Why we need test data generation in software testing?

Today's quality engineering leaders want secure test data to be made available on demand into different test environments that are On Premise and in the Cloud. The test data that is needed has to be of "high quality", meaning that the data is in a known "state" and condition so that the results of a particular test case are predictable and valid. Synthetic test data generation allows any volume and variety of data to be generated very accurately... not just "realistic" data, but unique / new data, negative and edge case data, all valid permutations & combinations, and any volume of data, in any format - not just data sitting in a SQL database. Almost none of the data just described is available from a production database copy (masked or synthesized) which means that most quality engineers resort to manual methods, spreadsheets, scripts, etc. to try to build the data - which is slow, cumbersome and defeats the purpose of "quality at speed". Synthetic data generation is changing how the software testing industry is able to test - reducing cycle time, increasing coverage and increasing productivity.