Search engines are ever-changing, increasingly shifting their focus on user intent and behaviour. As a result, keyword optimisation is also changing, and calling for a more ‘semantic’ SEO. As people are online more often than ever before, search engines are doing their best to adapt to the new ways in which users are searching on the web, and improving their algorithms accordingly. Keywords alone are no longer sure to match up the query with the user’s intent, and a different approach is required.
When selecting its results, Google now takes into consideration much more than the single keywords: it takes into account the semantic fields of the words, i.e. synonyms, connections, location, etc. The aim is to create a more predictive and personalised search experience; the rise in the number of answers appearing in answer boxes is just one manifestation of this.
Semantic search: The details
Semantic search works through semantic fields, which include synonyms, co-occurrence, related topics, and lateral ontologies. These allow search engines to understand the meaning of a certain word or phrase.
In the everyday world, humans use context to identify significance and narrow down the different possibilities of meaning that a word has. Search engines, however, do not have this luxury, and they give us a list of results which encompass all the possibilities. They then remember our choice to deliver more accurate results in future searches, taking into account things like user intent, country-specific preferences, and behaviour. This is why search engines give different search results for the same word in different countries.
To better meet the needs of users, keyword research is thus moving towards topic research. This gives search engines many more elements with which to understand the true intent of a query. In topic research, one doesn’t think of individual keywords anymore, but of correlated entities.
If you have a website, you need to keep in mind the degree of semantic proximity of the page content, as it considerably influences the domain name, sub-domain and sub-folder. If two topics are very close, it is useful to combine them in a single domain name to enhance the topic authority; however, if they are very far apart, their presence in a single domain could be detrimental to both. A semantic approach means rethinking website structures to have fewer but better-structured pages.
Recent changes in Keyword Planner
Currently, when searching the volume of a keyword, Keyword Planner alters the queries, combining different keywords instead of showing individual volume estimates for each keyword. Essentially, it groups keywords into ‘buckets’, thus reducing the keyword pool. The main reason why this is happening has to do with user intent. Search engines store information about user behaviour and use it when it comes to displaying search results. A downside of this change is that estimates on traffic and volume are becoming less accurate, showing averages rather than numbers.
It is worth pointing out, however, that the changes which are happening are not consistent. For this reason, it is important to involve language-native professionals who will be able to give you insights on the language, culture, and user intent.
It’s also important to remember that these changes are happening slowly, and there is no agreed best practice at present; carefulness is therefore required when venturing into semantic search.
Serena Mancini
Latest posts by Serena Mancini (see all)
- Three Ways To Combine PPC And SEO For A Successful Marketing Strategy - October 16, 2017
- Keyword Research Is Heading Towards Semantic SEO - April 4, 2017
- What To Consider When Setting Up An E-Commerce Presence In Italy – Part 2 - November 10, 2014
Hi, thanks a lot for the nice article
Do you always use Keyword Planner for it? Maybe you have some other suggestions?
I use SerpStat now, but is it good for semantic SEO. What do you think?
Hi Jordan, thank you for your comment. I am glad you liked the article.
I think it is always a good idea to not just rely on one tool and I personally like to use several, depending what exactly I want to measure.
If you want to incorporate semantics into your keyword research, take into consideration linguistic tools that will help you find synonyms and variations, but also use the SERP to test and verify what the search engine offers as a result for your chosen keywords. I sometimes find that even more helpful than any tool!