The human resources department of an international asset management corporation requested assistance in evaluating their site content to prepare for a migration into the new, internal corporate internet platform. As it was, the site's primary organization rested on providing regional versions for Australia, Hong Kong, Japan, Luxembourg, New Zealand, Singapore, the United Kingdom and the United States.
The site map above displays nearly every asset for every region. (Page titles have been removed.) Note that assets (i.e., boxes representing a web page) with only one "flag" in it, a colored smaller box with a regional abbreviation, represent an asset unique to that region. Each region was also assigned a unique background color for the parent box to highlight clusters of unique regional assets. (Exception: Australia and New Zealand were combined.)
As a result, high-level patterns reveal themselves immediately. For example, the far right column represented policies and procedures. As expected those assets were more uniform than not. The sprawling third column, on the other hand, reveal just how particular benefits information were, especially for the U.S. The great number of orange boxes for the Japanese variant show how much of their content was written in Japanese.
Assets with multiple flags represent pages or content shared across regions. The different colors of the flags indicate how "identical", "similar", "distinct" or "unique" that content may be relative to the other regions. Assets without a colored flag and regional abbreviations in a bold black font show that the asset existed but was void of content. Assets without a colored flag and regional abbreviations in a bold gray font indicate that no such asset existed for that region at all.
How in the world could all of that content be evaulated quickly and effectively?
In the simulation above, note how the text blocks show increasing differences from left to right, top to bottom (i.e., "identical", "similar", "distinct", "unique"). Each of those text blocks outlined in green represent two copies of text copied and pasted on top of each other into Visio. The more or less they matched up, the more or less similar or distinct they were despite purporting to be the same piece of content. (You didn't think I actually read the entire site, did you?!)
After evaluating and mapping the content I identified five major types of content shared by all regions. Organizing the site content into those major areas (or "primary navigation") came easily. Please see the proposed site index below, and compare its low-contrast simplicity to the kaleidoscope above.
After evaluating and mapping the content I identified five major types of content shared by all regions. Organizing the site content into those major areas (or "primary navigation") came easily. Please see the proposed site index below, and compare its low-contrast simplicity to the kaleidoscope above.
Because the new internet platform presented pages in a highly structured interface, I was also able to indicate in the proposed site map where certain types of content would be located in the new UI, including how to accommodate regional variants.
In order to demonstrate that the proposed site index included all existing content, the numerical ID for each asset (in gray ovals) from the new was mapped to the original site. Please see below.
Note how the proposed site index serves as a "Key" for mapping the new to the old and sums up the proposal. (The image below is an enlargement of the lower right quadrant of the site map above.)
