A disciplined approach to tags, categories and keywords for better user experience

When search engines first made their mark on the Internet, there was a large barrier to entry for website producers. The knowledge for building a web site and the cost associated were substantial, creating a small semblance of order among web site structures.

But as these barriers lowered, the Internet became the wild west – causing search engine companies to refine their practices, better serving their users. One of the most misunderstood legacy aspects of the early days is keywords – which are also known as tags or categories.

What Keywords Were, a History

In the days before Google began to reinvent search, engines used exact keyword matching to return results. The results basically became an index of that keyword. Didn’t list that eighth term that could describe your content? No one searching for it would find you.

For content creators, it became in their best interest to stuff as many keywords in as possible, but next came the spammers. Keyword stuffing became known as a form of “black-hat” SEO. Now search engines will penalize your ranking for listing too many keywords.

What Keywords Are Now

Since 2009, search engines have all moved beyond this keyword matching. Current SEO looks for keywords within meta descriptions, title tags and page copy – this would be user-facing text. No longer is hidden text considered helpful to your site.

Additionally, Google has been very transparent about it’s development of semantic search – with an algorithm that completely ignores keywords.

Semantic search determines the context of a page based on the language and words used. It could be described as understanding the idea behind a page, and it matches that idea to the intent of the user doing the search. This involves not only the results given to a search user, but the way a user can search.

Instead of “best pizza in Washington D.C.,” a user can ask “Where is the best pizza near me?”

A Keyword Approach to Taxonomy

Still, semantic search does not negate any benefit that websites might gain from smart use of keywords. Rather, a website with a well-structured taxonomy can greatly improve usability and even impact SEO. When creating a piece of content, rather than thinking “what keywords can I add to this post?,” the best frame of mind is to think “where does this post fit into my site structure?”

Let’s take a look at what this might look like, using the WordPress CMS as an example.

In WordPress, the highest level taxonomy is called “categories.” A direct parallel to news websites would be sections, such as “news,” “sports,” and “entertainment.” These are buckets that should be created when planning a website and should not change unless adding a new feature. A content producer should never create a category on a whim.

This level might be simple to understand, but the next level of “tags” is often abused. As mentioned earlier, even a tag system should be planned out conscientiously prior to publication. Here are some examples of wise use of taxonomy:

  • A national sports website adding tags for each team they cover
  • A local news website adding each town or city in their coverage area
  • A food blog adding the type of cuisine to recipes
  • A political site determining a fixed set of of topics which might be mentioned

The latter of these is open to editorial interpretation, while the former is a strict set of teams. Those taxonomies which require editorial choice are the ones which are in most need of a pre-determined structure. Lack of structure and a laissez faire approach only lead to poor user experience.

Benefits of Taxonomies

“User experience” is a very general term here when talking about the benefits of a smart taxonomy, but a user’s experience is affected by several results, which follow in turn.

Easier to find related content by preventing duplicate tags

How many terms can you think of related to the phrase “illegal immigration?” Abiding by a strict set of terms will keep content with related items. If we were to add “illegal aliens” and “unauthorized immigration” as terms, we would have three tags that are the same thing.

Prevents the creation of useless tags

Tags are meant to lead users to related content. They become stronger and more useful as more pieces of content are added. With no structure, our system ends up being loaded with dozens of tags that have only a single article. Imagine a user’s disappointment if he or she finds no additional reading contained therein.

Avoiding tag overload

Remember that the “old” way of assigning keywords created a benefit for higher quantities. Doing so now only lessens the usefulness for users. A growing list of tags becomes visual junk to users. Here’s an example of what to avoid:

Bots used for SEO can follow a smart structure more easily

Google’s semantic search benefits from one strong page regarding each subject – they work best if a site has a clear hierarchy. That said, you won’t be penalized be search engines if your content appears in multiple archives.

Multiple Taxonomies Work Better

In many content management systems, including WordPress, it is possible to create additional custom taxonomies. This is a best practice when creating a usable system, but we’re also getting a little advanced here.

Let’s combine some of our previous examples. A local media outlet tagging local municipalities would be one taxonomy, while an additional could be the sport.

This might be good for an editorial process because it may encourage writers to pick one selection from each taxonomy. In this situation, adding additional tags works because we are appealing to two different situations in which a user may want to access the information – by locality or by sport.

If a site is programmed well enough, we could allow users to select one element from each taxonomy, showing all baseball stories in Woodridge, for example. This goes beyond simple editorial decisions, however, and involves several complexities beyond the scope of this paper.

Designing for Taxonomies

With a well-planned structure in place – and its use implemented by writers and editors – the final step is to ensure that the design within our site is simple to access and comprehend.

In WordPress, for example, many templates simply list tags following the content of the post – leading to the aformentioned “visual junk.”

A design such as this does not necessarily encourage readers to visit more content for any of these tags; nor do they convey that more content actually exists. Instead, a tag could trigger additional content on the page.

Related headlines

With our primary taxonomy, programming a post to display other recent headlines in the same tag or category would provide more encouragement for readers to stay on the site.

Maps, logos, or other visuals

For our sports or location taxonomies mentioned previously, programming an outbox to appear adds an additional art element to the post – as well as providing context.

Displaying data

If we really want to delve into adding additional context, we could program something that connects to an API, or a database we maintain ourselves.


The ultimate goal of content is to make information accessible by readers. A properly built and maintained taxonomy can go a long way toward this goal – keeping users on site longer, and helping build brand loyalty.