When using the concept of microsites, it is still important to establish a consistent look and feel for the site as a whole. People should feel comfortable as they move from your support pages to your marketing pages to your employment pages. A consistent user interface breeds familiarity and generates a united front. The user need not know that the site is constructed in modules, and a changing interface can lead to a user becoming lost and confused while exploring. It helps to have one person designated as the overall Webmaster for the project, to coordinate the efforts of the various subgroups. The Webmaster's role is basically the same as a project manager on a large software project.

Who Is the Audience?

Of course, just having a purpose for a site isn't enough; you need to consider a site's audience. Are the users coming from within your organisation, or from outside? Is it young or old? What language do they speak? When do they visit the site? What technologies do they support? What browsers do they use? Figuring out an audience doesn't have to be that hard, but don't assume that your audience is too large. People from South America or the Sudan can visit your Web page--but do they? Should they? It is important to be realistic about the audience of the Web. There may be millions of users on the Web, but they aren't all going to visit a particular Web site.

Far too often, sites are designed around the needs or beliefs of their authors, not around the needs and beliefs of the actual users of the sites. This is a sure way to limit the success of the site.

When the idea of a web site's audience is discussed, do not think in terms of a nameless, faceless John Q. Cybercitizen with a modem. The serious visitor is an individual looking for specific information. They already have a vague expectation about what information they need and the site design must anticipate this need and present a readily recognisable link to that information.

Who Will Pay for It?

Web Sites cost money to produce, so they generally have to produce some benefit in order to continue. While people do put up sites for personal enjoyment, there is a limit even to this type of site in terms of an individual's investment of time and money: It is very important to understand the business model of the site. Only a year or two ago, many corporate Web budgets were not always the first concern due to the novelty of the technology. Today however, Web sites often have to prove that they're "worth it." The money has to come from somewhere.

A site's creator could pay for everything, but that probably isn't reasonable unless the Web site is for pure enjoyment or nonprofit. Typically some funds have to be collected, probably indirectly, to support the site. A promotional site for a movie may not directly collect revenues, but hopefully the site will influence the audience and have some impact on the success or failure of the film. What's interesting is that many Web sites are nearly as indirect as a movie promotion site. It can be very difficult to measure the direct benefit of having such sites. More directly measurable sites are those where leads are collected or goods are sold. Some value can be put on these transactions, and an understanding of the benefit of the site can be determined.

As information services become more pervasive, consumers will probably begin to value content on the Internet despite its lack of tangibility.

Another business model involves getting someone besides the owner or the intended audience of the site to pay. This model typically comes in the form of an advertising-driven site. However, what is interesting about advertising is that there is actually a good being sold--the audience.

Advertisers are interested in reaching a particular audience and are willing to pay for an advertisement based on the effectiveness of that ad reaching the intended audience.

The question is how can an audience be attracted, measured, and then sold to the advertisers? The obvious approach is to provide some reason that an audience will want to come to a Web site and identify themselves. This is very difficult. Furthermore, the audience must be accurately measured so the advertiser has a way to compare the size of audience from one site to the next and knows how to spend his or her advertising dollars. People often discuss the number of visitors to their site as an indication of value to an advertiser. The advertisers, however, may not care about the number of visitors unless those visitors are in their target audience. Furthermore, the number of visitors doesn't mean much of anything as far as price is concerned unless it can be compared against other advertising venues even those beyond the Web. Planning a business model around advertising is less than straightforward. Parties interested in funding their ventures in this manner would be well advised to study traditional advertising-based media, like magazines, to understand the pros and cons of this approach.

Whoever is paying for the site, there must be some understanding of the costs and the benefits of the site. How much does each visitor actually cost, and what benefit does he or she produce? Understand that the number of visitors doesn't count, even when using the advertising model. The value of the site transcends this figure and addresses the effectiveness of the visitation. In other words, many visits don't necessarily mean success. Having many visitors to an online store who nonetheless make few purchases may mean huge losses, particularly if it costs more to reach each visitor. Even the form of the Web site may change the cost. Since the amount of data delivered from a Web site is generally related directly to the site's variable costs, sending video would cost more than sending regular HTML text. High costs aren't a bad thing if there is a big payoff. This is why a goal must be set.

Defining Goals

A goal for a site is not the same as its purpose. A purpose gives a general idea of what the site is for, while a goal is very specific. A goal can help define how much should be spent, but goals must be measurable. What is a measurable goal of the site? Selling x dollars worth of product directly via the Web site is a measurable goal, as is selling x dollars of product or service indirectly through leads. Reaching a certain usage level per day, week, or month can be a goal. So is lowering the number of incoming technical support phone calls. There are many ways to measure the success or failure of a Web project, but measurements generally come in two categories: soft and hard. Hard measurements are those that are easily measured, such as the number of visitors per day. Soft measurements are a little less clear. For example, with a promotional site for a movie, it might be difficult to understand if the site had any effect on the box office sales.

Why are measurements so important? From a manager's perspective, measurements can be used to determine how much to spend. If a Web site's goal is to produce $10,000 of new sales, then spending $500,000 on the site is not acceptable unless the site has some other nonmeasurable value that can make up the other $490,000. While this seems like common sense, a clear return on investment or cost benefit is seldom determined for corporate Web sites. Soft measurements can make things difficult, since it is not clear what effect the Web site may have. In the advertising industry, certain rules of thumb apply, like spending ten percent of overall sales on advertising, a percentage of that would obviously go toward a Web site. Due to the hype surrounding the Web, very little business sense is exhibited toward Web development. Even if sites were considered as little more than online brochures, it is obvious that the more brochures printed on paper, the more expensive. The more pages the more expensive. The more complicated, the more expensive. Web sites are the same way.

