Is semantic analysis the ultimate weapon to boost your SEO?

The concepts of semantic analysis and, more broadly, of “SEO semantics” are often discussed in the SEO sphere. Today, it’s a top priority for defining effective online visibility strategies. In this comprehensive article, we’ll cover several points to fully grasp the various challenges tied to semantic search. First, we’ll look at how SEO semantics improve your organic visibility on Google. We’ll then take the time to define the concept of SEO semantics, and finally we’ll offer you a range of techniques to create optimized content.

Why semantics will supercharge your SEO

SEO semantics refers to an approach to organic search that aims to write high-value content for your audience: detailed, precise, and providing a relevant, complete answer to users’ searches.

Here are the various benefits:

• Make it easier for Google to understand a given topic: you grow your site’s traffic through rankings earned on long-tail keywords. The combined traffic generated by all of these queries can add up to substantial traffic.
• Multiply the entry points to your site – long-tail queries rank your pages on Google more easily. What’s more, the observed click-through rate is often better on this type of query: the results pages are generally less crowded with Adwords and/or Shopping ads.
• Boost your overall SEO traffic – ranking on a large number of keywords lets you multiply the “entry points” to your pages, improving your traffic in the process.
• Increase your domain’s popularity – Content seen as interesting may be shared on social media and picked up by blogs. The external links generated this way help optimize your link profile.
• Provide precise answers to users – If Google considers that your page answers a specific query, it may display the result in position 0.

Designed to improve the SEO performance of content, Semji enables a fine-grained understanding of user expectations. The platform’s artificial intelligence analyzes search results in real time to deliver the most relevant SEO recommendations.

The rise of voice search

In 2020, 30% of searches will be voice searches (Gartner). The rise of voice interfaces (smartphones, connected speakers, etc.) is driving conversational search. These new search habits give rise to new forms of queries, longer and more precise:

• “what’s the weather like in Lyon?”
• “find me an easy cake recipe”
• “how do I change a tire?”

Google can no longer be seen as a mere search engine, but as an answer engine. It encourages user retention on its pages by offering site excerpts in position 0 (“featured snippets”):

How do you answer the increasingly precise questions driven by voice search?

Working on semantics will let you create rich content that’s relevant to the user. Two actions to prioritize:

• Identify your main keyword (high search volume) and define its lexical field. Tools like Semji, alyze.info, sequantum.com or 1.fr can help you with this task.

• Work on your site’s internal linking and structure. Rich content will have far less impact if it sits on an orphan page. ( https://www.definitions-marketing.com/definition/page-orpheline/).

Work in depth on your pages’ semantic environment, the various relationships that exist between them (related topics, links), and your site’s architecture. The goal is clear: make your site as understandable as possible for Google in order to maximize your presence in the results pages.

What is semantic SEO?

With its algorithms dedicated to monitoring the quality of editorial content on the web (Hummingbird, RankBrain, etc.), Google is getting smarter and smarter. It works harder and harder to understand the user’s search intent in order to offer them the best content.

A definition of semantic search

To fully grasp the idea of “semantic SEO,” let’s rely on this apt definition given by Tamas Doszkocs of weblib.com ( http://weblib.com):

“Semantic search refers to a search, question, or action that will produce results meaningful to the user, even when the retrieved items contain none of the terms from the initial query, or when the search performed involves none of the words used in the initial query.”

In plain terms, Google will “understand” the user’s search intent based not on the meaning of each individual keyword, but by analyzing the meaning generated by the combination of search terms in a given context. Some Google results won’t necessarily contain the exact terms used to generate the search, but rather content that will meet your expectations.

These new techniques directly influence the work of SEO specialists and the way editorial content is optimized.

Let’s now take a look at Google’s Hummingbird algorithm, which is meant to improve the understanding of conversational queries.

Google Hummingbird: the goal is relevance!

In September 2013, Google announced it had rolled out Hummingbird, an algorithm whose mission is to improve the relevance of results by taking users’ search intent into account. Unlike Google Panda or Penguin, Hummingbird is not an update aimed at penalizing low-quality sites.

This algorithm is part of an effort to deliver relevance for the user: to offer them useful content that matches their initial search intent.

The first question to ask yourself as a content producer: is the query I’m working on navigational, informational, or transactional? To find out, look at the results page for your keyword. Are the results dominated by product pages? Buying guides? Analyzing the results page helps you understand what Google understands and favors.

