A Checklist to help you
https://www.newzdash.com/elections2020. This is a live SERP tracker of all election results in Google.
I have worked in so many publications and I meet with so many SEOs and so many publishers and it has been the one topic that keeps coming over and over. Can you really optimize for the Google Discover
Google Discover is an AI-driven content recommendation tool introduced by Google in December 2016 which was initially called Google Feed and the name changed to Google Discover in 2017. Google Discover focuses on relevant and engaging content. The operating words being - relevant and engaging because we are going to keep coming to these two words over and over.
Google Discover Feed includes a lot of information. It includes
In a way, this is Google’s way of Facebook feed.
Google Discover is different from search as you already know. In Google search, You have a keyword and this keyword will determine what results you will get. Where as in Google Discover there is no query, it's queryless and it matches users interests.
Usually most of the Discover content is based on newly published content.It's recently published content, but it's not only new content, a lot of times there is older content as well. Discover doesn’t limit rankings towards published more most recently, which means like to stories or Google news. It's very recent within twenty four to seventy two hours. And this is completely automated based on AI machine learning and based on user interests and also evolving using user interests.
Google Discover is highly personalized. It takes in consideration your search history, web history, app history, browser history, what type of device you're using, what type of interests and topics you follow or implicitly interested in your location history.
Also, if location tracking is enabled, you will see recommendations based on that as well.
This is a screenshot. I live in Berkeley Heights, New Jersey, I never entered my address anywhere but the tracking is enabled and it understands that and I get a lot of stories from ‘TAPinto’ which is a local publisher. So Discover gives users more or less info on every story and the control to see more or less of that story.
Also, you can customize Google Discover by blocking any topic or interest that you don’t like. If you click the three dots next to each story, this card appears and it says hey, you can hide the story. You are no longer interested in the topic that they identified i.e Berkeley Heights or you're not interested in the source which is ‘TAPinto’
This is something from Google blog and it's very important to understand what are topics.
When we talk about interests or topics, this is a top layer that may include hundreds and thousands of entities and keywords and so on.
“We have taken our existing knowledge graph which understands connections between people, places, things and facts about them and added a new layer on top of this called topic layer, engineered to deeply understand the topic space, how interest can develop over time as familiarity and expertise grow. The topic layer is built by analyzing all the content that exists on the web for a given topic and develops hundreds and thousands of sub-topics.”
So for the subtopics we can identify the most relevant articles and videos, ones that have shown themselves to be ever green and continually useful. Underline Evergreen and Useful as well as Fresh Content.
Google is telling us 3 things here: Fresh Content, Evergreen and Useful.
We then look at patterns to understand how these subtopics relate to each other similar to the knowledge graph. So we can more intelligently surface the type of content you might want to explore next so by understanding the topic layer and sub topics.
If you are interested in one of these sub topics, then they start determining. Are you interested in other sub topics under the same topic layer? If you look at the top keywords, ranking every year in Google zeitgeist - These are the top keywords for the year 2020:
The following are some of the types of entities found in the Knowledge Graph:
With the BERT and all the other Google algorithm changes, you know that topics is a very major concept now in SEO and when we talk about topics or entities, you're talking about four or five things you're talking about:
Google reports 15% of all searches are new searches, so topics and entities are new every day there are a lot of new topics that come in. iPhone 13 was not an entity a year ago, but now it's an entity. And galaxy pencils and so on.
There are new topics and new entities that come in every day. These are some of the supported schema types that Google highlighted on their site so there are books, book series, education, events and so on. If you are optimizing for news, you need to keep entities in your head and it's important because this is the majority of people who search using entities and qualifiers.
For example: An entity would mean Donald Trump and a qualifier would mean ‘twitter’, ‘joe biden rally’ and so on. So there is always an entity and then a qualifier.
Entities must be present in headlines and must be present throughout the story.
Should we reach a certain word count right? It's not about word count, it's about how comprehensive your article is and a part of being comprehensive is talking about these related entities in the article. I always give this example of how to boil an egg. If you're talking about how to boil an egg most likely you're talking about easy recipes and you're talking about breakfast and egg shells and so on right. These are all connected together. So when you're writing an article in News about a specific entity, what are the related entities at that time?
Because in all likelihood, you are already seeing the numbers in your Google search console and it's amazing to see for example in India and in Europe, some publishers that get 90% of their organic clicks in Google search console through Discover. It's like a fire hose of traffic for some publishers.
We believe that countries with android heavy consumption or phones have a very high penetration of Google Discover Also, certain topics like sports, new publishers focused on sports have a very high Discover traffic. Reaching drastically from 15% all the way to 90%.
If you look at your Google Discover, you will see that the average CTR in Google Discover is almost seven folds as your other keywords.
This is one of the very top publishers out there for a certain period of time and you see that they're 8M clicks that they got through Discover with an avg. CTR of 8.6%. Even though the impressions are lower, but it's a very high CTR
Theory
According to my theory, Google Discover also shows up in the SERPs. Remember we discussed entities and interests and so on.
If you have ever seen “Interesting Finds” a lot of times, 70% of the results have Interesting Finds. My theory is Interesting Finds are a part of Google Discover. I still don't understand how it is triggered, why it appears in certain ways, how it appears.
Also there were some users on twitter reporting the Interesting Finds start to show up on some knowledge panels, especially around businesses or around certain entities which also gives a signal like what i mentioned before that who will Discover and entities and topics and so on are extremely connected.
It only shows up in mobile and when clicked on “Interesting Finds” you will go to a page that looks extremely similar in design and layout as Google Discover.
On that page you will see sub topics on the top right. When searching for “Google Discover” you see the “Interesting Finds” on the SERP and when you click on one of them, it takes you to a page very similar to Google Discover.
I don't have any confirmation from anyone at Google but I believe Google Discover is making its way in the SERPs somehow based on all the personalization we mentioned and in a recent study, they found 70% of all the queries when analyzed had “Interesting Finds” block inserts.
The August trend gives a very clear sign that Google Discovered traffic is not reliable or it's not sustainable.
This can change for more for so many reasons. It's a supplemental traffic and Google keeps stressing about it. As Google Discover gets more and more popular, more content gets added to Google Discover it gets cluttered, you get less chances of getting into it.
Remember it's a social feed. If you work on social media, look back to when facebook first came out, it started sending huge traffic to publishers to the point that facebook traffic was almost the same or more than Google traffic for publishers.
Then in 2018 they changed their algorithm, they wanted to focus on meaningful interaction and engagement. So they turned the knob down and overnight publishers lost 50-60% of their traffic.
Take away - The catch with social feeds is that since they have no queries, the vendor keeps focusing on certain metrics that are important for them. Engagement, CTR, meaningful interactions and so on.
Google mentioned engagement a lot of times so that is why Discover traffic is not a reliable traffic but still we can talk about some ways to optimize some ways to uncover certain secrets of Google Discover.
So you have to do it. I created this and there are two ways to do this. One is you take all your articles that are ranking in Google Discover every day or like once a month and Google search console will give you the top thousand.
Then you take each article, you go to Google NLP (Natural Language Processing). Register for an API key and then have some developer or some tool do this for you and then you can collect all the entities about each of these articles.
Collect these entities in an excel or Google sheets and cluster the entities together based on the clicks you get from the articles and then you will get something very similar to this:
.png?width=50&height=50&name=segment%20(1).png)
.png?width=32&height=32&name=testimonial%20(2).png){kind=small}
