Coming strictly from Google’s mouth, John Mueller,
“you might create and optimize content to fulfill specific search needs for the search engine traffic, but there is no way to create content that explicitly targets Discovers interest matching”
This is true but there are certain best practices you can do to increase your chances of getting more of the Discover trend. Google Discover is highly personalized, which means that no two users will get the same result. But instead of focusing on individual users interest matching, you start building a persona for your Google Discover audience.
Here we use machine learning, we take all Google Discover URLs in Google Search Console and then we get all the entities for each of these articles and use machine learning to cluster these entities together and then we come up with this.
So it seemed that the most of the traffic in July 2020 was coming through face masks, recipes, zodiac, trump, taylor swift and so on.
If you understand the entities from the articles that drive most of the traffic to your site from Google Discover you really understand your audience from Google Discover.
You need to understand your Google Discover audience and their growth interests. This is a key to optimize for Google Discover.
Having content that’s timely for current interests, tells a story well, or provides unique insights.
Having content that is timely is a key. Content that tells the story well and it provides a unique insight is what matters the most. Operating word here is ‘unique insights’. This is mentioned in Google top stories, in Google News and Google Discover. which means that Google appreciates diversity in coverage.
So if everyone is covering a certain angle about a certain incident, that's happening and
if you can present a different angle or different insights to the same incident, Google appreciates that and wants to present diversity to their users.
Don't just cover, what millions of other publishers discovered. Can you cover it from a different angle, a different opinion?
While the majority of content in Google Discover is timely (Newsy/Trending), we still see evergreen. The majority of the content, maybe two thirds or more is timely content between one and three days old. But we still see content that is two months old, three months old and when you think about evergreen content. For example travel, new hobbies, shopping etc
Make sure that your evergreen content is refreshed often and relevant. So when you have a guide and this is the same strategy you do for Google SERPs. Track your evergreen content and see when it drops in traffic.
On Titles -
Avoid tactics to artificially inflate engagement by using misleading or exaggerated details in preview content (titles, snippets, images) to increase the appeal, or by withholding crucial information required to understand what that content is about. Avoid tactics that manipulate appeal by catering to morbid curiosity, titillation or outrage.
Remember images and visuals are very important.
AMP is not a prerequisite for Discover but I analyzed ten thousand Discover results from the Google Discover feeds. I found that 60% or more are AMP. Again - This is not scientific. This is me just going over and over and over many feeds and I found at least two thirds If not more, of all the feed content that is recommended by Google Discover is AMP.
So I would say AMP increases your chances of appearing in Google Discover but it's not approved and since Google Discover only appears in mobile devices so you should absolutely be mobile friendly.
Another thing also to look at is, if you go to Google search console, select Discover performance and then select Discover appearances and look at these three numbers.
You have your ‘Total Clicks’ for Discover, you have your ‘AMP articles’ and you have your ‘Videos’ and you see your non-amp articles are a very small fraction.
89% of all the clicks that came to this site came through AMP articles right so AMP is very important. It's not a prerequisite but absolutely helps with Google Discover.
In July 2020, Google added this paragraph to Google Discover:
“our automated systems surface content in Discover from sites that have many individual pages that demonstrate expertise, authoritativeness and trustworthiness (E-A-T). Those looking to improve E-A-T can consider some of the same questions we encourage site owners to consider for search. While Search and Discover are different, the overall principles for E-A-T as it applies to content within them are similar”
So it's very much the same thing, especially if you're working around your money your life (YMYL) kind of content, medical content, health content or financial content, would recommend you follow Lily Ray she is an expert on E-A-T.
She wrote an article on What Does It Mean to Have Good E-A-T? and it gives you some information about E-A-T.
Let’s discuss the 7 things that you should absolutely focus on
Mention every kind of information on the “About me” page that can help raise the E-A-T score.
