Datapools provide your tests with variable data when they run.
Generated tests are based on user emulation. With recording engaged, you perform a discrete sequence of steps that you expect the tested application's users to perform. From the recording, a test is generated that exactly mirrors your interactions with the application. If you run a generated test without modification, it uses the exact data that you used when recording the test.
For example, suppose you have recorded a test that searches for "Doe, John" in an employee database. If, without modifying the test, you run the test with 10 virtual users, each searches for "Doe, John." This is probably not what you want. Instead, you want the test to use variable data: each test instance should search for a different name. You accomplish this goal by using a datapool containing the variable data that, at runtime, you want to substitute for the actual data in the recorded test.
The following figure shows the test that is generated when you start the
recorder and enter the URL www.ibm.com/xx/, where xx is your country code.
The figure shows the US home page (xx is us).
If a highlighted request contains request values that you have associated
with datapool columns, these values are highlighted in dark green with white
lettering.
Also,
the Test Data table for the page is updated with a
summary of datapool information (green lettering).
Parent topic: Providing tests with variable data