is tested?

How do I prove every requirement is tested?

Write each requirement as sub-requirements in Analyze and link the test cases that prove each one. InSight › Test Health then counts a sub-requirement as covered only when a linked scenario checks a result and passed its last run, and it lists every sub-requirement that is not covered yet.

8 min read Updated CustomersAdmins

In short

On this page you can

  • Trace every sub-requirement to the test cases that prove it.
  • See Requirement coverage as a percentage against an 80% target.
  • Find where requirements drop out: no scenario, no pipeline, never ran, nothing checked.
  • List the sub-requirements that have no test at all.
  • Save the result as a report or print it as PDF.

How to use it

  1. In Analyze, write main requirements with sub-requirements, one folder per module.
  2. Select a sub-requirement and click Add under Test Cases, then Save.
  3. Put the test cases in a pipeline in Schedule and run it.
  4. Open InSight › Test Health and read Requirement coverage, Coverage funnel and Gaps.
  5. Click Save report to keep the numbers as they are today.

The short answer#

Coverage in SmartestQA is counted on sub-requirements: one rule with one expected result. A sub-requirement is proven when a test case linked to it checks a result and passed its last run. A test that only clicks through the screens and turns green does not count.

You do three things: link test cases to sub-requirements, run them in a pipeline, and read the Test Health tab of InSight. The tab shows the share that is proven, the step where the rest got stuck, and the list of sub-requirements with no test.

The problem this solves#

Many teams keep requirements in one place and tests in another. The link between them lives in a spreadsheet, a column in a document, or someone’s memory. Before a release or an audit, a person goes through the list row by row.

Two things are missing. The list is out of date soon after it is made, because tests and requirements keep changing. And a row that says “tested” often means only that a test exists, not that it ran, passed and checked the result the requirement asks for.

An example#

This is an illustration, not a customer story. A team builds an insurance claim form and has to show before each release that every rule of the form is tested.

  1. The QA lead writes the rules in Analyze as main requirements with sub-requirements, one folder per module. The claim form has 60 sub-requirements.
  2. A tester selects the sub-requirement for the upload size rule, clicks Add under Test Cases, picks two test cases in Test Case List and clicks Save. The message is “Requirement saved successfully”.
  3. In Schedule the tester adds the requirements to a pipeline with Add Requirement and clicks Save & Execute. The run executes the test cases linked to them.
  4. In InSight › Test Health the Requirement coverage card shows 70% and 42 / 60, under the marker at the 80% target.
  5. Coverage funnel names the biggest loss, and Gaps › Unlinked sub-requirement lists the sub-requirements with no test, riskiest first.
  6. Before the release the QA lead clicks Save report. PDF prints it with a header that names the report, its date and who saved it.

The team no longer keeps a separate list: the coverage number is calculated from the links and the last runs, and the open rules are named on the same tab.

With a spreadsheet, with a separate tool, and with SmartestQA#

QuestionA spreadsheetA separate requirements or test management toolSmartestQA
Where is the link between a requirement and its tests?In a row someone fills inIn the tool, often by an ID that the test scripts must carryOn the sub-requirement, under Test Cases
When does a requirement count as tested?When someone marks itUsually when a linked test reported a passWhen a linked scenario checks a result and passed its last run
Who keeps it up to date?A person, by handAn import or integration you set up and maintainThe numbers are calculated from the project when the tab opens or on Refresh
How do I find the requirements with no test?Filter the empty rows, if the sheet is completeA report of the tool, if the links are completeGaps › Unlinked sub-requirement, riskiest first
What do I hand over for a release or an audit?A copy of the sheet. This works well for a small, stable listAn export of the toolA saved report that opens exactly as it was saved, or its PDF

Steps#

  1. Write the requirements as a tree

    Open Analyze. Keep one folder per module, and under it main requirements with their sub-requirements. A main requirement is a feature or user flow; a sub-requirement is one condition with an expected result. See Write requirements, or let the AI wizard write them from a document.

  2. Select a sub-requirement

    Open the main requirement and click the sub-requirement in Sub Requirements. The right panel switches to that sub-requirement. Test cases link to sub-requirements only, so with the main requirement selected Add is disabled.

  3. Link the test cases that prove it

    In the Test Cases section of Detail, click Add. The Test Case List dialog shows the test cases of the project. Click each test case that checks this rule, close the dialog and click Save.

    “Requirement saved successfully”. The links are stored when you save. Details: Link an existing test case.

