Information Architecture is the art of building a structure for managing information. For example, you might use it to organize your company website's content into groups and create an interface to support those categories. Information architecture has unlimited application in the knowledge universe - you can use it to organize data on your system, make your email more manageable or simply compose a document. The ultimate objective is to provide users a smarter and easier way of assimilating information.



Metadata and classification are part and parcel of information architecture. Classification might sound dull as ditchwater, but it imparts the cutting edge to data retrieval. While expert classification takes years to master, there are a few simple ground rules that can be put into practice right away.

• Arrange the information in a document in such a way that it communicates rather than informs. Merely loading data does not do the trick.

• Structure the content in a manner that the user finds important points quickly.

• Having classified mountains of data, provide a facility whereby the user can easily create a new document by extracting information from several sources at one time. You would need tools such as Extensible Mark-up Language (XML) for this kind of work.

Information architecture has been the subject of many books, including “Information Architecture: An Emerging 21st Century Profession” and “Information Architecture for the World Wide Web: Designing Large-Scale Web Sites”. However, if we were to peel away the technical nuances of designing information architecture and focus on the practical aspects, we would be left with the following invaluable insight.

Set unambiguous goals: Clearly define the purpose of the information. Is it mainly for personal or business use or both? This will determine the level and type of classification you need to develop.

Don't be in a hurry: Consider each classification carefully and take time to review it objectively. Examine it from the users' perspective. Is this the way they are likely to look for information? Are there too many confusing categories? Too few? Also, design it with an eye on the future, and not based only on current requirements. This will require a broader perspective of the subject under study. Remember, speed is the enemy of quality classification, and you'll get nowhere if you don't get it right. Do not classify for the heck of it.

Don't change ships midstream: Once you've devised your master plan, stay with it. Don't keep changing your classification rules at every opportunity, as that only wastes effort and creates confusion.

Put it to test: The proof of the pudding is in the eating, and it is best that the chef lead by example! Take a trial run yourself before you spring it on others. Did you find what you wanted easily? Did it match your expectations? Was the time it took you to find the information justified? Software development templates like “Verification and Validation Plan” and “Test Plan Template” can also be of help. And most important, don't be satisfied until you receive good feedback from the actual end users.