If I understand you correctly, you are trying to find the Column U data that matches the column A data and extract that so that you are only left with numbers that do not match, would that be a correct assumption?