Introduction to Semantic SEO

In 2012, Google announced its shift from strings to things. But what does this mean?

If you turn back the clock on SEO history, you would see there was a time when stuffing keywords was enough to get ranked well on Google.

At the time, Google’s algorithms associated how frequently a string – a sequence of characters – appeared on a page with relevance.

So, if you stuffed your target keyword into an article, the page would show up high on the rankings.

This practice of force-fitting keywords to artificially manipulate the algorithm killed user experience and diluted the quality of search results.

But Google put an end to all that when it rolled out the Hummingbird algorithm. The algorithm shifted its focus from strings to the things those strings represent.

In the post-Hummingbird era of SEO, the context around a string became a prominent ranking factor.

How Google Analyzes Context

Let’s play a game. I want you to guess the word that occurs in the blank. To make the task challenging, I’ve blocked all the contextual words with [X].

In [X], during its [X] [X], tragedy struck when the ________ hit an [X] and sank. [X] [X] [X] captured the emotion of the [X] quite accurately in his [X] winning [X] [X] [X] . The [X] won [X] [X] ; but surprisingly [X] [X] [X] [X] didn’t win an [X].

Now, try guessing the word in the blank with the contextual words visible.

In 1914, during its maiden voyage, tragedy struck when the ________ hit an iceberg and sank. Director James Cameroon captured the emotion of the event quite accurately in his Oscar winning film The Titanic. The film won 13 Oscars; but surprisingly lead-actor Leonardo Di Caprio didn’t win an Oscar.

With the help of the contextual words, I’m sure you had no trouble filling the blank.

(If it wasn’t obvious, the answer is Titanic…)

Google’s current approach to understand content works the same way you did to figure out what goes into the blank.

Thanks to contextual cues, Google’s algorithm does not rely on the string-based approach anymore. Here’s an example to demonstrate that.

Strings vs Things – An Example

Consider these two queries:

• what’s the life expectancy of a mini bulldog
• how long do miniature english bulldogs live

If you look at just the sequence of characters in the two queries, the two queries are very different.

With Levenshtein distance as the base for comparison, these two queries are only about 18.18% similar.

(Levenshtein distance is the number of edits required to change one string to another.)

However, if you ask a searcher, the two queries are nearly identical in meaning. In other words, the two queries are semantically identical.

And if you search for the two queries on Google and compare the articles that are ranking on the first page of the SERP, you’ll find 8 of the 10 articles ranking for both queries.

In other words, Google is now able to tell that the two queries are very similar semantically.

Since Google can find semantic relationships between things, optimizing content for a specific query or phrase is an inefficient strategy that leaves a lot of potential traffic on the table.

Therefore, as a content publisher, it’s time you move from conventional SEO to semantic SEO.

Conventional SEO vs Semantic SEO

Conventional SEO optimizes a stand-alone article to target traffic for a search query. Semantic SEO optimizes a cluster of interlinked articles to target traffic for many – if not all – queries related to a topic.

I’ve covered the differences between conventional SEO and semantic SEO in detail in the article below.

That article will also compare the pros and cons of semantic SEO and conventional SEO, enabling you to decide whether semantic SEO will work for you.

I recommend you read that article next.