Google Discover Optimization: Eligibility, Images and Reporting

Most site owners think of Google traffic as the result of someone typing a query. Google Discover works the other way round: it is a feed of articles, videos and other content that Google shows people based on their interests and activity, before they search for anything.

This guide explains Google Discover in plain terms: who can appear in it, what Google actually asks of publishers, how to prepare images and titles that suit a feed, and where to see your numbers. It is written for owners of blogs, news sites, shops with editorial content and small businesses that publish regularly. You will finish with a short preparation routine and a way to read the Discover report without over-reading it.

No switch, tag or markup enrols a page in Discover, and nobody outside Google can promise a placement. Everything below is about removing obstacles.

Short answer: Google Discover is a personalised feed in Google Search. Pages are automatically eligible if they are indexed and meet Discover's content policies, but eligibility does not guarantee appearing. You can improve your chances with accurate titles, helpful content, large high-quality images and the large image preview setting, then monitor results in Search Console.

Before you begin

You need a verified property in Google Search Console, because the Discover report lives there. If you have not set one up, follow Set Up Google Search Console: Complete Website Owner Checklist first. You also need edit access to your page templates or CMS settings (to adjust robots meta tags and image fields) and a few recent articles to use as test cases.

Allow about an hour. The risk is low, but a wrong robots meta tag value can limit how pages appear in Search generally, so note your current settings first.

What is Google Discover and who can appear in it?

Google describes Discover as part of Google Search that surfaces content based on interests and Web and App Activity. It is a feed, not a results page, so there is no query to match.

Google's documentation states that content is automatically eligible if it is indexed by Google and meets Discover's content policies. It also states that no special tags or structured data are required. Being eligible is not the same as being shown: Google says plainly that eligibility is not a guarantee of appearing.

That has three practical consequences:

Indexing comes first. A page that is not indexed cannot be a candidate. If you are unsure, use the steps in How to Check If a Page Is Indexed by Google.
Policy compliance is a gate. Google says Discover may not recommend certain content types, including job applications, petitions, forms and code repositories, and that it filters content seen as shocking or unexpected. Policy violations can lead to Discover manual actions, which appear in Search Console under Security and Manual actions. See Search Console Manual Actions for how that area works.
Older content is not excluded. Google notes that older content can also appear if it is helpful and relevant.

How is Discover different from ordinary search traffic?

Treat Discover as a bonus channel that comes and goes. Google says traffic from Discover is less predictable or dependable than keyword-driven search visits, and that it can change because people's interests change, because the kinds of content shown shift, and because of Search updates. Google adds that a change is not necessarily a sign that content quality or publishing frequency changed.

Question	Ordinary search	Google Discover
What starts the visit?	A typed or spoken query	A feed that appears without a query
Can you target a keyword?	Yes, by matching intent	Not in the same way; interests drive it
How steady is it?	Usually more repeatable	Google calls it less predictable

So do not build a forecast on Discover numbers. Use them as a signal about which topics and images connected with readers. For keyword-led visits, start with Search Intent: Match Your Page to What Searchers Want.

Step-by-step instructions

Step 1: Confirm your best pages are indexed and clean

Pick five to ten recent articles and inspect each in Search Console. If any is not indexed, fix that first. Excluded by "noindex" Tag in Search Console and Robots.txt Blocking Google or Bing? cover the usual causes.

How to confirm it worked: each page shows as indexed in URL Inspection, and Google can fetch the page and its main image.

Step 2: Make the headline say what the page delivers

Google's guidance is to use page titles and headlines that capture the essence of the content, and to avoid clickbait, misleading or exaggerated previews and sensationalism. A title the page cannot back up may breach the content guidance.

If the first paragraph does not deliver what the headline implied, rewrite one of them. The wider rules for titles are in SEO Title and Meta Description Guide.

How to confirm it worked: a colleague who has not seen the page can say, from the headline alone, what the page is about, and is not surprised by the content.

Step 3: Publish content that is timely, well told or genuinely insightful

Google describes content that does well as timely, telling a good story, or offering unique insight, written for people first with a good page experience. There is no word count or posting schedule to hit. Use the questions in Helpful Content Checklist to review each page: does it show first-hand experience, does it answer the whole question, and would you be comfortable putting your name on it?

For a small business, unique insight usually means your own photos, data or customer questions.

How to confirm it worked: you can name what each article offers that similar pages do not.

Step 4: Prepare a large, relevant image

Images are the heart of a feed card. Google's Discover guidance asks for large, high-quality images that are at least 1200 px wide. Its description of "high resolution" is more than 300,000 total pixels, with a 1280 by 720 image (16:9) given as an example. The guide also says to avoid generic images such as a site logo and, for best results, text-heavy images.

Specify the image with either schema.org markup or the og:image meta tag. If you crop a vertical photo to 16:9, make sure the important part survives the crop in the version you point to. A typical head snippet looks like this:

<meta property="og:image" content="https://example.com/images/harbour-market-1280x720.jpg">

For wider image advice, see Google Images SEO Checklist and Open Graph and Twitter Card Meta Tags: Setup Guide. If your files are heavy, Compress Images for the Web shows how to shrink them without ruining quality.