And it’s in order to serve users even better that results pages can take various forms:

• When you search for information about a public figure: their profile is displayed in a dedicated knowledge graph showing the celebrity’s date of birth and, if they’re an actor, the films they’ve appeared in. In that case, why click on a classic organic link?
• When you type a city name + weather: the weather snippet is displayed with the day’s temperatures and those of the coming days. So why take your search any further?
• When you search for information about a company: here too a snippet can be displayed and give you enough information to save you from clicking directly on the organic result.

For certain specific types of queries (weather, local time, math formulas, etc.), Google turns into an answer engine and no longer displays a single blue link. Even though Danny Sullivan indicated it was only a test, this clearly shows Google’s intent to become an answer engine.

A few “Hummingbird friendly” actions

Hummingbird therefore fits perfectly into Google’s vision of shaping a higher-quality web. Google understood the stakes of semantic SEO as early as 2013. It’s therefore important to understand this logic in order to produce quality content.

Here are some actions to take to “please” the algorithm.

• Whenever relevant, integrate microdata within your content: breadcrumbs, user reviews, product sheets, etc.
• Rather than focusing on a single keyword, broaden the lexical field of your articles: synonyms, related topics, etc.
• Work on long-tail queries of more than 2 words
• Use the semantic tags offered by HTML 5 to build your sites: <header>, <nav>, <article>, <footer>.
• Structure your content using the HTML tags <h1>, <h2>, <h3>, etc.
• Where relevant, adapt your content to conversational queries: short sentences, bulleted lists, HTML tables, etc.
• Don’t forget universal search: add optimized images (lightweight, with alt text filled in) to your content and, if you can, videos from your YouTube channel.

Named entities, to better understand content

With voice search, queries have become more complex: whole sentences, multiple word forms (plural, singular, semantically related words, etc.). Google is making progress in the linguistic analysis of user queries, notably by developing indexing techniques based on phrases: it no longer analyzes keywords one by one, but the relationships that may exist between each word.

Since that isn’t enough to perfectly understand content, Google also relies on named entities.

What is a named entity?

The term first appeared in 1995, during the MUC (Message Understanding Conference) campaigns ( https://cs.nyu.edu/cs/faculty/grishman/muc6.html), funded by DARPA in the United States in the ’90s (source – Wikipedia) https://fr.wikipedia.org/wiki/Entit%C3%A9_nomm%C3%A9e

The expression “named entity” refers to a textual object with various properties. It often refers to a real-world entity: a person, a place, a date, an event, etc.

In concrete terms, an entity contains all the terms that relate to the same idea. Let’s take the city of Paris as our named entity. Several expressions can define this place:

• capital of France
• administrative center of Île-de-France
• city of Paris
• major French city, etc.

All these terms are grouped under the single entity Paris.

How does Google use named entities?

In 2012, Google bought the Metaweb database (the Freebase project (https://developers.google.com/freebase/), which had nearly 12 million named entities. Why? Quite simply to maintain a technological lead over other search engines and build user loyalty to the “Google” brand. The ultimate goal is to keep the user on its results pages as long as possible (knowledge graph, position 0, “Popular destinations” snippet, etc.) so that they might potentially click on paid ads.

Today Google is no longer a simple search engine based on basic indexing techniques. The company puts an enormous amount of energy into understanding and also learning from the millions of queries generated on its engine every day.

RankBrain and machine learning: artificial intelligence joins the dance!

We now know that Google uses the RankBrain algorithm, an intelligent system that helps deliver relevant results to users.

RankBrain sees 15% new queries every day, i.e. 500 million queries searched for the first time each day!

According to Greg Corrado (Senior Research Scientist at Google): “RankBrain is a way for Google to better understand ambiguous queries that are open to multiple interpretations. It uses artificial intelligence to accurately guess what topic the query relates to, as a human would.”

https://www.seroundtable.com/google-rankbrain-21095.html

This algorithm is based on machine learning (a subfield of artificial intelligence research): the goal, then, is to guess what the user is looking for and answer it. That’s how Google manages to generate completely different results for very similar queries.

Let’s take a telling example with the query “le bachelor.” The search engine displays a knowledge graph presenting the TV show, along with associated organic links:

If we remove the pronoun “le,” all the results relating to the show disappear, and the results page is then completely different:

Position 0 and PAA (People Also Ask) results appear. The results shown therefore focus on detailing this academic degree. The search phrase is almost identical. Yet Google manages to interpret the nuance introduced by the single pronoun “le” and offers the user results that match their search intent.

How to use semantic SEO for your website

To rank on targeted keywords, it’s no longer enough to include those keywords within your pages. You need to go further by broadening the lexical field around the main keyword.

