Mr. Jeff Starr, the WordPress developer behind the Wutsearch Search Engine launchpad and a couple of plugins that we use at The New Leaf Journal, published an instructive post on the differences between taxonomies, categories, and tags in WordPress (note we are referring to WordPress as a content management system). This article came to my attention in my feed reader because I was in the process of reviewing all of our tags and ensuring each had meta descriptions when I came across it (that process is now completed). My work on our tags came not long after I re-organized our Categories in late 2022. With Mr. Starr’s article serving as my inspiration, I will explain my thoughts on how to best use WordPress categories and tags and how my thinking on the issues has evolved over three years of running The New Leaf Journal.
Explaining WordPress Taxonomies, Categories, and Tags
By default, WordPress includes three taxonomies, that is, ways of organizing individual posts: Categories, Tags, and Post Formats. We use Categories and Tags at The New Leaf Journal. While WordPress offers a number of custom post formats, we only use regular posts and pages here (one could argue my Leaflet, Leaf Bud, and Letter custom post types have some taxonomy characteristics, but custom post types are a distinct issue). It is possible to create additional taxonomies – through a plugin or manually. We once used a plugin to implement Post Series as a separate taxonomy, but I abandoned it because I could not make the archive pages display correctly with our theme.
Explaining WordPress Categories and Tags
Mr. Starr aptly explained the difference between the taxonomies that we make the most use of: categories and tags.
[B]oth Categories and Tags are types of Taxonomies. The only technical difference is that Categories are hierarchical, while Tags are not. So with Categories, you can create sub-categories (or child categories). With tags, you cannot. Tags always have a ‘flat’ organizational structure.Jeff Starr
There are no rules on how one can use categories and tags, if they are used at all. For example, Mr. Starr only uses tags to organize his posts, while we use both categories and tags. The only technical difference between categories and tags involves hierarchy (many WordPress themes handle how categories and tags are displayed differently, but that is beyond the scope of this article).
(Note: WordPress technically requires every post to have a category. It does not require posts to have tags. Uncategorized posts go, by default, into an Uncategorized category. However, as Mr. Starr shows, a site can easily rely on tags exclusively by allowing all posts to go to Uncategorized.)
While there are no rules governing categories and tags, their structure suggests their ordinary purposes. Mr. Starr describes categories as being “used to broadly organize posts into groups” and tags as being used to “denote any specific post characteristics.” For example, I have a sub-category of my Visual Novels category called VN Reviews. This category is for all of my articles reviewing visual novel games. I use tags to describe aspects of these visual novels. For example, I have tags for three visual novel scripting engines – ONScripter, KiriKiri, and Ren’Py – which I apply to visual novels that use one of the three languages. These tags can also be used on non-reviews. For example, I have written several guides to using ONScripter on Linux which have the ONScripter tag but are not in the Visual Novel Review category.
The purpose of categories and tags
We have explained categories and tags in a technical sense. Moreover, we have explained what they are generally used for. But here I pose a different question: What is their purpose?
Webmasters should use categories and tags to improve the experience for site visitors. That is, every category and tag should be useful to a guest. Because this post is about The New Leaf Journal, I will use our humble site as the example.
We have published more than 800 articles in three years. Unlike many sites, we are not restricted to a single topic. I may re-print an old poem one day, write about an anime series on the next, and then publish a post on Roman history. While I hope that you enjoy all of our posts, I will venture that most readers may be interested in some of the things we write about but not others. It does not follow that someone who enjoys my report on BLOB DYLAN graffiti in Bushwick will be interested in my essay on citizenship and nationality in U.S. immigration law. Someone who comes for Victor V Gurbo’s article about building a Big Joe Williams-inspired nine-string guitar may not be interested in his Nintendo Switch game review. There are many such cases at The New Leaf Journal.
The main purpose of categories and tags is to help our guests find articles that interest them. The anime fan who finds our site through my post on hair color in Kimi ni Todoke can use our Anime category to find other articles about anime or our Hair Color and Styles in Japanese Anime, Manga, and Games tag to broaden his or her knowledge of a growing academic field. Categories and tags can also serve a broader discovery purpose. For example, our hair color and styles tag is very specific. That tag will turn up articles focusing on the same subject matter as my Kimi ni Todoke piece. The Anime category is a bit broader – the reader will still see anime posts, but most of my anime articles are not mono-focused on hair color and styles. But there is one broader tag – Japan. The Japan tag includes many non-anime articles. Moreover, I do not apply it to every anime article – only those which discuss Japan in some way outside of the anime context. Thus, following the Japan tag link would allow the reader to discover non-anime articles featuring Japan in a meaningful way.
The essential purpose of categories and tags at The New Leaf Journal is to help our guests find their way around our site and make their own unique New Leaf Journal reading experience. I will add here that both categories and tags have RSS, ATOM, and JSON feeds, which allows readers to fine-tune how they follow The New Leaf Journal on their own terms.
