SEO And The Taxonomy Of Topics via @sejournal, @martinibuster

Communicating with expertise and authority is a top priority and there’s no better starting point than outlining your content with a topic taxonomy as a way for creating content that has the best chance of ranking.

A taxonomy is a system for classifying, organizing, and naming things based on shared characteristics. It is used in science to categorize things but it’s a word that is also used to categorize anything, including topics, making it an effective SEO concept to understand.

Topic Taxonomy

In the context of content, a topic taxonomy refers to the organization of information into different topics and subtopics. Organizing content this way creates an outline of the inherent topics and subjects covered within a webpage of content.

The goal of a topic taxonomy is to provide a structured framework upon which to craft a meaningful document that is semantically relevant to a given topic.

This is a way of looking at content from a high level, seeing the forest, not the trees.

How To Build A Topic Taxonomy

Technology like deep learning, neural networks, and RankBrain have helped Google evolve away from keywords and toward understanding the topics that are in content.

Search queries are about topics can can be categorized by topic like this:

Beauty > Makeup > How-to

A webpage can be categorized in a similar way. Webpages are rarely ever about one topic because there are always subtopics.

So, when thinking about a topic, you can also think about what the parts of that topic are.

If it’s a clothing product page, then the parts of that topic (subtopics) are:

• size
• material
• fit
• fabric
• care
• description
• reviews

Those subtopics are the kinds of things a consumer expects to see when visiting a webpage about clothing.

If the webpage is about how to build links then the subtopics of that page can be:

• research
• analysis
• outreach
• templates
• subject lines

Google Identifies Topics

Google’s John Mueller once remarked about headings that they’re about communicating what the topic is about.

Mueller said:

“And when it comes to text on a page, a heading is a really strong signal telling us this part of the page is about this topic.

…whether you put that into an H1 tag or an H2 tag or H5 or whatever, that doesn’t matter so much.

But rather kind of this general signal that you give us that says… this part of the page is about this topic.

And this other part of the page is maybe about a different topic.”

Google’s Martin Splitt talked about this process in a 2021 seminar where he introduced Google’s internal concept of webpage topics, called the Centerpiece Annotation of a webpage.

Martin explains:

“That’s just us analyzing the content and, I don’t know what we have publicly said about this, but I think I brought it up in one of the podcasts episodes.

So I can probably say that we have a thing called the Centerpiece Annotation, for instance, and there’s a few other annotations that we have where we look at the semantic content, as well as potentially the layout tree.

But fundamentally we can read that from the content structure in HTML already and figure out so “Oh! This looks like from all the natural language processing that we did on this entire text content here that we got, it looks like this is primarily about topic A…”

In the above quote, Martin says there are additional annotations that are related to:

• The semantic content
• The layout tree

What’s important is that Martin Splitt made it clear that Google is looking to identify what the topic of the webpage is.

So, your job is to outline the topic of the webpage and the subtopics that are relevant to that topic.

Example Of Importance Of Topic And Subtopics

I once did one of those live website audits that they do at SEO conferences and this person said their website can’t rank for a search term like Financial Information.

They had the keyword phrase Financial Information in the title, in the headers, in the alt text, all over the place, right?

I looked at the website and the topic is clearly about Financial Advice, not financial information.

So I asked the website publisher why they were trying to rank for Financial Information if the website was about Financial Advice?

And the person says, “Because the keyword volume is higher for Financial Information.”

So, the reason website couldn’t rank for Financial Information, despite all the links and keywords, was because the topic of the content was about Financial Advice.

This is why it’s super important to begin with understanding the webpage topic then outline what the subtopics are based on what the user is expecting to find.

As I think most of us know by now, Google is no longer matching search queries to keywords. Google is matching search queries to relevant answers and the answers are always about topics and subtopics.

Accuracy Over Comprehensive

Lastly, avoid making the mistake of being comprehensive.

Providing a direct answer is always best because conciseness offers the benefit of clarity and it saves time. When people want an answer they want it fast and easy to understand.

This is true for consumer product page or on a review site. Don’t linger on what’s not of direct importance, be precise.

Google is all about precision and being fast. That’s one of the reasons why they chose the word Hummingbird for their 2013 update because the core algorithm rewrite made it perform faster and precise. Most subsequent updates had the feature of being precise.

That’s why it is important to also be precise, not because Google is precise but because that’s what users want, which is why Google focuses on precision.

Thinking in terms of a taxonomy of topics will help you stay focused on being precise and complete without going overboard with too much content and becoming off topic.

Stay on topic with a taxonomy of topics.

Featured Image by Shutterstock/Dim Tik

Google Explains The Index, Follow Meta Tag via @sejournal, @martinibuster

Google’s John Mueller answered a question on Reddit about a commonly used robots meta tag and what would happen if it was missing. Mueller’s answer, while it makes sense and is documented, may still come as a surprise to many publishers and SEOs.

Robots Meta Tag

The HTML meta element communicates metadata. Metadata is machine readable information that a crawler like Googlebot can read.