Let’s say you have an online store offering home decor and outdoor furniture: garden sheds, pools, spas, barbecues, etc. You’ll need to help Google understand what your site is about and the meaning of the internal anchors placed on your pages. To do this, rather than staying focused on the phrase “garden shed,” you can work on words belonging to the lexical field: “garden gazebo,” “garage,” “garden greenhouse,” “carport with shelter,” etc. The following steps explain how to leverage semantics in SEO.

Identify your semantic corpus

Keywords are still very useful in SEO and are the foundation of any SEO strategy. Ranking trends act as a barometer for analyzing the impact of SEO actions.

Here are a few tips for finding semantically relevant keywords:

• Find keywords that users are searching for: tools like Adwords, Semrush, Ahrefs and Yooda provide information on monthly search volumes.
• Think “conversational search” and find questions your target audience might ask: for this, take a look at Ubersuggest and AnswerThePublic to round out your initial list
• Take a look at your competitors: which keywords are they ranking for?
• Have pages where you can work on the various chosen queries: if needed, you’ll have to consider adding new pages
• Leverage your internal search tool: what better way to learn the exact searches your visitors are making?
• Work on longer-tail keywords: staying focused on highly competitive keywords can cause you to miss out on a significant audience
• Understand your target’s search intent: commercial? informational? transactional?

Build a semantic architecture

Depending on the client challenge at hand, we recommend that our clients reorganize their site structure into thematic silos.

What are the expected SEO impacts?

• Satisfy the visitor’s search intent: the searched-for topic is presented in its entirety in one place on the site, covered across several pages of varying levels of detail.
• Satisfy search engines’ indexing bots through effective internal linking: deep pages support the main page by linking to it and vice versa. The main page’s ability to rank on more competitive keywords is strengthened. What’s more, crawling deep content is made easier: this way, Google’s crawl focuses on pages that are useful for SEO and therefore strategic for your business.
• Work the long tail: deep pages covering more specialized topics will rank on keywords with lower search volume that are less competitive but easier to rank for. You multiply the entry points to the site and, potentially, organic traffic.

A sealed silo for greater SEO effectiveness

Simply organizing your pages well isn’t enough to maximize SEO gains. The defining feature of a silo is that it must be sealed. Ideally, links should be made between pages of the same level. Subpages should send link juice to your parent page.

The silo’s semantic consistency is thereby strengthened: the topics covered are better understood by search engines. Be sure to apply this linking technique across all of your content.

Finally, don’t forget to add a breadcrumb trail within your pages. There’s a twofold benefit:

• Improve internal link structure,
• Help a user coming from Google find their way around your site more easily.

Organizing into silos thus lets you work on the on-site contextualization of your content.

You now have a solid architecture… but the essential piece is missing: content that’s interesting and sought after by users!

Write optimized content

Multiply the semantic richness of your content!

The old technique of repeating the same keyword several times within a page is over! Today it’s important to work on your content in a way that makes it richer in editorial terms. Starting from the main keyword you’re working on, you should enrich the topic by drawing on its lexical field.

Here we’re talking about “co-occurrence,” that is, terms that are frequently associated. This is how keywords considered secondary will help you rank better. It will be harder to rank for “baby product” than for “cheap organic baby product.” The latter query refines the initial search while allowing you to explore the lexical field more deeply.

By working with this approach, you’ll grow your long tail and increase the number of potential “entry points.” A tool like “Seoquantum” will let you easily find words to enrich your content semantically. You can also use a synonym dictionary to complete your lexical field. To help you in this quest, other tools exist such as 1.fr, Alyze or Semji.

How to work on semantic SEO?

Semji saves you time

Discover the Semji platform in 2 minutes! (French) - YouTube

Tap to unmute

Discover the Semji platform in 2 minutes! (French) Semji

Semji 926 subscribers

Watch on

A proper semantic analysis is essential and takes time. Time that marketing and SEO teams don’t always have. Semji was designed precisely to help teams:

• Prioritize the pages to optimize based on their potential for traffic or conversion gains.
• Understand users’ search intent at a glance
• Analyze competitors on a given keyword (and do better!)
• Guide writers in designing and writing relevant, semantically optimized content.

Google works to offer its users the best possible experience on its search engine. Google’s algorithm is increasingly complex and sophisticated, and it requires constant adaptation on the part of SEOs and marketers. On top of this comes the evolution of practices with, in particular, voice search, which influences the way we consume and therefore produce content.

To learn how to produce 100% optimized content tailored to your audience and boost your SEO performance, don’t wait any longer and request a demo of Semji!