Automation

What Is Data-Driven Testing?

Data-driven testing is a test automation approach where the same test logic is executed multiple times with different input data sets, separating test data from test scripts.

Start a Free Mock Interview

Free to start · 7-day trial on paid plans

In Depth

Instead of writing separate tests for each input combination, data-driven testing stores inputs and expected outputs in an external source (CSV, Excel, JSON, database) and feeds them into a single test template. A login test might run ten times with different combinations of valid/invalid emails and passwords from a spreadsheet. The test logic stays the same; only the data changes.

Most modern frameworks have built-in parameterization. JUnit 5 has @ParameterizedTest with @CsvSource, pytest uses @pytest.mark.parametrize, Jest uses test.each, and TestNG has @DataProvider. These mechanisms iterate over data sets and report each combination as a separate test result, making failures easy to trace to specific inputs.

The main benefit is coverage amplification. A single parameterized test with 20 data rows provides the same coverage as 20 separate tests but with a fraction of the maintenance cost. The challenge is managing test data at scale: keeping data sets versioned, ensuring data does not go stale, and avoiding combinatorial explosion where too many parameters create millions of meaningless combinations.

Why Interviewers Ask About This

Interviewers want to know that you can design efficient, scalable test suites. Data-driven testing is a direct indicator that you think about coverage systematically rather than brute-forcing individual tests.

Example Scenario

A payment form accepts 12 currency types. Instead of writing 12 test functions, a data-driven test reads currency codes and expected formatting rules from a JSON file. Each row produces a separate test case in the report. When a 13th currency is added, only the data file needs updating.

Interview Tip

Mention the specific parameterization mechanism in your framework of choice. Discuss how you decide which data to include and how you avoid data-maintenance overhead.