Friday, February 14, 2014

Week 3 - Labeling in Web Information Architecture

Labeling, simply put, is one of the various forms of representation available out there. Large chunks of information are represented on websites through labels, with their primordial goal being simply to communicate information efficiently, with meaning, without taking up too much of a page’s vertical space or a user’s cognitive space.  Most web developers or anyone developing content for a web site creates labels without even realizing it. This usually makes us to take the labeling act for granted, thereby leading to poor labeling on web sites to the detriment of the end users of these sites. IAs should therefore take more time and pain to create good, informative and useful labels because these are often the most obvious way to clearly show the user your organization and navigation systems.

There are various varieties of labels:
·       Contextual labels – this is usually a series of hyperlinks that lead to chunks of information on other pages or to another location on the same page.
·         Headings – they simply describe the content that follows them
·         Navigation system choices – they represent the options in navigation systems
·         Index terms – keywords, tags and subject headings that represent content for searching or browsing.
·         Iconic labels – they represent information in much the same way as text can, but constitute a more limited language than text; hence they’re more typically used for navigation systems or small organization system labels, where the list of options is small.

Designing labels might prove to be the most difficult task an IA ever performs. This part of the job has proven to be more of an art than traditional science, hence the more it’s performed or rehearsed, the better the architect will become in realizing it. Several guidelines are available in designing good labels. The IA should always keep in mind that as with all aspects of an IA, content, users and context are of prime importance. Some of these guidelines include:
·         Narrow scope whenever possible – try to limit the audience and stick to fewer subject domains
·         Develop consistent labeling systems, not labels – labels are systems in their own right & should be designed as such, taking into consideration style, presentation, syntax, granularity, comprehensiveness and audience.

When designing category names for your labeling system, there are several dos and don’ts to take into consideration:
·         Clever category names cause doubt and hinder site exploration.  The more confident people feel about their decisions, they more likely they are to engage with your website.
·         Descriptive category names have a higher chance of being discovered and clicked on than clever made up words or internal jargon because people understand them.
·         Check for overlapping categories
·         Use classification schemes that communicate attributes your users can decipher.
·         Explicitly state the difference between options at the fore-front of things.

There exist various sources of labeling systems:
·         Your site – an existing website with labeling systems can be your first stop to your label design. You might want to capture all the existing labels in a single document as a reference point for editing and redesign.
·         Comparable and competitive sites – if you have to develop and design from scratch, you might want to carry out a needs assessment from the end users, and then try to judiciously study comparative sites with similar needs as the first reference point to starting your labels design.
·         Controlled vocabularies and thesauri – these are professional vocabularies often readily available and have been designed for broad usage. So rather than trying to reinvent the wheel you can copy from what others with library and subject-specific backgrounds have already spent judicious time on. Some of these sources are:
o   Taxonomy warehouse
o   ThesauriOnline
o   Controlled vocabularies
o   Web Thesaurus Compendium

Creating new labelling systems from scratch might not be the most enticing of tasks to be performed by an IA, but we generally do not have a choice when there exist no prior labelling systems or when you need to do more customizing than expected. The most important sources in this case are usually your content and your site’s users. You’ll have to carry out crisp content analysis, consult with content authors and also consult with advanced users or user advocates who can speak on the user’s behalf.

After all the above has been effected, there still remains a lot of tuning and tweaking to get your labels/labeling system right, correct and efficient. Tuning and tweaking will involve processes like first, sorting the list of terms alphabetically, review the list for consistency of usage, punctuation, letter case, etc…, which terms to include in the labeling system and are they comprehensive enough to encompass all the possibilities that your site may eventually need to include? You should also plan ahead for scalability and consistency within the labeling system.

No comments: