WHAT IS GOOGLE DİSCOVER AND HOW DO YOU GET INTO IT?

Google Discover is a feed that recommends content without the user searching for anything. In classic search there is a query and you answer it; in Discover there is no query, and Google brings content to the user based on their interests. In a mobile-heavy market it carries serious traffic potential, but the nature of that traffic differs from classic search. No application, no registration. Getting into Discover has exactly two conditions in 2026: being indexed, and complying with the content policies. Below we cover, based on Google's own documentation, how Discover works, the eligibility conditions, image requirements, content policies and why traffic spikes and collapses so sharply.

What Is Google Discover?

By Google's official definition, Discover is a part of Google Search that shows people content related to their interests, based on their Web and App Activity. The core mechanic is this: the user runs no search, and Google presents content it judges might interest them as a feed, based on past behaviour. Discover therefore works on recommendation rather than demand. In classic SEO work your goal is to be the answer to a query; in Discover your goal is to have content that catches the attention of a user with a particular interest. The distinction changes everything from content selection to headline writing and requires a different way of thinking from classic search engine logic.

Where Does Discover Appear?

Discover is a mobile-weighted surface. It appears in the Google app on Android and, on many devices, in the feed to the left of the home screen, in the Google app on iOS, and on the new tab page in mobile Chrome. No official Google confirmation could be found that the Discover feed runs on desktop, so do not plan around expectations of desktop Discover traffic. The mobile nature has practical consequences: image size and legibility become critical, the headline must be understandable without truncation on a small screen, and the user who clicks lands directly on your mobile page. Your page's mobile performance is therefore part of Discover success; we covered mobile-first indexing and page experience in our technical SEO guide.

The Core Difference From Classic Search

The core difference fits in one sentence: classic search answers demand that already exists, while Discover manufactures it from a guess about interest. The difference between the two surfaces is not only the presence of a query. In classic search the user arrives with an intent and you meet it; in Discover the user has no intent, only an interest. Keyword targeting therefore does not work directly in Discover. The second difference is measurement: the Discover report in Search Console has no query dimension and no average position metric, because there is no query to rank for. The third is timing; in classic search a good page can bring traffic for years, while in Discover the traffic ends when the content leaves the feed. The fourth is predictability, which Google states plainly in its own documentation. Keyword research and search intent are decisive on the classic side, while in Discover the appeal of the content and the strength of the image come forward.

Is Anything Special Required to Appear in Discover?

Let us correct a common misconception up front: you do not register for Discover and no special markup is required. In Google's official wording, content is automatically eligible to appear in Discover if it is indexed by Google and meets Discover's content policies. The same document adds another clear sentence: no special tags or structured data are required, and being eligible is not a guarantee of appearing. So part of what is sold as "Discover optimisation" is selling a mechanism that does not exist. Eligibility comes down to two things: being indexed and complying with the policies. The indexing side is ordinary technical SEO work: crawlability, indexability and page health. The only Discover-specific technical setting is the large image preview tag described below.

What to Check If You Never Appear in Discover

If you get no impressions at all, the problem is usually eligibility rather than ranking. Work through these five in order:

• Indexing: is the page indexed by Google? The URL Inspection tool in Search Console answers that in one step.
• Image preview: is max-image-preview:large set in the robots meta tag, and is the featured image wider than 1200 pixels?
• Byline block: are the publication date, author name, publisher identity and contact details visible on the page? The policy asks for them directly.
• Headline honesty: does the headline promise something the page does not deliver?
• Report threshold: if the Discover report never opens, your property may not have passed the minimum impressions bar. The threshold is not a penalty; it is a consequence of having no visibility yet.

Running the list once takes half an hour. After that the only variable left is the content itself.

Image Requirements and the max-image-preview Tag

The image is the strongest element determining clicks on Discover cards, and Google states the requirements numerically. For a large image the criteria are: at least 1200 pixels wide, more than three hundred thousand total pixels and a 16:9 aspect ratio, with the documentation's example being 1280x720. Enabling the large preview depends on a robots meta tag: in Google's robots meta specification, the max-image-preview:large value means a larger image preview, up to the width of the viewport, may be shown. Without the tag Google may show a default-sized, that is small, preview and you lose your visual advantage in Discover. The same directive can also be delivered through the X-Robots-Tag HTTP header. A detail that often gets missed in the documentation: Google requires either the max-image-preview:large tag or an AMP page for the large preview. Either route works. We covered where AMP stands today, and why most sites no longer choose it, in our Google AMP guide; in practice the robots meta tag is the right road. The practical task is to make sure max-image-preview:large is set across your site and to produce featured images at least 1200 pixels wide.

Content Policies and the Risk of Removal

Discover is subject to a stricter policy set than search results, and a violation can result in content being removed from the feed. The Discover content policies cover Google's general policies: dangerous content, deceptive practices, harassment, hate speech, manipulated media, medical content rules, sexual content, violence and profanity. Three Discover-specific items deserve extra attention. First, disguising or misrepresenting sponsored content as independent editorial content is prohibited. Second, previews that mislead users by promising details not actually in the content are prohibited. Third, sources must provide a clear date, byline, author, publisher and contact information. The last point matters especially, because many sites leave author and contact details incomplete; the same transparency signals are also weighed on the E-E-A-T side.

