Using data envelopment analysis to measure Web site efficiency

Christopher Ward Iervolino, Pace University


This study introduces a prototype managerial tool for measuring and managing the efficiency of business to consumer (B to C) Web sites. By defining each Web site as a separate Decision Making Unit (DMU), this tool is able to utilize Data Envelopment Analysis (DEA) to determine which sites are the most efficient and provide guidelines to improve the efficiency of the sub-optimal sites. ^ Within the context of this study, an efficient Web site is defined as a site that generates the most output, using the smallest amount of inputs. Inputs refer to measurements representing the amount of effort required to build and maintain the site. For the purposes of this study, they include metrics which track the amount of Web site content, the frequency of content update and the complexity of the content and its arrangement. This study defines Web site output as amount of traffic the site generates. These outputs are measured as the average number of daily hits and the average number of daily peak hits the site generates. This data is then adapted and incorporated into the DEA-based model to measure efficiency, and provide recommendations, for comparable B to C Web sites. ^ Data is collected from 12 Web sites, using two different Web site analysis software tools and services. Only static e-commerce Web sites which contain catalogs, which are created using exclusively static content, are included in this test. In order to limit scope, this study uses three separate Web site characteristics to select only simpler Web sites, more typical of small business. Firstly, the study examines only sites containing catalogs comprised of exclusively static content. Secondly, it considers only those sites that have less than 250 hypertext markup language (HTML) pages. And thirdly, it uses only sites whose traffic is measured by a single objective third party Web-tracking service. The resulting DEA model then generates relative efficiency scores and related recommendations for improvement for each Web site. The results are interpreted against those of typical Web site efficiency measurement methods and appropriate generally accepted Web design best practice recommendations. ^

Subject Area

Computer Science

Recommended Citation

Iervolino, Christopher Ward, "Using data envelopment analysis to measure Web site efficiency" (2002). ETD Collection for Pace University. AAI3064836.



Remote User: Click Here to Login (must have Pace University remote login ID and password. Once logged in, click on the View More link above)