WebRef Update: Featured Article: Software Developers Have Something To Teach You
Software Developers Have Something To Teach You
Way back, years ago when the Web was still in diapers, software engineers slaved away long into the night to produce applications. Specialized applications designed to do one thing and do it well.
Well, those days are over now, brought short by the era of the interconnected online everything. For example, Web sites like The Bureau of Atomic Tourism or Philip Greenspun's photo.net once tremendously popular Web sites, have given way to multi-billion dollar electronic commerce Web sites with bottom-line expectations like Ebay or Yahoo. However, despite this tremendous business shift, the vast majority of Web sites still lack the clear purpose of software applications. The problem is crystal clear: developers often build a site without clearly defining the problems they are trying to solve.
Before you write your first line of code, design your first page template, or polish off your first content piece, you must first strive to get a clear idea of what your goals are (yes, even open source projects need goals, unless you feel really, really lucky). The best sites on the Internet are more like software applications than your average Web site! They have one or more clearly defined core functions, and they do them really well.
Consider a few examples. When you think of Hotmail, what do you think of? Free e-mail, right? That's because that's Hotmail's core service. How about Excite? Well, search is their core service. You might also think of classifieds (they bought a company called Classifieds 2000 a while back, which used to provide classifieds to almost all the big search engines), e-mail (they have a great e-mail service), or my.excite.com. Ebay is for auctions. Amazon is great for books. Progressive.com is for auto insurance (they have a great comparison engine). Mapquest gives you driving directions. Notice that each of the examples has one or more narrowly defined core functions. Some businesses have expanded their services over the years like Excite, but one lesson is clear: successful Web sites are successful because they provide solutions to problems.
It follows then, that the first step to designing a successful Web site is deciding what problems you're trying to solve.
Let's pretend that you are a consultant brought in to rebuild a company Web site. As with every new project, the most difficult part is figuring out where to begin. The most important question to ask is why. "Why are we putting video on the site?" "Why are we making these arbitrary divisions on our Web site to match our organizations structure?" "Why are we putting all of our FAQs in their own section instead of help?" In order to solve problems, first you must define what they are. Write down your questions, especially if you're not 100% sure how to answer them.
Ah, detective work. There is nothing more tedious or rewarding. Here's a process that might help. First, find out who has been meddling with the Web site. Try to get an idea of an organization's sense of Web site ownership. This will help you establish where your site modifications are coming from (sometimes it's spread out across an entire organization, which can create huge challenges). Now would be a good time to whip out the questions you wrote down and interview the builders. I don't recommend the "What the heck were you thinking?!!!" approach. For some silly reason, that question usually gets a rather negative, somewhat heated response. Just go through your list, perhaps expanding your questions to promote a better dialogue. Listen. Understanding the motivations for the choices of the original developers unlocks a world of information (you may even learn a thing or two about the management). You'll learn a lot more than you started out looking for and this will enable you to build a successful plan for moving forward.
Revised: May 16, 2000