Things such as:
Also use a person schema on “About the author”. Refer this example of a person schema:
<script type="application/ld+json">
{
"@context": "http://schema.org",
"@type": "Person",
"email": "email@site.com",
"image": "https://author-image.jpg",
"jobTitle": "Job Title",
"name": "First Last",
"birthDate": "YYYY.MM.DD",
"url": "https://www.site.com/author-profile-page",
"sameAs": [
"https://www.facebook.com/author",
"https://www.linkedin.com/in/author/",
"https://twitter.com/Author/"
]
}
</script>
The next thing may not be mentioned anywhere else but there is a buzz factor in Google Discover. Google keeps mentioning ‘engaging’ over and over again.
The articles that Google Discover and fifteen other metrics, i did let study in january 2020 and i did it again in september 2020 and the one striking thing that i noticed is twitter interactions of these articles, the correlation is 0.95.
In general, correlation value of anything above 0.8 or 0.9 is a very high indication that there is a very strong correlation here because the maximum correlation value you can get is 1. When thousands of pages analyzed across a long period of time came back with the results that twitter interactions have the highest correlation to Google Discover traffic.
Not saying twitter and Discover but Google is looking at some kind of metric that provides engagement or buzz around content that they display in Discover,
It's not about twitter. It's about something. Some source in your country, in your place, that is influencing the buzz so start looking in your country. What are the most popular social platforms. start doing some correlations and see If there are specific platforms that are highly correlated.
Google Discover is more or less a social feed:
If you are a business, an entity, a brand, make sure you get your knowledge panel. This is information from Google. knowledge panels or entities have a follow button. Users can follow and this is an explicit signal to Google that I care about this entity. I care about this business and then they start getting more information from them
Allows users to follow Entities from Knowledge Panel
If you go to your phone in Google Discover settings, you will see Google’s telling you here is the information we collected about you.
So you cannot get access to the user information right. But if you do what i told you in the beginning, collect the Google Discover articles, run NLP on top of these articles, Google has a free NLP tool up to five thousand calls a month for free. Collect these entities, cluster them and understand your goal.
I took this article that showed up in my Google Discover Feed. I clicked on the three dots and it says,
All I did is I went to the article, I got the article content. I went to Google NLP. I put it there and you will see it showing that New Jersey is an entity. I live in New Jersey. Anthony Fauci is the most important entity. It's a person. Covid It's very strong entity.
So you can call it all that information, cluster them and understand the people that the interest graph of the audience that come from Google Discover to your site and keep doing this.
You can do this If you have tags on your site, this is i did this in like twenty minutes. collected all the tags from all the articles that drove traffic, did some excel magic very quickly and very easily.
I can tell that the top tags on our site.
This is how I identified: I don't like to use editorial tags, I'd rather use NLPand entities and so on. But these are when we write about news or politics or trump or coronavirus were really good.
High traffic from Google Discover right verses like we keep writing about hollywood. You can do the same analysis by section and so on. Now finally most of the traffic coming from Google Discover will last for like a day or so. There are some instances where the traffic lasts for a month or sometimes more But this is very rare. Most of the traffic will last for one day's spike.
You can track Google Discover by looking at your Google APIs, John Mueller said, not all Google Discover is identified in Google Analytics. That was in 2019. Not sure if it's still valid or not.
Most of the hacks i saw for tracking Google Discover in Google analytics are not perfect but this is a good one. Remember Google search console reports on clicks. Google Analytics reports in sessions so two different metrics.
This is a great article and slides. I have the slides later that tells you there are identifiers for every segment in Google Discover that tells you, what is this part in Google Discover
This is a john saying. We will not be expanding the Google Discover and Google search console data further than where we have right now and saying that the Google Analytics doesn't have everything
So utilize all the different aspects or different channels to reformat your content in different ways. Podcasts just started appearing a couple of days ago, Ads will be coming to Discover for sure. Evergreen content is being given more and more in Google Discover. Some users even said that they saw Reddit comments in Google Discover
Google is looking at certain platforms. I have seen or has been reported on twitter, there are short videos coming out in Google Discover more or less like tik tok and many more.
So this are some 6 articles that I encourage you to read with way more details:
.png?width=50&height=50&name=segment%20(1).png)
.png?width=32&height=32&name=testimonial%20(2).png){kind=small}