What Works in Discover and the Clickbait Trap

Google lists what it expects for Discover in its own documentation: headlines that capture the essence of the content, avoiding clickbait and tactics that artificially inflate engagement, staying clear of sensationalism, content that is timely or well told or offers unique insight, and compelling high-quality images. The tension in that list is notable: Discover rewards content that sparks interest but penalises exaggerated headlines. The balance that works in practice is a headline that raises curiosity but delivers on its promise on the page. A headline that does not deliver earns clicks in the short term and grows both policy risk and user dissatisfaction over time. On content type, timeliness, human-interest topics, visually strong storytelling and pieces offering a clear point of view come forward. Helpful content principles apply here as in classic search; Google's helpful content guidance provides the basic frame.

The Nature of Discover Traffic: Sharp Spikes, Sharp Drops

What surprises people most about Discover traffic is its instability, and Google acknowledges this itself. In the official documentation's wording, traffic from Discover is less predictable or dependable when compared to keyword-driven search visits. The same document lists three reasons for the fluctuation: changes in user interests, Discover adjusting the types of content it shows, and improvements made in Google Search. The Search Console help page phrases the difference similarly: Discover traffic is somewhat less predictable than Search traffic, in part because it is proactively served by Google rather than dependent on user queries. The practical result is that a page bringing tens of thousands of visitors one day and zero the next is normal and usually not a penalty. Google's recommendation is clear too: treat Discover traffic as a complement to keyword-driven search traffic rather than a replacement. A publisher who builds a business model solely on Discover becomes dependent on a variable they cannot control.

The Search Console Discover Report

The Discover report in Search Console appears conditionally: in Google's words, the report is visible only if your property has reached a minimum number of impressions in Discover. The numeric value of that threshold is not published, so disregard sources quoting a specific figure. The report contains three metrics: impressions (counted when the card is scrolled into view), clicks and average click-through rate. There is no average position metric and no query dimension, because there is no query. The available breakdowns are page, country, date and Discover appearance type, with a data window of the last sixteen months. The table is limited to a thousand rows, and rows below the threshold are not shown but are included in totals. The two most valuable analyses when reading it are seeing which page types enter the feed, and inferring the performance of your image and headline from the relationship between impressions and click-through rate.

• The Discover appearance breakdown separates AMP articles, News Showcase, Web Stories and videos. Which format enters the feed becomes a reading rather than a guess.
• Only one impression is counted per result per session. If the user scrolls back and sees your card a second time, the counter does not move; impressions measure unique encounters, not raw views. That is exactly where people go wrong when they compare Discover click-through rate against page-view numbers from a news site.

Discover and E-E-A-T

There is no known list of ranking factors for Discover and Google does not publish one, so claims about definite factors are speculation. What can be derived from the policy text, however, are clear expectations, and most of them overlap with trust signals: a clear publication date, author byline, publisher identity and contact information are directly required by the policy. Those same elements are core components of the experience, expertise, authoritativeness and trustworthiness framework, which we detailed in our E-E-A-T guide. The second connection is topical consistency: because Google recommends by user interest, sites that publish consistently on a topic become more strongly associated with that interest. That association is fed by your brand being clearly recognised on a specific topic. Google's February 2026 Discover core update stated plainly that expertise is judged on a topic-by-topic basis, so a site that goes deep on one subject and a site that covers several are each measured per topic. Which signals actually carry authority, and which are only tool metrics, we separated in our real authority signals article. In short, there is no separate trick for Discover; trust and topical clarity do the same work here.

Discover in a Mobile-Heavy Market

Markets where internet use skews mobile and the Google app is widespread are favourable ground for Discover. The practical consequence for publishers is that mobile experience and image quality are far more decisive than a desktop-oriented approach would suggest. On localisation, the point to note is that Discover works by user interest, so if you publish in a given language, topics inside that audience's interest ecosystem carry a natural advantage. One caveat is needed. Google's February 2026 Discover core update was the first core update aimed solely at the Discover feed, and it launched for English language users in the US. The announcement named three goals: showing users more locally relevant content from sites based in their own country, reducing sensational content and clickbait, and surfacing sources with demonstrated expertise in a given area. The same announcement says the rollout is planned for all countries and languages in the months ahead. So for every other market the question is when, not whether. Until a rollout is confirmed where you publish, disregard content that pins past traffic swings on that update, and note the direction of travel: local publisher identity and topical focus are a safer bet than a sensational headline. Read your own data from the Search Console Discover report; the trend of your own property is a better guide than general claims.

Turning Discover Traffic Into Lasting Value

The biggest risk with Discover is that incoming traffic vanishes without trace. A sudden spike brings tens of thousands of visitors and none of them return the following week. The ways to convert it into lasting value are these. First, internal linking: from the page receiving Discover traffic, give clear links to other content that sustains the topical interest, since the visitor came precisely because they care about that topic. Second, a durable channel offer: an email newsletter or a social follow turns a one-off visitor into someone who can come back. Third, a conversion path: keep a single clear next step on the page. Fourth, measurement discipline: record which content types entered the feed and extract the repeatable pattern. Finally, set expectations correctly; Discover traffic is a complement to your organic strategy rather than the whole of it. We covered lasting growth on the search side in our organic traffic guide. The smallest step you can take today: open the Discover report in Search Console, list the first five pages that entered the feed, and give each one a single durable next step. Forty minutes of work. It also decides how many of those visitors come back after the next spike.