We've been developing successful Internet solutions for over ten years. During this time we've discovered what works and what does not when it comes to interactive design and Web communications. The following best practices demonstrate the breadth and depth of experience that we bring to the table. This listing of 12 best practices is not our formal methodology, but rather a POV on the things that we have proven will improve a project's chance of success.
1. Get outside your box
Designing a Web site should involve far more than simply creating the visual design component. Behind the mystery of the Internet, there are really only seven major things you can do with the technology. Any Web design effort should examine those seven things in conjunction with your goals and objectives (see next rule). You should be asking yourself how you can improve your company using Internet technology. How can we use the Web to move more business through the channel and increase revenues or lower the costs associated with developing, selling, and supporting our product/service. Use your brain. Think.
2. Set goals, objectives, and success criteria
Goals help turn dreams into reality. If you want to be successful in your Web design efforts, you need to have a clear understanding of what you are trying to accomplish. Before you ever start a project, every person on the team needs to have a clear picture of the objectives at hand and how they are responsible for contributing to the projects success. And just as important to setting clear goals, is establishing predefined success criteria that can be measured in quantitative terms. Remember that success isn't random. You can't hit targets that you don't know are there.
3. Balance long term vision with iterative releases
While it is important to have a long term vision for your Web initiatives, it is just as important to move toward that vision in a series of iterative site releases. You cannot do everything at once. Break down your intentions down into a series of feature function sets and release them in phases. Taking this approach will help you get to market quicker and will provide you will valuable feedback for later efforts. Take small bites.
4. Design a tool for the actual user
Think like an actual user. Who is the target audience for the site and what things do they need in their day-to-day activities? What appeals to them? What will they find useful? Too often, web design efforts are influenced by the personal preferences by someone on the Client's side (colors, style, functionality, layout) rather than by focusing on the wants and needs of the target user. If you want to be successful, you should design things from the user's perspective.(Also see rules #7 and #11). People who claim to know what the user wants but don't have the evidence to back it up, are simply guessing.
5. Set and manage expectations
Managing expectations is critical. And we are not just talking about managing the expectations of the primary client. We are talking about managing the expectations of everyone involved in the project as well as management on both sides. Everyone needs to understand schedules, process, milestones, deliverables, scope, changes in scope, open issues, and unforeseen developments that change the project in any way. To this end, setting up and maintaining a central project extranet is an important component in managing expectations. (Also see rule #9)
6. Consider the technical issues early
From a technology perspective almost anything is possible, but technical issues can be tricky. What may seem simple in concept could turn out to be a significantly complex and costly undertaking. On the flip side, some seemingly complex functionality may be extremely simple to implement. Complicating things further are the advances in technology itself. What was difficult six months ago might be simple today. Only a true software engineer (not the self taught type), who is up to date on the latest technologies AND who has a clear understanding of what you are trying to accomplish, is qualified to give you a technical perspective on cost, complexity, and timeframe.
7. Talk to existing users
If you have an existing website, talk to your current user base. Ask them to participate in your redesign efforts. If this is your first Web site, then talk to existing customers about what they like and dislike about the Internet. Remember that this is really about them. It's about their true needs and wants. It's about what they find useful, appealing, and compelling. These are the people you are trying to serve. It is better to listen to users directly, rather than listening to a possibly skewed perspective of what someone within the organization "thinks" the user wants.
8. Study the competition
Doing an in-depth competitive analysis is one of the most strategic and insightful research activities on the Internet. It is amazing what you can find if you know where to look and how to read between the lines. At a surface level you can evaluate the competition in terms of their online initiatives, site features/functions, and the type of impression they are making on your "shared" target customer. At a deeper level, you may find holes within their organization, weaknesses to be exploited, and product details that management would rather keep under wraps. On the flipside, you should be careful not to make any of these same mistakes.
9. Keep the communication channels open & flowing
Regular and structured business communications and project reporting are critical to keeping a project on track. Don't try to hide a growing elephant. If there are issues or problems, shine a light on them and try to get them resolved before they become unmanageable. Using a project extranet with varying levels of access, is a good method distributing information such as project parameters, statements of work, change requests, team members, resource plans, invoices, status reports, deliverables, milestones, issues, problems, and workarounds.
10. Don't underestimate content requirements
Within the world of Internet professional services, almost everyone would agree that content is a killer. Content refers to the information presented on a web site (text, pictures, video, sound, illustrations, animations, etc.). Whether the client is providing raw content that will be reworked for the web or the client intends to provide Internet-ready elements (which is seldom possible), content related issues are responsible for over 50% of the delays experienced on a project. To avoid this problem it's important to take the content requirements seriously and come up with a checklist, assignments, and a plan for meeting the content needs.
11. Conduct formal usability analysis
There is no substitute for real live usability testing. There is something eye opening about sitting behind the scenes and watching someone fumble through your Web site trying to accomplish a series of seemingly simple tasks. It becomes clear very quickly if your site works or it doesn't. Formal usability testing lets you see if your user interface is really connecting with a user or frustrating him/her to the point where you are driving them away from your Web site and possibly toward the competition. Don't skimp here. Test your existing site and test your new site ... well before the anticipated launch date.
12. Quality Assurance is mandatory - test, test, test
Most Web initiatives get down toward the end of a project and then start to think about testing their work against various browsers and platforms. To be blunt, this is not what testing is about. Quality assurance and quality control should be embedded throughout the development process. Good QA saves time and money. Web sites and Web applications should conform to a test plan that takes into consideration the computing environment of your users (hardware, software, and network configurations).
