Design of experiments (DOE) is a systematic method to determine the relationship between factors affecting a process and the output of that process. While it is frequently applied into many industries such as agriculture, clinical trials, chemical engineering, manufacturing, service, etc, only the simplest concepts in DOE are usually applied in information technology as such A/B and multivariate testing using full factorial designs. We apply some more advanced methodologies in DOE into our web page designs, mainly the fractional factorial design ideas to design, test and optimize the websites more  accurately, economically and efficiently. Some of those ideas and methodologies are very similar to the Taguchi method (robust design method) developed by Genichi Taguchi decades ago in manufacturing industry. Some different criteria will be discussed for the optimal design selection in fractional factorial designs, and examples will be given for illustration and evaluation.