Categories and tags are not the only way to fulfill these goals. For example, I regularly include internal links in my articles – when relevant and appropriate. These links serve as a way of discovering new articles, including Leaflet and Leaf Bud short posts which exist outside of our category/tag system. I also (as of May 5, 2023) include nine algorithmically generated “related posts” (relatedness may vary – refining that is on my to-do list), which include our Leaflets and Leaf Buds along with regular articles. Finally, once a reader has a better idea of what our site has to offer, he or she can use the search box to look for specific words in all of our articles, Leaflets, Leaf Buds, and pages.
Our category and tags history
When I first started The New Leaf Journal, I lacked foresight. My original category concept featured non-topical categories. However, as we amassed articles, this became useless. What good is an “Essay” or “Discourses” category with a ton of posts? How is a catch-all Reviews category without further sub-categorization helpful to users? I realized as we published a significant number of articles that my original category concept was a failure.
Back in (I want to say mid-to-late) 2021, I completely re-did our categories and tags, starting de novo. My initial efforts with respect to categories involved creating more topical sub-categories to organize content, while not totally abandoning my original vision. For tags, I went through all of our articles to create tags based on a large sample of published material (instead of my guessing which tags may end up receiving multiple articles). I eventually tried to fold series (having abandoned our series-creating-plugin) into categories and tags.
In 2022, I switched our SEO solution from Yoast to The SEO Framework. I decided to allow search engines to index our categories and tags after having previously designated them as no-index. I know that there is some debate over whether this is best practice, but we can set that aside for now. What is mostly beyond debate is that if one allows tags and categories to be indexed, he or she should also, at a minimum, provide meta descriptions for search engines to consume. I had neglected to write meta descriptions for everything in my 2021 reorganization because at that time I had given categories and tags no-index tags. I only slowly added meta information to tags and categories after allowing them to be indexed.
I nibbled around the edges of our categories and tags throughout 2022. By late 2022, I decided that my 2021 innovations were no longer working. I sat down and restructured our categories in December 2022, almost wholly eliminating our general-purpose categories in favor of topical categories, and further sub-dividing large topical categories into smaller, more specific categories, some of which are similar to what I had once organized into series. There were a few instances wherein I converted a broad, category-like tag, into a category. I am now looking for new opportunities to further sub-divide big categories (e.g., our category for NLJ photography) in order to ensure that our category structure aids readers in finding articles that they are interested in reading.
The category project was irritating, but not overwhelming because we do not have too many of them (there are 63 categories as of May 8, 2023). Tags were another matter. I was aware that we had some useless tags, awkwardly named tags, and a couple instances of duplicate tags. The problem was that there were more than 400 tags. You may imagine how dealing with tag problems in a big way is daunting when there more than 400 to review.
In early 2023, our site was blacklisted by Microsoft Bing before being fully de-indexed by the same. While I am confident that it had nothing to do with our tags and categories (see my assessment), Bing highlighted that many of my tags, which were indexed, lacked meta descriptions. While many guides recommend designating tags (and categories) as no-index for search engines, my tags are already indexed and I am disinclined to rock the boat at this time (especially since we are doing fairly well, by our standards, outside of Bing). Thus, I decided to embark on a task to write short meta descriptions for every New Leaf Journal tag. Even if I decide to designate most or all of the tags as no index at some later stage, it does not hurt to have brief descriptions for each.
Over the course of two weeks, I wrote a meta description for each of our tags. This process necessarily required me to review every one of the 400-plus tags. I found a few that I could delete and others to rename or consolidate. I also added a couple of new tags that were appropriate. While going through this process, I thought about how to ensure that all of our tags would be useful for readers looking for interesting articles based on whatever they were reading.
My tag re-ordering left us with 400 tags as of the publication of this article (less than we had before, although I forget the exact number). We have 21 tags with only a single article. While single-article tags should be disfavored, I determined that I would be returning to the one-tag topics in 2023. In a way, my surviving one-article tags serve as a small to-do list.
Figuring out how to best use categories and tags has been a process for me here at The New Leaf Journal. My original solution, which featured extremely broad categories and specific tags, rendered the categories pointless. Over my three years running The New Leaf Journal, I have improved upon my approach and we now have a system which I think adds value for visitors.
It is hard to offer specific category-tag lessons based on my experience since every site is different, and most writing sites are a bit more targeted in what they cover than The New Leaf Journal. As a general rule, webmasters should think first about how to use categories, tags, or any additional taxonomies to add value for someone navigating their sites. That is, it is important to ask how a site’s category/tag/taxonomy structure actively aids readers in finding interesting or useful articles. I see that some guides reference search engine optimization. While this is important for visibility (I noted how it factored into my thinking), the first priority should be creating a site that someone visiting from a search engine will find interesting or otherwise worthwhile.