In an ideal world, testing would be a pretty straightforward process.
A test team takes the product requirements, writes a test specification document,
reviews the tests, and then runs them all for each version of the product.
The team is composed of full-time staff, and everyone knows exactly what is expected of them.

In practice few organisations have that luxury.
There is not time to run all the tests on every product version - especially on fix-releases
that need to be rolled out quickly. Requirements are constantly changing, and the tests have to be changed in step.
Test staff come and go.
There are misunderstandings over who was supposed to run which tests, so some get missed.
Management suddenly wants a status update at seven in the evening.

In these situations you need the support of a test management tool, such as TestLink.
The purpose of TestLink is to answer questions such as:

- For which requirements do we still need to write or update test cases?
- Which tests do you want me to run for this version?
- How much progress have we made on testing this release?
- Which test cases are currently failing, and what are the errors?
- On which version was this group of test cases last run, and is it time we ran them again?
- And ultimately: is this version of the product fit for release?
TestLink helps you to keep the test process under control.
It forms a repository for requirements and test cases, and relates these to builds,
platforms and staff. You allocate tests to staff who carry them out and record the results.
A wide variety of reports provide information on what has been done and what still needs to be done.