Scope equals money. The more that is put into a Web site, the more it costs. Because of the flexible nature of the Web, many developers want to add as much as possible to the Web site. However, more isn't always better. The more that is added to the site, the more it costs. Furthermore, having too much information makes it difficult to find essential information. To think about scope, return to one of the first steps in the process. What is the main purpose of the site? Shouldn't the information of the site reflect this purpose? Looking at the Web, this doesn't seem to always be the case. Have you ever gone to a site and not understood its point?

Finding the essentials of a Web site might not be easy, particularly if there are many purposes or many parties involved in its development. One approach is to have a brainstorming session where users provide ideas. Each idea is then written down on a 3 x 5 card. After all the cards have been created, ask the users to sort the cards into piles. First, sort the cards into like piles to see how things are related. Next, sort the piles in order of importance. What is important can eventually be distilled out of the cards. Remember to cut down the number of cards to make people focus on what is truly important.

Instead of coming up with ideas of what should go into a site to meet a particular goal or goals, you may be tempted to take existing materials such as marketing pieces and convert them to the Web. Unfortunately, creating the content of the site based solely on all the text of all manuals, brochures, and other support materials rarely works. Migrating text from print to the Web is troublesome because the media are so different. Reading onscreen has been shown to be 25 percent slower than reading from paper. In practice, people do not tend to read information online carefully. They tend to scan it quickly and then print out what they need. In this sense, writing for paper tends to go against screen reading. Think about newspaper or TV news stories: the main point is said first and then discussed. This goes against the slow build up of many paper documents, which carefully spell out a point. With visitors skimming the site, key bullet points get read while detailed information might get skipped. The main thing is to keep the point obvious and simple. Even if information is presented well, organisation can ruin all the hard work in preparing the information. If a viewer can't find the information, who cares how great it looks or how well it reads?

Once the purpose, content, and structure of the site have been determined, a plan can be developed. The plan should consist of a flow chart of the site, a time line, and a document that describes what is needed and how the site will be put together.

Far too many Web sites are viewed as glorified brochures. A page is just a bunch of words and pictures marked up with HTML. If this is true, then all books are just a bunch of words and pictures marked up with their own formatting language. They, too, should be easy to produce--but the reality is far more complicated. Maybe the real issues are novelty and complexity. Print designers don't charge the same rates as interactive designers, and many question why. One reason is the newness of technology. Today, relatively few people know how to design Web pages well. Looking beyond this situation, however, there is a much different answer. Many Web sites are not brochures. Some sites let the user purchase something, others provide searching features, and some even let the user play games. Very advanced sites automatically configure themselves depending on the visitor 's preferences or browser type.

If you extend these ideas to a Web site, it becomes a lot more than a brochure. By adding technology--such as a database, interactive forms, or programmed objects like Java applets--to the mixture, the Web becomes more like software and less like print media. In this sense, some effort must be made to select the appropriate technology for the job, and to properly integrate the look and content. As sites become more complex, look will still be important, but much of the effort will go into the technology. The shift away from a page paradigm of Web sites to a program paradigm is fast becoming a reality.

Sites are born, live, and die. Far too often, sites are born full grown and slowly die due to neglect. A Web site is not a building that is rarely expanded once it's finished. It's more like a tree. Most sites continually grow and change. Occasionally, whole sections might be pruned, while other sections grow out of control like a sucker at the trunk of the tree. Many times the original creators of a site don't think about the future, and stunt the growth of a site by not designing with room to expand. The original creators may move on to other sites or get bored of the task and assign it to another person. Sometimes the new owners rip the site up and start afresh. Other times they take the site in a whole new direction.

A: Planning Setting goals, specifying content, organising content, and setting the user interface to navigate content

B: Implementing Creating content, implementing navigation and the user interface, and coding the site, which may include HTML, programming, and database development

C: Testing User, browser, and system testing

D: Maintenance Maintaining and updating the site, questioning old goals, and returning to the planning stage

Each phase listed above has distinct subphases that could be expanded to provide more detail. For example, planning could include determining an audience, setting a budget, and performing other project-oriented tasks.

The process of creating a large Web site is more than just putting together some text with images and HTML. While a simple site with a few pages might not take a great deal of forethought, a little planning can go a long way even for a small site. What is the site supposed to do? How should it look? What should it say? What is its goal? Who is the target audience? What is the site's value? Setting a goal can help ensure the future success of the site, particularly if the value of the site can be measured. Determining what the site should do, given a particular goal, might be hard, particularly with many competing purposes. Isolating the core essence of the site is possible using some known techniques. Once content and purpose are determined, the shape and structure of the site can be set. At this point, the Web page author has a blueprint to explore building the site by gathering the raw materials that make it up (images, text, and other content). The content and navigation can then be assembled using Web technologies like HTML.

Once built, the site must be thoroughly tested, since it is created for the benefit of users, not authors. Web sites are not static--they evolve. The mission of the site may change over time. Its look and technology may become dated. The process of Web publishing is an endless loop.