Thread by @GiordMarco96

Some daily considerations about Semantic #SEO, #Python, and strategy for your next project.

This time I am going to include new stuff and write not-so-obvious considerations.

Your updated thread for improving your knowledge about data 🧵

Semantic SEO is different from the simple topic cluster approach. Here you are interested in connections about entities and ontologies.

In other words, you want to have a clear idea of how to create links between your pages and proper navigation based on evidence. >>>

>>> The traditional cluster approach has no mention of all these elements. Creating content is not going to make the difference if you cannot build a network.

This is true in highly competitive environments where you need some "authority" to be deemed worthy.

You can see it starts working when you notice a sharp increase in Impressions. This should lead to an increase in clicks later or at the same time, according to my experience.

This amount of time is not fixed and can change depending on several factors.

Semantic SEO is not even necessary for some niches where competition is pretty low but it's so convenient that you should either way.

It's better for the user as well, because you're planning a structure that makes sense and your content is connected according to some logic.

You don't need huge topical maps. A shallow one is more than fine. Recall that topical authority doesn't mean you have to be perfect!

As an experiment, I tried competing with Wikipedia on some long-content pages and I noticed that it is quite possible to be 2nd or even first with the proper topical authority.

However, competing for such short queries is usually not worth it and not even lucrative.

If you have a blog or an online magazine you can create specific pages for entities with attributes.

Short example, large videogame websites have pages for each videogame, listing all their attributes. They don't know this is a very good approach! >>>

>>> Having such pages is super good and a godsend for internal linking. Add the proper schema markup and supporting content and you will get good results.

There are alternatives, of course, you don't need to copy what other people do all the time.

You can cover all the contexts around a certain topic.

If you sell fridges, you can create articles that contain comparisons, information about different details, buyer guides, and good category pages.

You're covering the full customer journey here!

Competing in the USA/UK is super hard compared to non-English-speaking countries most of the time.

However, much larger markets have a lot of untapped opportunities!

You can check foreign websites to understand what they do. Underrated methodology but extremely deadly if applied correctly.

Some search intents are common across the globe and you can get more data by looking at some other countries.

Google tries to understand websites with the help of the Knowledge Graph, an efficient way to store and retrieve data.

Your website should be able to contribute to a given topic and provide some degree of new information, if possible.

You can clearly see some cases where competitors rank higher just for backlinks.

Don't despair, it takes time to see the results of Semantic SEO.

And this is why it's better for long-term advantage rather than immediate momentum!

There are instances when it's faster, it depends on your content velocity as well.

Your publishing rate may make the difference in some niches.

To play it safe, recall that it may take quite some time for Google to process your information, on average.

There is no clear answer without looking at competitors in given domain knowledge.

Try to understand who are your competitors by checking who is ranking most of the times for common queries.

Then, check if you have the same model and what you can do.

In my case I adopt a different strategy with B2B, as I am still learning and it's challenging.

In the case of online magazines the methodology is totally different.

Some criteria can help you to gain an edge over competitors >>>
to to to

>>> Instead of going after news content, you can build topical authority around evergreen content.

The supporting content that is not suitable for traditional search can be reinvented for Discover!

https://twitter.com/GiordMarco96/status/1501271544118460416

Repurposing content is another important element of content management.

Most popular articles can become YT videos and vice versa. Learn to expand your content and how to use Semantic SEO to amplify it.

If you have no clue on how to do semantic SEO then use your brain. How would I describe a topic in the best way possible?

What are its properties and related topics? Can you imagine talking about Marvel movies and not mentioning The Avengers?

This is a short example.

If that weren't enough, consider this before you launch a website.

What is the scope of this project? Will I cover just this one niche or I will take things further?

I prefer domains with a unique and original name rather EMDs. Even though there are cases where EMDs rule.

Being honest about the scope of a project is a good way to solve a lot of problems.

In my experience, a lot of site owners are not dealing with technical issues rather content and business problems.

Don't compete in hard niches without the budget!

Ideally, you should aim at an easier set of keywords and then move to the next layer.

I think this can be bypassed in some scenarios, yet I recommend this mindset as a rule of thumb.

Play it safe, no reason to risk.

Understanding what to do is extremely important and that's why I am a big fan of scripting and analysis.

You can look at things differently and just press one button. That's why I like coding so much.

This is another case where you can scrape Wikipedia to find entities or use Open-AI for outline generation.

There are a lot of ideas and I will be sure to cover them on my Twitter account with these threads. It will take some time though
,,,,

Python can help you, it's not only about automation.

Data analysis is even better imho. NLP is another application to use NER (extracting entities) and POS Tagging.

Plenty of options here, don't limit yourself to automation!

https://twitter.com/GiordMarco96/status/1499842120068177920

With the increasing quantity of data and the relative easiness of high-level programming languages, it could be beneficial to learn some data skills.

This is super true for younger people like me who have a lot of time to choose the path they like the most.

Google will become gradually more focused on entity understanding, as shown several times.

MUM is the next step into a unified approach to web information. Curate your website as it was a library storing the most important information.

If your goal is to build a solid brand that will last for years, you should have a plan and ethics.

You cannot hope to achieve success without

If you liked this thread, consider linking or retweeting it. Do you have any suggestions for future topics? Let me know!