The present invention extends to methods, systems, and computer program products for isolating the evaluation of actual test results against expected test results from the test module that generates the actual test results. A results evaluation sub-system receives actual test results resulting from the execution of a test in a test environment. A results retrieval sub-system receives environmental data indicating that a test was performed in the test environment. The results retrieval sub-system selects one or more expected results from a results database based on the received environmental data. The results retrieval sub-system sends the selected one or more expected results to the results evaluation sub-system. The results evaluation sub-system receives the one or more selected results. The results evaluation sub-system evaluates the actual test results against the one or more expected results to determine if the test was successful.