Learn how to investigate quality before automating checks. Manual testing and automation solve overlapping problems and work best together.
The useful question is not which kind of testing will win. It is which approach gives the team reliable information about a particular risk. A person exploring an unfamiliar feature and a script repeating a known check contribute different kinds of evidence.
For learners, the right starting point depends on your current skills. If you cannot yet explain a test's expected result or why it matters, adding code can make the activity faster without making it more useful.
- Explore the behaviour
- Choose what to repeat
- Review the results
Understand the strengths of each approach
Manual testing lets a person adapt as they learn. You might notice confusing wording, follow an unexpected route or change a test after discovering an unstated assumption. This is useful when requirements are evolving or the questions are not yet well understood.
Automation executes programmed actions and checks consistently. It can provide fast feedback on stable journeys after a change, but it only evaluates what has been encoded. Someone still needs to decide what to check, interpret failures and update the suite as the product changes.
| Situation | Useful starting approach | Why |
|---|---|---|
| A new, unclear workflow | Explore manually | Learn behaviour and identify risks |
| A stable critical journey | Automate selected checks | Repeat important checks after changes |
| A usability concern | Observe and investigate | Understand where people become confused |
| Many predictable data combinations | Use data-driven checks | Compare outcomes consistently |
Automate a valuable question
A good first candidate is important, repeatable and sufficiently stable. A booking confirmation appearing after a valid submission may be useful. A frequently redesigned visual detail may cost more to maintain than the evidence it provides.
Consider the entire cost: test data, environment, debugging, execution and maintenance. If a test fails so often that the team ignores it, its speed has little value. Before writing code, describe the risk, the expected result and the evidence the check must collect.
An automation decision
Risk: a valid learner cannot register after a release. Check: submit a unique test registration, confirm the success state and verify the expected stored record in an authorised test environment.
Choose your next learning step
If you are new to testing, start with requirements, risks, test design and defect reporting. Build a small manual testing project. Once you can explain the cases and results, learn a programming language and automate a few stable checks from the same project.
If you already test professionally, identify where repetitive work delays useful feedback. Learn the language and framework your target team uses, then practise test isolation, data setup and diagnosis. Avoid trying to replace every manual activity during your first automation project.
Make automated results worth trusting
A passing test should mean that a meaningful outcome was verified. Clicking a submit button without checking the result is not enough. Similarly, adding retries can hide an unstable test rather than explain why it fails.
Playwright's guidance recommends user-visible behaviour, isolated tests and assertions that wait for expected states. These principles help keep checks connected to the product experience. Use debugging evidence when a check fails, then distinguish a product defect from a test or environment problem.
Further reading: Playwright: Best practices
Show both kinds of thinking in your portfolio
Use one project to demonstrate exploration and repeatable checks. Write down a risk you discovered manually, explain why you automated part of it and record the questions that still require investigation. This shows that you understand automation as one part of testing.
When comparing courses, check whether the curriculum includes test design and troubleshooting alongside syntax. Ask what project you will complete and what feedback you will receive. A tool name on a certificate does not by itself demonstrate that you can maintain useful checks.
Choose three checks to automate
- List ten checks from a practice project
- Score each for user impact, repeatability and stability
- Choose three and explain the evidence each must produce
- Keep an exploratory session for questions the scripts do not answer
Build these skills with support
Explore live courses with guided practice, projects and space for questions.
Create Your AccountOur practical examples are learning exercises. References are linked beside the relevant sections. Course fees, schedules and access terms are maintained on the course pages. Tell us if something needs correcting.
