Investigating Web size metrics for early Web cost estimation

Abstract

This paper’s aim is to bring light to this issue by identifying size metrics and cost drivers for early Web cost estimation based on current practices of several Web Companies worldwide. This is achieved using two surveys and a case study. The first survey (S1) used a search engine to obtain Web project quote forms employed by Web companies worldwide to provide initial quotes on Web development projects. The 133 Web project quote forms gathered data on size metrics, cost factors, contingency and possibly profit metrics. These metrics were organised into categories and ranked. Results indicated that the two most common size metrics used for Web cost estimation were “total number of Web pages” (70%) and “which features/functionality to be provided by the application” (66%). The results of S1 were then validated by a mature Web company that has more than 12 years of experience in Web development and a portfolio of more than 50 Web applications. The analysis was conducted using an interview. Finally, once the case study was finished, a second validation was conducted using a survey (S2) involving local New Zealand Web companies. The results of both validations were used to prepare Web project data entry forms to gather data on Web projects worldwide. After gathering data on 67 real Web projects worldwide, multivariate regression applied to the data confirmed that the number of Web pages and features/functionality provided by the application to be developed were the two most influential effort predictors.