“TestData” plays a vital role in Software testing whether it can be white box or black box testing. It depends on what we test and scope of testing. Without having proper testdata, application might introduce critical defects in production. Also, redundant testdata used in testing may not cover the intended scenario. So quality of testdata is more important. Each scenarios need to be verified with appropriate testdata. It can be generated in multiple ways such as manually, scripting (automated), mocking and also from real customer data. However, each of the testdata generation has its own challenges. Manual Generation: This is a simple and easy way to generate basic set of test data for any application using popular techniques such as Boundary Value, OATs. It covers valid, invalid, null, outbound values to test the application. Scripting: Instead of manually generating random test data, we can create a script with a defined logic and generate various possible test data. This will reduce manual effort and can be triggered as a batch in non-working hours. By the time testers come back to work, test data will be ready for the defined scenarios. Mocking: Applications we test may depend on other services. When other services are unavailable, we may not be able to test the service where new changes made. Instead of relying on dependent services, mock the response. Especially this will be useful for the exception flow. This would reduce the time taken to identify the test data for all possible scenarios. Realtime data: Customers are the real testers. We all know that 100% test coverage of any application is impossible due to less number of resources. We can setup a dashboard in-order to identify various hidden paths and understand the flow. Later, we can analyze the impact caused to the application. These data can be pushed to test environment to simulate the user behavior in early stage.