A cornerstone of Web project management is sound effort estimation, which is the

process by which a person, group of people, tool, model, or a combination of these

predicts the amount of effort needed to accomplish a given task. The importance of

having realistic effort estimates at an early stage in a project’s life cycle is widely

recognised by Web project managers and developers since accurate estimates are

paramount to manage resources effectively, and to considerably increase the probability

that projects will be finished on time and within budget.

However, recent findings showed that numerous Web projects worldwide are still

not finished within time and budget, and one of the main reasons for that is the

use of poor project management practices, which includes poor effort estimation

(Ginige, 2002).

Given that the field of Web effort estimation is relatively new, with its first paper

published only in 2000 (Mendes, Counsell, & Mosley, 2000), it would not be surprising

that research findings in this field may not have reached industry widely.

Effort estimation, also known as cost estimation, is a necessary part of an effective

process, whether this process is Web authoring, design, testing, or development

as a whole. An effort estimation process involves the following steps (Fenton &

Pfleeger, 1987).

1. The identification of factors perceived to influence effort (e.g., size measures,

cost drivers) and the type of relationship they have with effort

2. The gathering of past project data using as basis the factors identified in Step

1. If gathering a Web company’s own data is not possible, this phase can in

volve the use of large cross-company data sets with data on numerous diverse

projects.

3. The use of one or several effort estimation techniques to be applied to estimate

effort for new projects. The effort estimates generated by these techniques can

also be adjusted using expert opinion.

4. The assessment of how effective the prediction technique is

It is important that all these steps are followed; otherwise, the effort estimates obtained

may not be as useful as they should.

Another important point is that effort prediction is generally based on the following.

• Knowledge of previous similar projects and applications managed by an organisation.

This means that data from past projects may be essential in helping

organisations estimate effort for new projects.

• Other project and application characteristics that are believed to be related to

effort. Examples of project characteristics are the size of the development team,

its experience with working on similar projects, and the number of different

programming languages being used in a project. An example of an application

characteristic is the size of the problem to be developed, which may encompass

different attributes such as the total number of new Web pages, total number

of images, number of features that were reused, and so forth.

The objective of this book is therefore to provide Web companies, researchers,

and students with the necessary knowledge on Web effort and cost estimation. It

includes step-by-step guidelines on how to use and compare several effort estimation

techniques, which may considerably help companies improve their current effort

estimation practices, and help researchers and students understand the process that

needs to be carried out to estimate development effort.

The effort estimation techniques that are detailed in this book are those that have

been to date the three mostly used effort estimation techniques in the Web effort

estimation literature, namely, stepwise regression, case-based reasoning, and classification

and regression trees (CART).

Throughout this book, we take the view that the use of past data on finished Web

projects can be extremely useful and necessary to help obtain accurate effort estimates

for new Web projects and also to help Web companies understand how they

currently estimate effort for their new projects.

All chapters are self-contained, and whenever applicable, the chapters present a literature

review of previous studies on the topic being explained in order to ensure that

readers are familiarised with what has been previously published in the literature.

We also present detailed chapters on principles of statistics and empirical studies to

provide readers with additional knowledge not only useful for Web effort and cost

estimation, but also useful within the context of Web engineering.

It is important to note that despite Web effort estimation being one of the main components

of project management, there are at least two other components also related

to effort estimation that can indirectly benefit from the use of more sound estimating

processes. These two other components are project productivity and development

processes. Once a Web company identifies important factors that have a bearing

on effort and gathers data on past projects, it is also possible to identify current

productivity trends between projects and also between developers, and to assess if

differences in development processes have any effect on the effort estimates proposed

and learn how to improve processes. Therefore, a change to a more systematic way

of estimating effort can also drive a company to improve its productivity and current

development processes, which is always a positive outcome.