Google Discover SEO: every Google best practice, and the ones I add from newsroom data

What Google Discover is, and why it decides news traffic

Google describes it like this: “Discover is a part of Google Search that shows people content related to their interests, based on their Web and App Activity.” (source) It is the feed under the search box on phones. Nobody types a query.

For news sites it is the biggest Google channel. In six newsroom accounts I worked with from 2023 to 2026, Discover brought about 60 to 80 percent of all clicks from Google. Across three news sites over the year to October 2026 it was about 74 percent. All AI sources together sent 0.02 to 0.04 percent of sessions in the same data.

So I say: “For News Site SEO is not about Google Search, but Google Discover Optimization.”

How the feed decides: the social-media-ization of news

I call it “The social-mediasation of news content.” Google’s documentation describes the eligibility. My model, built from newsroom data and workshops, describes the behaviour:

• Interaction drives reach. “As soon as you interact with an item, as soon as you scroll slower … you get more of the same. Additionally you get the same stuff as other users that have a similar usage and behavior profile as you.”
• It is a popularity contest. A reader clicks and interacts, the story is shown to more readers like them, and it starts again.
• It is spiky. In one newsroom’s 16-month data, the best single Discover article brought as many clicks as about two and a half weeks of the whole site’s Search traffic. The best Discover day had about eight times the clicks of a normal day.
• Headline and image are the whole card. “Headline (which is the H1, not the SEO title), Image and Timing. The strongest leverage you have … is the headline and the image. Because that is what shows up in the Google Discover feed.”

The hard part is alignment: a clickable H1, an image that stops the scroll, and a story that delivers what both promise.

Every Google best practice for Discover

This is everything Google’s own Discover documentation recommends, read on 2026-11-05, with what I do with each point.

Eligibility: indexed and within the content policies

• “Content is automatically eligible to appear in Discover if it is indexed by Google and meets Discover’s content policies.” (source) The policies are on Google’s Discover content policies page.
• “No special tags or structured data are required.” (source)
• “Note that being eligible to appear in Discover is not a guarantee of appearing.” (source)
• “If your site violates one or more Discover content policies, Discover manual actions may appear in your Search Console under Security and Manual actions.” (source)

What I do: check the Manual actions report first in every audit, then the 3 SEO tests, because “indexed” has to be true for the rendered page, not only for the HTML.

Helpful content, the same systems as Search

• “As part of Google Search, Discover makes use of many of the same signals and systems used by Search to determine what is helpful, people-first content.” (source)
• Since February 2026, Google says “our systems are designed to identify expertise on a topic-by-topic basis” (source).

What I do: build the sections you want to win in Discover as real sections, with enough stories, their own feed and their own authors.

Titles and headlines: no clickbait, no sensationalism

• “Avoid clickbait and similar tactics to artificially inflate engagement by using misleading or exaggerated details in preview content (title, snippets, or images) to increase appeal, or by withholding crucial information required to understand what the content is about.” (source)
• “Use page titles and headlines that capture the essence of the content.” (source)
• “Avoid sensationalism tactics that manipulate appeal by catering to morbid curiosity, titillation, or outrage.” (source)

What I do: keep the click trigger, drop the false promise. The H1 names who, what, where, plus the reason to click.

Timely stories, told well

• “Provide content that’s timely for current interests, tells a story well, or provides unique insights.” (source)
• “Older content may appear if it’s helpful and relevant to individuals based on their interests.” (source)

What I do: an evergreen update routine that never changes the URL. See below.

Images: large, relevant, enabled

• “Include compelling, high-quality images in your content that are relevant, especially large images that are more likely to generate visits from Discover.” (source)
• Google’s specification: “At least 1200 px wide” (source), “High resolution of more than 300,000 total pixels” (source) and a 16x9 aspect ratio.
• “Google tries to automatically crop the image for use in Discover.” (source) If you crop yourself, keep the important details in the crop you name in og:image.
• “Enabled by the max-image-preview:large setting, or by using AMP” (source).

What I do: one source image, three crops, linked to its large version. See large, linked images.

The preferred image: schema.org or og:image, never the logo

• “Use either schema.org markup or the og:image meta tag to specify a large image that’s relevant and representative of the web page, as this can influence which image is chosen as the thumbnail in Discover.” (source)
• “Avoid using generic images (for example, your site logo) in the schema.org markup or og:image meta tag.” (source)
• “For best results, avoid using text-heavy images in the schema.org markup or og:image meta tag.” (source)

What I do: both, consistent: the same story image in the NewsArticle image array and in og:image.