There are many kinds of meta elements like the meta description element but the Robots Meta Element is different in that it can control the search engine crawlers.

The information communicated by the robots meta tag is called a directive, which means that robot crawlers are obligated to obey the instructions in the robots meta tag.

There are many directions to pass along through the robots meta but the following meta tag is one that is relevant to the question John Mueller answered.

The noindex, no follow meta tag:

The above meta tag tells the search engine crawlers to not index the content on the webpage and to not follow any links.

One of the most common meta tags is this one, which commands search engines to index the content and follow all the links:

While the above meta tag is common, there is a significant amount of misunderstanding about it. There’s a line of reasoning that because Google supports nofollow then it must imply that Google supports the follow directive.

I found a number of authoritative websites that say that Google uses the meta robots index, follow meta tag.

But that’s not actually how Google uses those directives, as John Mueller makes clear in his answer.

What’s The Effect Of Leaving Out The Meta Robots Index Tag?

The person on Reddit asked the following question:

“I’m a bit confused with a website I’m working on.

So, this is what the meta snippets on most of the websites I work on look like:

But, on the website at hand, it’s missing the ‘index’ tag.

My question is: What’s the effect of the site missing the ‘index’ tag.”

John Mueller answered:

“The “index” robots meta tag has no function (at least in Google) – it’s completely ignored. Also “follow”.

Google has https://developers.google.com/search/docs/crawling-indexing/special-tags & https://developers.google.com/search/docs/crawling-indexing/robots-meta-tag to document the meta tags that have functions. You can use anything else, it’ll be ignored. is an option, if you want to throw people off.”

Why Google Ignores Robots Index & Follow

The simple reason why Google ignores the robots index and follow meta tag is because indexing and follow are the defaults.

Indexing and following links is what search engine robots do, they don’t have to be told to index content and follow links because that’s their purpose.

Google’s documentation on robots tags advises:

“The default values are index, follow and don’t need to be specified.”

The full list of valid directives for Google is here.

If the robots meta you want to use isn’t’ listed there then Googlebot is going to ignore it.

Is Index, Follow Completely Useless?

It’s true, documented and official that when it comes to Googlebot, is a waste of HTML space and is ignored by Googlebot.

Bing treats index, follow in a similar way but with a slight difference, as described in the official Bing documentation for meta tags.

This is what Bing’s documentation says about the index directive:

“By default we assume “index”, but if needed you can use to explicitly state that we may index the page.”

And this is what it says about the follow directive:

“By default we assume “follow”, but you can explicitly state “follow” if so desired.”

In my 20+ years of working in SEO creating websites, optimizing them and ranking them, I’ve always considered it a good policy to give the bots what they expect and try not to give them anything that’s unexpected. So if a meta description is not necessary then my impulse would be to leave it out because the whole point of optimizing is to make it as easy as possible for the search engines to index and understand the content, which means to get rid of anything that might work against that goal.

In this case, it’s highly likely it’s not going to have an effect one way or the other.

But… There’s another way that follow and index trip people up.

Some publishers use this robots meta tag:

Some sites advise that if the page isn’t indexed, the use of the “follow” directive compels the search engine to follow the links.

But that’s not true if there’s a “noindex” directive for the simple reasons that Google cannot follow a link on a page that it is not indexed. If it’s not in the index then the links on those pages are not in the index.

Featured Image by Shutterstock/Bangun Stock Productions

Brave Launches Mistral-Powered CodeLLM In Search Results via @sejournal, @MattGSouthern

Brave Search has announced the integration of a new feature, CodeLLM, enhancing search results for programming-related queries.

CodeLLM combines Brave’s search capabilities with the power of large language models (LLMs) to provide relevant code snippets, detailed explanations, and citations for programmers’ search queries.

A New Resource For Coding Assistance

CodeLLM is designed to assist programmers by providing AI-generated code snippets based on search results, with step-by-step explanations and source citations for reference and validation.

It’s built on the foundation of Mistral, an advanced language model capable of generating code from text prompts.

Brave automatically deploys CodeLLM for programming queries and generates a widget above normal web search results.

The widget triggers CodeLLM to return an AI-generated response summarizing key information and providing sample code grounded in the search results.

In an announcement, Brave highlights CodeLLM’s ability to tap into the contextual power of a dedicated search engine in real-time.

According to the company, that’s what sets it apart from generalized language models:

“As the models get more accessible, and more affordable to run, the value-add is the context we’re able to provide to them.”

Availability

CodeLLM is immediately available for all Brave Search users on both desktop and mobile without needing to toggle settings or download additional software.

Brave plans to continue improving CodeLLM’s capabilities and offer its results through the Brave Search API for use by third-party applications in the near future.

About Brave Search

The launch of CodeLLM comes two years after Brave Search was introduced as an independent alternative to dominant search engines like Google.

Brave claims its search engine doesn’t track users or collect personal data.

