“For each anchor element that has a title attribute, verify that the title attribute along with the link text describes the purpose of the link.”

fun fact:

The title attribute is actually a global attribute, meaning it is common to all elements.

That means you can put a title element on a paragraph

italics and even on a heading element

place.

If you use a title attribute for an element, a tool tip is displayed.

So if you have a heading element

,

etc., add a title attribute, when a reader mouses over the heading, a tooltip containing the words in the title attribute will appear from the words in the heading element.

2. Why extra-long anchor text is bad

Google’s new guidance on links states that long anchor text is considered bad practice and recommends being concise (to the point).

This shouldn’t come as a surprise since, according to the W3c, the official purpose of the anchor text is to describe what the link is about.

“The goal of this technique is to describe the purpose of a link by providing descriptive text as the content of the a element.

The description allows a user to distinguish that link from other links on the web page and helps the user determine whether to follow the link.

The destination URI is generally not sufficiently descriptive.”

This underscores the importance of knowing valid and proper usage of HTML.

If this is valid HTML, if the various elements and attributes are used as they are intended to be used, then chances are Google will respond positively.

This shows how important it is to know how to use HTML properly.

If in doubt, check the W3C or Mozilla HTML developer sites, which I think are easier to use.

I prefer the Mozilla developer pages because they are more organized than the official W3C resource.

3. Context and natural language important for link anchor text

Using natural language is important in my opinion to ensure content is properly search-optimized.

Every AI and machine learning algorithm coming out of Google today focuses on natural language understanding.

Google’s algorithms do not award “points” based on the keywords in the content.

So if Google’s algorithms interpret text in a certain way (looking at entities, verbs, context, etc.), then it makes sense to write clear and easy-to-understand content.

According to Google’s new guidance on links. Thinking about the context and using natural language is a best practice for link anchor text.

The new guide recommends:

“Write as naturally as possible and resist the urge to stuff any keyword related to the page you are linking to (remember keyword stuffing is a violation of our spam policy).

Ask yourself, does the reader need these keywords to understand the next page?

If it feels like you’re forcing keywords into the anchor text, then it’s probably too much.”

It used to make sense to stuff keywords into the anchor text.

Because Google uses technologies like BERT to understand what sentences and phrases mean, it makes sense to write natural anchor text that Google can understand.

Google uses more than BERT to understand search queries and webpages. I’m just using BERT as an example of why well-written natural language is important.

BERT’s official 2019 consumer facing announcement stated the importance of this context is for understanding natural language:

“Especially for longer, more conversational searches or searches where prepositions like “for” and “to” are important for meaning, the search can understand the context of the words in your search query.

You can search in a way that feels natural to you.”

BERT’s 2018 Official Scientific Announcement talks about the importance of “context” in understanding the meaning of content.

That’s what it says:

“…Pre-trained representations can be either context-free or contextual, and contextual representations can also be unidirectional or bidirectional.

Context-free models like word2vec or GloVe generate a single-word embedding representation for each word in the vocabulary.

For example, the word “bank” would have the same context-free representation in “bank account” and “bank of the river.”

Instead, context models create a representation of each word based on the other words in the sentence.

For example, in the sentence “I accessed the bank account,” a one-way contextual model would represent “bank” based on “I accessed this,” but not “account.”

However, BERT stands for “bank” using both its previous and next context—“I accessed the…account”—starting at the bottom of a deep neural network, making it deeply bidirectional.”

Of course, context and natural language are important to Google’s algorithm. With what I just wrote about just one of Google’s algorithms, BERT, Google’s advice on writing anchor text takes on an extra layer of meaning:

“Write as naturally as possible…”

4. Chain links do not

Chaining links means you add links close together, so each linked word doesn’t adequately communicate what the linked page is about.

Also, the text surrounding a link and providing context for it is lost when you concatenate links.

Google’s new guide explains:

“Don’t chain links side by side; It’s harder for your readers to differentiate between links, and you lose the surrounding text for each link.”

This advice goes back to understanding what the correct use of HTML elements and titles is to write valid HTML that Google understands.

Again, I highly recommend reading the Mozilla developer pages about HTML.

Search optimized links

There’s a lot of interesting insight in Google’s newly expanded guidelines on best practices for links.

It is absolutely worth taking the time to read.

Check out Google’s advanced guide:

Link best practices for Google

Featured image from Shutterstock/Asier Romero