Anything that has low certainty or has a lot of impact should be tested early. — Scott D. Anthony Copy Share Image