Page experience

• “Provide an overall great page experience.” (source)

What I do: PageSpeed Insights at least 80, better 90, on the article template, after every release.

What Discover filters out

• “For example, Discover might not recommend job applications, petitions, forms, code repositories, or satirical content without any context.” (source)
• “Discover makes use of SafeSearch but beyond that filters out content that might be seen as shocking or unexpected.” (source)
• “Discover regularly shows content including but not limited to sports, health, entertainment, and lifestyle content from across the open web.” (source)

Traffic swings are part of the channel

• “Traffic from Discover is less predictable or dependable when compared to keyword-driven search visits.” (source)
• The reasons Google lists: changing interests, changing content types and updates to Search. “If someone is no longer as interested in a particular topic”, their feed shows other content (source).
• “However, it may be that there’s nothing you need to do after an update.” (source)
• “Our ongoing work to improve Discover’s user experience means sites may see changes in their traffic that are unrelated to the quality or publishing frequency of their content.” (source)

The February 2026 Discover core update

Google lists three changes (source):

• “Showing users more locally relevant content from websites based in their country” (source)
• “Reducing sensational content and clickbait in Discover” (source)
• “Showing more in-depth, original, and timely content from websites with expertise in a given area” (source)

It started with English in the US: “We’re releasing this update to English language users in the US, and will expand it to all countries and languages in the months ahead.” (source)

Measuring: the Discover performance report

• The report “shows impressions, clicks, and CTR for any of your content that has appeared on Discover in the last 16 months, as long as your data reaches a minimum threshold of impressions.” (source)
• “The Discover performance report includes traffic from Chrome, and fully tracks a site’s Discover traffic across all surfaces where users interact with Discover.” (source)

My Discover best practices, beyond Google’s documentation

Google’s documentation is the floor, not the playbook. These are the practices I teach newsrooms.

They need to click you

“Its a social media algo for news … the more clicks it gets, the more impressions it gets … To do 1: They need to click you!”

• The click trigger goes into the H1 and into the image.
• The reader sees the headline without the context of your home page. It has to work alone.
• If the headline promises something the page does not deliver, the reader bounces back and does not return soon.

They need to click something

“The more interaction it gets, the more impression it gets … To do 2: They need to click something!”

• A read-more link after the first paragraph, at the latest after the second: on my slides, “the most important editorial process” for more traffic.
• Polls, videos and galleries added by a second editor.
• A block of links to other stories, so every reader has a next click.

Image language

“Discover is a visual medium.” The rules I give the photo desk:

• A person beats an object. Action beats stillness.
• Your own footage beats the agency photo.
• Colour beats brown and grey. Authentic beats generic.
• Use what your social media team knows about images.

Large, linked images

1. max-image-preview:large in the robots meta tag.
2. One source image, cropped server-side to 1200 px in 16:9, 4:3 and 1:1, all three in the NewsArticle image array.
3. An og:image at 1200 x 675 with width and height.
4. The image linked to its large version. The click may be cancelled with JavaScript.
5. srcset and sizes, compressed files, a descriptive file name and alt text, width and height.
6. Only the hero image is not lazy-loaded, with fetchpriority="high".

<meta name="robots" content="index, follow, max-image-preview:large">
<a href="/images/mayor-opens-new-bridge-1200.jpg">
  <img src="/images/mayor-opens-new-bridge-640.jpg"
       srcset="/images/mayor-opens-new-bridge-640.jpg 640w,
               /images/mayor-opens-new-bridge-1200.jpg 1200w"
       sizes="(max-width: 640px) 100vw, 640px"
       width="640" height="360" fetchpriority="high"
       alt="The mayor cutting the ribbon on the new bridge">
</a>

NewsArticle, feeds and the Google News sitemap

• NewsArticle markup is critical for Discover in my practice; breadcrumbs are good to have; everything else is nice to have. Validate it in the Rich Results Test.
• One stable publisher name on every page.
• An RSS feed per section, linked in the <head> of the start page and of every article.
• A Google News sitemap, submitted in Search Console and in the Publisher Center.
• AMP is optional. If you run it, it must validate and carry the same NewsArticle markup.

