Skip to content
Tutorials Center
My learning

Topics

Lesson 3 of 60 ยท Testing Fundamentals

QA, QC and Testing

Understand the difference between quality assurance, quality control and testing, verification vs validation, quality attributes, and who does what in a team.

Job ads say "QA Engineer", managers talk about "quality control", and developers say they are "just testing". These words overlap in everyday speech, but they name different things. Knowing the difference helps you talk to your team, read job descriptions, and understand why a test team alone cannot make a product good.

In this lesson you will separate quality assurance, quality control and testing, learn the pair verification and validation, meet the common quality attributes a product can be judged by, and see how roles are split in real teams. The lesson builds on What Is Software Testing? and the seven principles.

Quality, in one sentence

Quality is the degree to which a product meets the needs and expectations of the people who use it, pay for it and depend on it, within the limits of cost, time and law. Notice that this is wider than "no bugs". A fast, bug-free app that nobody understands has poor quality for its users, and a feature-rich app that crashes daily has poor quality too.

QA, QC and testing

The three terms form a nesting, from the broadest to the narrowest:

Term Focus Question it asks Examples
Quality assurance (QA) The process Are we working in a way that tends to produce good results? Defining coding standards, review rules, a definition of done, training, process audits, learning from past defects
Quality control (QC) The product Does this output meet the quality criteria? Inspecting a build, running acceptance checks, release sign-off
Testing Evaluating the product by examining or running it What does the product do, and where does it differ from expectations? Executing test cases, exploratory sessions, reviews

QA is preventive: it tries to stop defects from being created. QC is detective: it looks for defects in what has been built. Testing is the main activity that QC uses, so testing sits inside QC, and QC sits within the wider goal of quality assurance.

flowchart TB
  QA["Quality assurance<br/>process: prevent problems"]
  QC["Quality control<br/>product: detect problems"]
  T["Testing<br/>run and review the product"]
  QA --> QC --> T

Each layer is narrower than the one above it: testing is a part of quality control, which is part of the overall effort to assure quality.

Many companies call their testers "QA", and in practice the label does not matter. What matters is remembering that quality is built by the whole team. If only the testers care about quality, defects are found late and the cost curve from the first lesson takes over.

Prevent, detect, check

QA Improve how the team works to prevent defects.
QC Check that the product meets the criteria.
Testing Run and review the product to find out.

Process on top, product below, tests as the main tool.

1 / 2

Sorting activities

Sorting real activities into the right bucket is the best way to make the terms stick. The stand-in below holds a list of everyday project activities, each tagged by the question it answers, and prints them grouped. It is only a classification table, not a real quality system, but changing a tag shows how you would argue for it.

JS Which bucket does it belong in?

Press Run (or Ctrl + Enter) to see the console output here.

No output.

Verification and validation

These two words describe two different questions you ask about a product, and good testing covers both.

  • Verification: "Are we building the product right?" It checks that a work product conforms to its specification: the code follows the design, the design follows the requirements. Reviews and most test cases verify.
  • Validation: "Are we building the right product?" It checks that the finished product actually satisfies what users need in their real situation. Trial use by real users, demos and acceptance testing validate.

A simple example: the requirement says "the report shows total sales per month". The page shows exactly that, so verification passes. But the sales manager actually needs sales per region, so validation fails. The product is correct against a document and still wrong for the user, which is the absence-of-errors fallacy from the previous lesson.

A common memory trick: verification is about the document ("did we follow the spec?"), validation is about the value ("does it do the user any good?").

Quality attributes

"Does it work?" is only one of many questions about quality. Standards such as ISO/IEC 25010 list the characteristics by which a software product can be judged. Names vary between versions of the standard, but the usual set looks like this:

Attribute Question
Functional suitability Does it do what is needed, correctly and completely?
Performance efficiency Is it fast enough and does it use resources sensibly?
Compatibility Does it work with other products, browsers and systems?
Usability Can the intended users learn and use it, including people with disabilities?
Reliability Does it keep working and recover from faults?
Security Does it protect data and resist misuse?
Maintainability Can it be changed and tested without breaking things?
Portability Can it be moved to another environment?

Not every attribute matters equally for every product. A banking app weights security and reliability heavily, a game cares about performance and usability, a one-off internal script may care about almost nothing but correctness. Choosing which attributes matter, and how much, is part of the test strategy (see Test Strategy and Test Plan). The types of testing that target each attribute are previewed in Test Types.

The next stand-in is a small scorecard. You give each attribute a weight (how much it matters for this product) and a score from the latest evidence, and it computes a weighted result and lists the weakest important attributes. The numbers are examples; a real team would base scores on actual test results.

JS A weighted quality scorecard

Press Run (or Ctrl + Enter) to see the console output here.

No output.

The average looks respectable, but the last line points at the two attributes that matter most and are weakest. That is the value of looking at attributes separately: one weighted number can hide a serious problem.

Roles in a team

Titles differ a lot between companies, so judge by responsibilities instead of names:

  • Developers write code and also unit tests and reviews; in many teams they own quality at the source.
  • Testers (manual testers, QA engineers) design and run tests, explore the product, report defects and give the team information about risk.
  • Test automation engineers (SDETs) write and maintain automated checks and the tools around them.
  • Test lead or test manager plans testing, tracks progress and reports to stakeholders.
  • Business analysts and product owners define what the product should do and take part in acceptance.
  • Release, DevOps and operations staff run pipelines and monitor production.

In modern agile teams the line between these roles is deliberately blurry: testers join planning, developers write automated tests, and everyone shares responsibility for quality, as Testing in Agile Teams and CI/CD explains later.

One important idea about independence: people tend not to see their own mistakes, so a fresh pair of eyes finds more. Independent testing can range from a developer's colleague to a separate team or even an outside company. Independence adds objectivity but can add distance and delay, so teams balance the two.

Check your understanding

Quick check ยท 5 questions

of 5 answered You scored / 5 ยท perfect!

  1. 1. A team introduces a rule that every pull request must be reviewed by a second person. Which term best describes this activity?

    A review rule changes the process to prevent defects from reaching the product, which is quality assurance. QC would examine the finished output.
  2. 2. The software exactly matches its specification, but users say it does not help them do their job. Which check failed?

    Verification asks whether the product was built right (matches the spec), which passed. Validation asks whether it is the right product for users, which failed.
  3. 3. How are quality control and testing related?

    Quality control checks that products meet criteria, and testing is the principal technique it uses, so testing sits inside QC.
  4. 4. For a children's drawing app, which of these quality attributes is likely to carry the most weight?

    Quality attributes are weighted by context. For an app used by young children, ease of learning and use is central.
  5. 5. Why is some independence between author and tester useful?

    Authors share the assumptions that created a defect, so others often see what they miss. The trade-off is communication distance, which teams manage.

Summary

  • Quality means meeting the real needs of users and other stakeholders, which is wider than having no bugs.
  • QA improves the process to prevent defects, QC checks the product, and testing (running and reviewing) is the main tool of QC.
  • Verification asks "built right?" (against the spec), validation asks "right product?" (against real needs).
  • Quality attributes such as functionality, performance, usability, reliability and security are weighted differently by context.
  • Titles vary, but everyone on the team shares responsibility for quality; some independence between author and tester helps find more defects.

Next up: Testing in the SDLC.