How to confirm it worked: open the live page source, find og:image, load that URL in a browser and check that it is at least 1200 px wide, relevant to the story and not a logo.

Step 5: Allow large image previews

The large preview matters because Google's page lists the max-image-preview:large setting as the way to enable large images (the page also mentions AMP as an alternative). The robots meta tag documentation defines the values: none allows no image preview, standard a default-size preview, and large a larger preview up to the width of the viewport.

<meta name="robots" content="max-image-preview:large">

Check your templates for an existing robots meta tag before adding another. Google's robots documentation notes that setting standard or none prevents larger thumbnails in Discover, which is useful only if you do not want them. Read Meta Robots Directives Explained before editing templates; a stricter rule can override your intent.

How to confirm it worked: view the source of an article and find a single robots meta tag containing max-image-preview:large, with no conflicting tag.

Step 6: Protect the page experience

Google asks for a good page experience in general. A reader who taps a card and meets a slow page or a layout that jumps has wasted the click. Check mobile layout and speed with Mobile-First Indexing Checklist and Core Web Vitals Explained. These are not Discover switches.

How to confirm it worked: the article opens quickly on a phone and the layout does not jump.

Step 7: Find and read the Discover report

In Search Console, the Performance report has a Discover view. According to Google's help page, the report is visible only if your property has reached a minimum number of impressions in Discover. If you do not see it, that usually means there is not enough data yet, not that something is broken.

When it does appear, it shows impressions, clicks and average CTR. An impression is counted when a Discover item scrolls into view (once per session per item), and a click when a user taps it; shares do not count. Rows below a minimum impression threshold are left out of the table but still count toward totals. Metrics are attributed to the canonical URL, and the default range is the last three months. Google's Discover page adds that the report includes traffic from Chrome and covers surfaces where people use Discover.

Today's data, and sometimes yesterday's, can be preliminary and may change. Look for patterns over weeks, not single days. Google Search Console Performance Report: Queries, Pages and CTR explains the general layout, and GA4 and Search Console Integration shows how to compare on-site behaviour with these numbers.

How to confirm it worked: you can open the Discover view (or explain that your property has not yet reached the threshold), sort pages by clicks and name the three pages that attracted the most impressions.

What the report can teach you

Once data exists, ask practical questions rather than chasing totals. Which pages appear most, and what do their images and headlines share? Is the click-through rate on one page much lower than on others, which can point to an unclear image or an over-promising headline? Did a spike follow a change you made, or a general shift? Google warns that fluctuations can be unrelated to your actions. A simple sheet with URL, publish date, image, headline and Discover clicks will show patterns after a few months that no single week can.

Final verification checklist

Your key articles are indexed, crawlable and not noindexed.
Headlines match the content and avoid clickbait or exaggeration.
Each article offers something original, timely or useful.
A relevant image at least 1200 px wide is specified through og:image or schema.org markup, and it is not a logo.
A robots meta tag allows max-image-preview:large with no conflicting directive.
You know where the Discover report is, or that your property has not yet qualified for it.
You have checked Security and Manual actions in Search Console for any Discover-related issue.
You are treating Google Discover numbers as variable, not as a forecast.

Common mistakes to avoid

Looking for a Discover tag. None exists; Google says no special tags or structured data are required. Spend the time on content and images instead.
Using the site logo as the page image. Google's guidance says to avoid generic images. Use a picture that belongs to the story.
Cropping the subject out. A vertical photo forced into 16:9 can lose the point of the picture. Crop manually and point to that version.
Writing sensational headlines. Misleading previews go against the content guidance. Make the headline a fair summary.
Blocking large previews by accident. A template that sets max-image-preview:standard or none stops large thumbnails. Check the live HTML, not only the CMS field.
Reading one good week as a trend. Discover is described as less predictable. Compare months.
Ignoring policy messages. Discover manual actions appear under Security and Manual actions; check there if visibility disappears.

Frequently asked questions

Do I need to apply to be in Google Discover?

No. Pages are automatically eligible if they are indexed and follow the content policies. There is no application, and being eligible does not mean you will appear.

Does structured data help me appear in Discover?

Google says no special structured data is required. You can use schema.org markup or og:image to specify the large image, but the markup does not enrol the page. See Article Schema Markup for general use.

Why is my Discover report missing?

Google says the report is visible only after your property has reached a minimum number of impressions in Discover. If you have not reached it, the report will not show, and you can keep publishing and check again later.

Why did my Discover traffic suddenly drop?

Google says Discover traffic is less predictable and that changes can come from shifting interests, different content types being shown or Search updates. Check for indexing, policy or technical issues first, then keep to the guidelines.

Official references

Next step: Try the free website sample to see how your own site does on a few of these checks. This guide provides technical information, not a guarantee of rankings, traffic or index inclusion.