Brave’s search index draws from the Web Discovery Project (WDP), which contains over 20 billion quality pages, with 700 million pages indexed daily.

Per the company’s data, Brave Search serves over 25 million queries per day.

Featured Image: Mamun sheikh K/Shutterstock

Google Alters Purpose Of SpecialAnnouncement Structured Data via @sejournal, @martinibuster

Google’s updated guidance to the Special Announcements structured data shifts the focus toward a more general purpose use that is different from what Schema.org originally intended.

SpecialAnnouncement Structured Data

The SpecialAnnouncement structured data was created by Schema.org in March 2020 as a way to communicate special announcements that are specifically related to Covid-19. In fact, the Schema.org documentation states that it currently remains as a work in progress that is specifically focused on the then current Covid-19 crisis.

This is how Schema.org defines this specific structured data:

“A SpecialAnnouncement combines a simple date-stamped textual information update with contextualized Web links and other structured data. It represents an information update made by a locally-oriented organization, for example schools, pharmacies, healthcare providers, community groups, police, local government.

The motivating scenario for SpecialAnnouncement is the Coronavirus pandemic, and the initial vocabulary is oriented to this urgent situation. Schema.org expect to improve the markup iteratively as it is deployed and as feedback emerges from use.”

Google’s Implementation Of SpecialAnnouncement

The SpecialAnnouncement structured data was introduced by Google in April 2020 as specifically for Covid-19 related situations. It was created as a way for businesses, government entities, schools and other organizations could communicate special announcements such as quarantines, hours of operations, closings, and restrictions that are related to the Covid-19.

SpecialAnnouncement was released as a Beta feature and it currently remains in Beta status, officially in a testing phase, meaning that Google’s implementation of it was subject to changes or complete removal which is what quietly happened in the first week of 2024.

How SpecialAnnouncement Structured Data Documentation Changed

There are multiple changes throughout the document that are too numerous to document. Google’s previous version of the Special Announcement structured data documentation as of December 2023 had 38 references to Covid-19 and a total of 221 removals of text.

After the change the new version of the SpecialAnnouncement documentation contains only 13 references to Covid-19.

Most of the changes are similar to the following examples.

The previous documentation for this section of the guidance:

“How to implement your COVID-19 announcements
There are two ways that you can implement your COVID-19 announcements:”

Was rewritten to this:

“How to implement special announcements
There are two ways that you can implement your special announcements:”

The changes to the above passage transforms the purpose of the SpecialAnnouncement structured data into one with a wider application than the formerly narrow one of Covid-19 but still remains focused on local medical-related events (more details on this below).

Here’s another example that has a similar broadening effect.

This previous passage:

“Information about public transport closures in the context of COVID-19, if applicable to the announcement.”

Was rewritten to remove references to Covid-19 like this:

“Information about public transport closures, if applicable to the announcement.”

The changes, while subtle, represent an evolution of the SpecialAnnouncement structured data specifications away from being specific to Covid-19.

Perhaps the most notable change in the documentation is at the very top of the page in the very first paragraph.

The previous version of the documentation at the top of the page featured this:

“COVID-19 announcement (SpecialAnnouncement) structured data (BETA)
Note: We’re currently developing support for COVID-19 announcements in Google Search, and you may see changes in requirements, guidelines, and how the feature appears in Google Search. Learn more about the feature’s availability.

Due to COVID-19, many organizations, such governments, health organizations, schools, and more, are publishing urgent announcements…”

The updated and considerably shortened passage of the guidance now looks like this:

“Special announcement (SpecialAnnouncement) structured data (BETA)
Many organizations, such governments, health organizations, schools, and more, may need to publish urgent announcements (such as COVID-19 announcements)…”

Impact Of Changes To SpecialAnnouncement Structured Data

The dozens of changes to the SpecialAnnouncement structured data has the effect of making it more flexible for use in situations beyond Covid-19.

But the changes as they currently stand do not widen the scope of the SpecialAnnouncement structured data beyond what they previously were. This is because the nine examples of how the structured data can be used remain exactly the same.

Nine Examples Of Special Announcements

1. Announcement of a shelter-in-place directive
2. Closure notice (for example, closing a school or public transportation)
3. Announcement of government benefits (for example, unemployment support, paid leave, or one-time payments)
4. Quarantine guidelines
5. Travel restrictions
6. Notification of a new drive-through testing center
7. Announcement of an event transitioning from offline to online, or cancellation
8. Announcement of revised hours and shopping restrictions
9. Disease spread statistics and maps

The guidance continues to encourage the use of this structured data for extraordinary situations related to widespread communicable diseases and other disruptive emergency situations.

Given that the examples remain the same it’s probably safe to say that it may be inappropriate to use this structured data for something isolated like the rescheduling of a theater performance due to unforeseen circumstances. But because this structured data is in Beta it may at some point evolve

Read the updated SpecialAnnouncement structured data guidance:

Special announcement (SpecialAnnouncement) structured data (BETA)

Featured Image by Shutterstock/Lauritta