Following are some good practices that need to be followed while performing database testing:

• Validate every single PIECE of data that includes metadata and functional data based on the requirement specifications.
• Test data should be verified and created after consultation with the development team and only then the VALIDATION needs to be done.
• Output data needs to be validated by using both AUTOMATED and manual processes.
• Perform boundary value analysis, graphing techniques, equivalence partitioning techniques to get the required test data inputs.
• The referential integrity of the data in the tables also needs to be carefully evaluated.
• The selection of default values in the table needs to be validated too to check for consistency of data.
• Check if appropriate logging events have been tracked and added to the database.
• Are the jobs scheduled to run are executed on time?
• Ensure that timely backup of the test database is taken to ensure test cases do not fail.
• Ensure that the test server is periodically refreshed with close to real-time data so that testing can be as accurate as POSSIBLE.