Data validation and testing are both crucial when migrating data to any system. It’s important to understand the difference as they are frequently mistaken to mean the same thing. If validation and testing are not performed thoroughly or not performed in conjunction with one another, the ensuing issues can easily derail an implementation.
What is Data Validation?
Data validation is the process of checking the accuracy, quality, and integrity of the data that was migrated into the target system. When performing data validation, it is important to understand not all of the data will look exactly as it did in the legacy system. The structure of the master data may have changed and values may also be different between the legacy and target system. Payment terms, for example, may be stored as ‘Net 30’ in the legacy system, but changing to ‘N30’ in the target system. Understanding how and what is changing is going to be important when performing data validation activities and the resources performing the validation should have that understanding ahead of trying to perform validation.
What is Testing?
Throughout the implementation process of a new system there ought to be multiple opportunities to perform testing in the target system with the migrated data during planned test cycles. Preparing for an implementation is an iterative process. Issues will be identified in each test cycle, whether it is with the system configuration or the migrated data. The various teams will then work together to resolve the issues identified in preparation for the next test cycle with the goal being much improvement from the previous. This does not mean, however, that there will not be any issues. Again, it is an iterative process and typically with the resolution of one issue, it can uncover additional issues that may have been hidden by the original issue. By the User Acceptance Testing (UAT) cycle, the major issues should have been identified and resolved with the goal being minimal issues and changes post UAT heading into the go live.
Testing is not data validation. Testing is much more involved than performing basic data validation. Since data can drive system functionality, it is important to perform data validation and testing in tandem. The data may look correct, but that does not mean the system is functioning as expected. Testing involves more advanced navigation of the target system along with executing typical business processes in the new system. The results of executing those processes should confirm if the system and functionality is working as expected and meets the requirements needed by the business. All areas of the business should execute end-to-end testing.
Validation and Testing for Success
After witnessing and being a part of many, many implementations, I can confidently say, a good indicator of a successful go live can be partially based on the amount of thorough data validation and testing that was performed throughout the project ahead of the go live. Ensuring your project has a comprehensive and robust data validation and testing plan will put the project on the path to success!
