Top 10 Tagging Best Practices for Anything Web 2.0
The nice thing about having a fledgling blog like mine with a readership in the single digits is that I can pretty much make wholesale changes to my blog and most people won’t notice a thing. And since tonight I had a couple hours to kill, I decided to clean up and revamp my Wordpress Tags and Categories.
Like any resourceful latecomer to the blogging game, I googled for other people’s tagging best practices. Perhaps because tagging is such a nebulous and subjective art, there was not a lot of information I can find besides a couple useful articles by Ian Beck and Nick Santilli. The most common advice is to be consistent, which I totally agree. Second most common advice is to find something that works for you. Umm… OK, I’ll try.
With tagging so prevalent in anything web 2.0 today, my goal is to find a tagging system or best practices which not only can be applied to blogs, but other things as well, from personal file organization to social bookmarking.
With that, here are my Top 10 Tagging Best Practices for Anything Web 2.0:
- Be consistent. This rule is so important that it needs to be repeated here again. Whatever happens, pick a system and stick to it.
- Always use lower case letters. No exceptions. And spaces and hyphens are the only punctuation that I will use to separate different words in a phrase.
- Use nouns whenever possible. I always prefer “stupidity” to “stupid”, and “developer” or “development” (depending on the context) over “develop”.
- Try to limit the use of abbreviations. Although the goal is to make the tags as succinct as possible, I will still expand out abbreviations unless they are universally recognized either as a brand (like IBM or HP), or an industry buzzword (SOA or GTD). Ambiguous abbreviations such as KM or PR I will expand them out as “knowledge management” and “public relations” respectively. The only exception is for extremely long abbreviations like “tokyo international anime fair”, in which case I will use the shortened form “taf”, since the abbreviation is used frequently to market the event as well.
- Tag important company and brand names. If my blog entry (or article I’m submitting to a social bookmarking site) talks about any companies, I will include them as tags. The only exception (based on my discretion) is if the company was only mentioned in passing as an example and is not central to the main idea of the article.
- Tag important product names. I will tag product names that appear in the blog entry, but I will strip out the vendor name in the tag and apply that as a separate tag. For example, if my article talks about IBM Lotus Notes, I will tag the article separately with the “lotus”, “ibm” and “notes” tags instead of “ibm” and “lotus notes” or even worse, just “ibm lotus notes”.
- Tag ideas, concepts, locations, and events. Any main ideas, concepts (however nebulous or concrete), locations or events must be tagged and captured. If the article has three sections each with a different idea, try to extract a tag from each section. Examples: “productivity”, “social graph”, “enterprise”, “singapore”, “blogs”, “security”. If I am tagging an event, I will always put the year of the event if it’s an annual event (like using “taf 2008″ instead of just “taf”).
- Don’t tag individuals. I will never tag a blog entry with another person’s name, even if I talk about them or link to their blogs. This is simply a matter of personal choice. I feel that linkbacks are more than sufficient to give them credit if I do refer to their work. The only exception is if the person is a central idea for my blog. E.g. if my blog was a blog on Microsoft, Bill Gates will be a central idea, and thus I wouldn’t mind using the tag “bill gates” when talking about him.
- Use a plural base. I will always choose “blogs” over “blog” or “blogging”. There has been some debate on whether to use a singular base or a plural base as a default, but to me, a plural base just sounds more natural, especially when tags are mostly being used as categories in the Web 2.0 world.
- Don’t use more than 10-12 tags per entry. This can be an entirely arbitrary number up to you. For me, if I need so many tags to accurately provide the metadata for a blog entry, then it may be a sign for me to break up the article into smaller, manageable parts.
- Bonus Rule: Remember that tags are metadata that helps other people find your content using search engines. So think from the point of view of the user – If I am to type in search terms using a search engine to find this content, what search terms will I likely use? Those can also be considered as tags.
So these are my tagging best practices. How about yours? Please leave me a comment if you have your own set of tagging best practices that you want to share.
Last note on when to use Categories and when to use Tags in Wordpress. In general, I subscribe to Lorelle’s school of thought: Treat Categories as your blog’s table of contents, and Tags as your blog’s index. Categories help you organize your posts into manageable sections in your blog, and Tags help your readers do a search to locate the specific information that they are looking for.
Update: About eight months after I wrote this article, I have changed my opinion on some of the best practices above, specifically the plural base (#9 above) and number of tags to use (#10). Anyway, at the end of the day, it is what works for you.

Hey Jonathan! I really like this post. I frequently struggle with what to tag my entries, so this helps. I’m curious about #6 important product names. Why tag “ibm” “lotus” and “notes”? instead of ibm lotus notes? I don’t really understand the rationale. It seems to be that someone searching specifically on that would google “ibm lotus notes”, which is what you advise against tagging?
Also regarding #9, why choose “blogs” over the other terms? why not just tag all the terms? I must admit I’m not sure how these may or may not be picked up by searches, so would love to know more =)
Thanks Daryl, for your feedback.
For rule #6, the rationale is basically to keep the tags as succinct as possible. Ideally, I want to avoid repeated words across different tags. Imagine if my blog is about enterprise software. If I don’t strip out vendor names in products, I will soon have tags such as “microsoft exchange”, “microsoft outlook”, “microsoft office” and so forth. Not to mention the mandatory “microsoft” tag which I need to distinguish between content between Microsoft and IBM, for example. That’s an awful lot of “microsoft” in the tag cloud, don’t you think?
From an SEO perspective, having one tag of “ibm lotus notes” should be similar in search performance to tagging each word separately. However, if you break up the tag phrase into shorter word tags, the added benefit is that should your readers search on just the tag “lotus” in Technorati or Google, the chances of your post appearing should also be much higher than if you tag it with the full phrase “ibm lotus notes”.
As for rule #9 – I generally want to use as few tags as necessary to describe the blog entry, and the plural base form of the noun sounds the most natural to me. Let me try to illustrate: When I tag something, I am basically filing it in a particular category or group. Now what sounds more natural to you in spoken English?
“This tag represents a group of shirts”, or “this tag represents a group of shirt”?
Search engines in general treat the plural and singular forms of search terms as the same, so this is entirely a personal choice. Same goes for “blogging” vs. “blogs”. In fact, do a Google search twice – once on the term “blogging”, and once on the term “blogs”. Notice that the results you get are more or less identical?
I get #6, still a bit iffy about #9. I must say I feel enlightened about search a little bit! How do we know that the results aren’t identical because those bloggers tagged it as both “blogging” and “blogs”?
Actually, for Google Web Search, it seems that their algorithm is smart enough to know that when you are doing a search for “blogs”, you will likely be interested in results for “blogging” as well. So this is handled for you automatically.
However, if you look at Technorati or (surprisingly) Google Blog Search, you do get different results. Technorati somewhat mitigates that by showing related tags. Google Blog Search I presume puts more emphasis on blog content than tags, which I guess can help explain the different results.
At the end of the day, I guess tagging still comes down to personal preference. For me, I prefer a more concise tag cloud, and thus I like to just pick one tag variant and stick with it. To me, yes – tags definitely have their SEO value, but more importantly, their primary function should be to help readers search for the right content they are looking for *after they arrive at your site*.
Therefore, I rather have a cleaner tag cloud to give users a better user experience, even if it means I miss out on some search traffic.