  4. Run the test cases in a pipeline

    A link alone proves nothing: the test has to run and pass. In Schedule, add the test cases to a pipeline with Add Test Case, or add the requirements themselves with Add Requirement, and click Save & Execute. A pipeline that includes a requirement runs the test cases linked to its sub-requirements. See Add test cases, requirements or pipelines.

  5. Read coverage in Test Health

    Open InSight › Test Health. Requirement coverage shows the percentage and the fraction behind it, with a marker at the 80% target. Risk coverage counts the same for CRITICAL and HIGH sub-requirements.

  6. Find what is not proven

    Coverage funnel follows the sub-requirements through five steps: Sub-requirements, With scenario, In a pipeline, Ran, Proving. The card names the biggest loss. Modules × coverage shows the same per module, and Gaps › Unlinked sub-requirement lists each sub-requirement without a test, riskiest first.

  7. Keep the evidence

    Click Save report, name it and click Save. A saved report opens exactly as it was saved. PDF prints the tab. See Saved reports.

Watch it#

The video shows the requirement tree, linking two test cases to a sub-requirement, and the Test Health tab with its pillars, the coverage funnel and the module table.

Video transcript

The tree starts at the project and has one folder per module, marked M, with a count. Open a module and select a main requirement. Its details open beside the tree: status, severity, dates, linked test cases and comments. Assessment (AI) is the next tab. A sub-requirement is one rule. Select it to see its details on the right. Test Cases lists what proves this rule. Nothing is linked yet. Add opens the test cases of the project. Add the cases that check this rule: here, two tests of the reset form. They now show under the sub-requirement, and their results count for it. In InSight, Test Health answers the next question: how much of the product is really proven? Automation maturity is one score built from four pillars, each with an 80% target. Fix these first lists the three actions that raise the score most. The coverage funnel shows where requirements drop out on the way to a passing test. Modules × coverage ranks every module by risk, coverage and last results.

From the requirement tree to the coverage funnel in InSight.

What you get#

  • A link from every rule to its tests. Each sub-requirement lists the test cases that prove it. One test case can prove several sub-requirements, and one sub-requirement can have several test cases.
  • Four coverage numbers, each against an 80% target: Requirement coverage, Scenario coverage, Risk coverage and Action coverage.
  • One score. Automation maturity is a score out of 100 with a band: Good (80 and up), Needs work (50–79) or Low (under 50).
  • What to do next. Fix these first lists at most three actions, sorted by how much each one moves a pillar.
  • The list of gaps. Gaps shows up to 100 rows per tab: sub-requirements without a scenario, scenarios in no pipeline, and actions no test uses.
  • A frozen copy. A saved report keeps the screen as it was on that day, with the name of the person who saved it.
Requirement coverageShown in: InSight › Test Health
Requirement coverage = proven sub-requirements ÷ sub-requirements

proven = a linked scenario checks a result and passed its last run

Example: 60 sub-requirements, 42 of them proven. The card shows 70% and 42 / 60. All formulas: The four pillars.

Questions and answers#

When does a requirement count as covered?Customers

When a scenario linked to the sub-requirement checks a result and passed its last run. Checking a result means the scenario has, for example, an assertion with an actual value or a visibility check with a text. See the four pillars.

My tests pass. Why is requirement coverage still low?Customers

A passing scenario counts only if it checks a result and is linked to a sub-requirement. Trust issues lists the scenarios that turn green without proving anything, and Gaps lists the sub-requirements with no scenario. See Scenarios pass but requirement coverage stays low.

How do I list the requirements that have no test?Customers

Open InSight › Test Health and go to Gaps. The Unlinked sub-requirement tab lists them with their main requirement and risk, riskiest first. Open opens the sub-requirement. See Gaps.

How do I show coverage for a release or an audit?Customers

Click Save report on the Test Health tab before the release. The saved report opens exactly as it was saved, and PDF prints it with a header that names the report, its date and who saved it. See Saved reports.

Can an AI assistant write the missing tests?Customers

Yes. Generate scenario on a row of Gaps and Copy MCP prompt on a recommendation copy a ready prompt. Paste it into Claude or another MCP client that is connected to SmartestQA. See MCP & AI assistants.

What about requirements that cannot be automated?Customers

Sub-requirements marked as not automatable are left out of every pillar, so they do not lower the percentage. See Coverage — four pillars.

Read more#

Didn’t find your answer?Write to support with your project name and what you see on screen.

info@smartestqa.ai