{
  "@context": "https://schema.org",
  "@type": "NewsArticle",
  "headline": "The same words as the H1",
  "image": [
    "https://example.com/images/story-16x9-1200.jpg",
    "https://example.com/images/story-4x3-1200.jpg",
    "https://example.com/images/story-1x1-1200.jpg"
  ],
  "datePublished": "2026-11-05T08:00:00+01:00",
  "dateModified": "2026-11-05T11:30:00+01:00",
  "author": [{ "@type": "Person", "name": "Author Name", "url": "https://example.com/authors/author-name" }],
  "publisher": { "@type": "Organization", "name": "Your Publisher", "logo": { "@type": "ImageObject", "url": "https://example.com/logo.png" } }
}

The 3 SEO tests

1. PageSpeed Insights: at least 80, better 90, mobile and desktop. If you cannot pass it, AMP is the fallback.
2. “JS turned off”: above the fold and the main content visible without JavaScript; visible links must be links.
3. Search Console live test: Inspect URL, Test Live URL, View Tested Page. The screenshot must show the rendered page.

Evergreen, updated

• Update evergreen articles at least once a season or once a year.
• Change everything but the URL. Never change the URL after publishing, and never the targeting.
• Update the date and move the story back to the start page.

Read each channel apart

• Discover, Search and Google News are separate reports in Search Console.
• Google News results that appear inside Google Search are reported as Search.
• Keep a “did work / did not work” list of your own Discover stories.

Where my practice goes further than Google’s documentation

Google’s documentation says	What I do, and why
No special tags or structured data are required	NewsArticle markup is critical: it carries headline, dates, author and the three image crops
No Google News sitemap is mentioned for Discover	A Google News sitemap and managed Google News visibility, or Discover stays luck
Titles and headlines should capture the essence	The H1, not the SEO title, is the feed headline: it gets the click trigger
Treat Discover as supplemental to Search	For news sites it is the main Google channel, so run it as one
There may be nothing to do after an update	Wait until it is over, take the new baseline, then act on the data

Run Discover as a process, not a checklist

Discover is a channel, and a channel can be run. The cycle never goes away:

1. Topic setting from your own Discover data.
2. Publishing aligned: H1, image and story checked together.
3. Measuring impressions, clicks and click rate per story, desk and market.
4. Feedback loop: what worked, what did not, back to the desks.
5. Testing: one change, one desk, a start date and a stop rule.
6. Repeating, and after every Google update a new baseline.

Every newsroom needs its own task lists and processes for the editorial, development and marketing teams, specified together with them. That is what my Google Discover & News Feed Workshop builds.

What I measured in newsroom data

• After one site change, the same URLs got about 75 to 80 percent of their Discover clicks back in every market but one, and under 2 percent in the one large foreign market. The cause was on Google’s side.
• In that market, impressions fell to under 2 percent while the click rate stayed at about 10 percent: the feed stopped showing the stories, readers did not reject them.
• Headlines with sensational markers made up about 30 percent of that loss.
• During one core update, a newsroom’s weekly Discover clicks about halved for three weeks, then came back. “you can not ‘fix’ Google”.
• In one audit, the article images were about 600 px wide; in another, the srcset promised about 3,800 px but served at most about 1,000 px.

The full cases are on the workshop page.

Sources: Google’s Discover and your website (last updated 2026-03-09) and Google’s February 2026 Discover Core Update (published 2026-02-05), read on 2026-11-05; my own audits, workshops, slides and second opinions from 2023 to 2026, with client numbers rounded and anonymised.

Frequently asked questions

Do I need special markup for Google Discover?

Google says no special tags or structured data are required. I still treat NewsArticle markup as critical, because it carries the headline, the dates, the author and the three image crops.

Do I need a Google News sitemap for Discover?

Google does not require it. I do, if you want Discover as a channel you can repeat. Discover and Google News are closely linked; without news visibility Discover stays an occasional hit.

Why did my Discover traffic drop?

Date the drop and check Google's update dates first. Most big drops I analysed were on Google's side. Wait until an update is over, read impressions, clicks and click rate per market, take that as the new baseline and act on it.

Which image size does Discover need?

Google recommends images at least 1200 px wide, more than 300,000 pixels, in 16x9, enabled by max-image-preview:large. I ship one 1200 px source image in 16:9, 4:3 and 1:1 and link it to its large version.

Does clickbait work in Discover?

No longer. Google's February 2026 Discover update reduces sensational content and clickbait, and clickbait headlines lost the most in my own 2026 data. Keep the click trigger, drop the false promise.