How to Optimize for Google Discover and News SEO

TL;DR

Google Discover reaches hundreds of millions of people monthly and news domains capture a disproportionate share of its clicks, but eligibility runs through E-E-A-T signals, max-image-preview:large, 1,200px+ images, and a pruned news sitemap. Publish fast, cite sources, avoid clickbait, and keep structured data clean.

Google Discover and Google News operate on different mechanics than traditional organic search, and most SEO teams still treat them as an afterthought bolted onto their regular content workflow. That is a mistake. Discover has reached hundreds of millions of people monthly since Google's original rollout, and for publishers who qualify, it routinely outperforms organic search in raw traffic volume (Search Engine Land). News domains, despite representing less than half of the URLs that appear in Discover feeds, capture a wildly outsized share of the clicks, which tells you something important: recency, authority, and topical trust matter more here than they do almost anywhere else in Google's ecosystem.

This guide covers what actually moves the needle in 2026, technical setup for News eligibility, the editorial signals Discover rewards, and the specific mistakes that get otherwise-qualified content excluded. None of this replaces a solid technical foundation. If your technical SEO is shaky, Discover and News amplify that weakness because both surfaces are less forgiving of crawl and rendering problems than standard search results.

Why Discover and News Deserve Their Own Playbook

Google Discover is not a search result. There is no query. It is a personalized, algorithmically curated feed on the Google app and Chrome's new-tab experience that shows users content based on their interests, browsing history, and location, not something they typed. Google News and the Top Stories carousel work differently again: they are query-triggered but run on a compressed timeline where freshness and structured data quality decide who gets shown in the first hour after publication.

Both surfaces reward different signals than a typical ranking factor checklist.

Discover cares about topical affinity, image quality, and freshness of interest, not keyword matching.
Top Stories cares about publication speed, timezone-correct timestamps, and clean NewsArticle markup.
Neither surface uses HowTo or FAQ rich-result markup for placement. Google removed HowTo rich results in 2023 and FAQ rich results on May 7, 2026, though the schema still helps AI systems extract and understand your content even without a SERP rich-result payoff.
Both are increasingly influenced by Google's "Preferred Sources" feature, which lets users manually pin outlets they want prioritized in Top Stories, and as of an April 2026 expansion, in every language Google Search supports (Search Engine Journal).

The strategic implication: you cannot game your way into Discover or News with technical tricks alone. You need genuine editorial trust signals plus flawless technical execution. One without the other caps your visibility.

The Technical Foundation: What Google Actually Requires

Before any editorial optimization matters, your site needs to clear a set of hard technical gates. Google's own developer documentation is explicit about the image and markup requirements for Discover eligibility (Google Search Central).

Image Requirements

Minimum image width of 1,200 pixels.
Total resolution above 300,000 pixels, so a 1280x720 image clears the bar comfortably.
16:9 landscape orientation performs best in the Discover card format.
The max-image-preview:large meta robots tag must be present, or the page must be valid AMP, or Discover will suppress large image previews entirely.
Avoid logos, watermarks, or heavy text overlays in the featured image; Google's guidance explicitly discourages generic branding images as the primary visual.

News Sitemap Rules

If you are pursuing Top Stories or Google News inclusion, a dedicated news sitemap is close to mandatory, separate from your standard XML sitemap. Google's documentation lays out firm limits (Google Search Central):

Only include articles published in the last 48 hours. Once an article ages past that window, strip the <news:news> block or remove the URL entirely.
Cap each sitemap file at 1,000 URLs; use a sitemap index if you publish more.
Every <news:publication_date> needs a full, timezone-aware ISO 8601 timestamp. A missing or ambiguous timezone is one of the most common reasons articles silently fail to appear in Top Stories.

Structured Data

Use Article, NewsArticle, or BlogPosting schema with complete headline, datePublished, dateModified, author, and image properties. This is a floor requirement, not a differentiator. Google explicitly does not use HowTo or FAQ schema for rich-result placement anymore (HowTo since 2023, FAQ since May 7, 2026), but both schema types still help AI crawlers and answer engines parse and extract your content accurately, even when it does nothing for your classic SERP appearance.

Core Web Vitals and Page Experience

Google dropped the AMP requirement for Top Stories in 2021, but page experience did not become optional, it became the actual gate. Slow-loading article pages, layout shift from ad injection, and render-blocking JavaScript that delays the main content from appearing are common reasons technically "eligible" articles never surface. If you have not run a full technical pass recently, our 40-point technical SEO audit checklist covers the Core Web Vitals and crawl-health issues that affect Discover and News eligibility specifically, not just standard organic rankings.

Every minute between publish and Googlebot's first crawl narrows the Top Stories eligibility window.

Editorial Signals: What Discover's Algorithm Rewards

Google's own guidance for Discover ties eligibility directly to E-E-A-T, the same trust framework it uses for core ranking systems (Google Search Central). This is not boilerplate. Google explicitly instructs publishers to avoid content designed to exploit "morbid curiosity, titillation, or outrage," and to write headlines that reflect the actual content of the page rather than manufacturing false urgency.

In practice, that translates into a short list of behaviors that separate publishers who get consistent Discover traffic from those who get occasional, unpredictable spikes.

Write accurate, specific headlines. Curiosity gaps without payoff get suppressed over time as engagement signals sour.
Publish original reporting or original analysis, not aggregation. Rewritten wire copy or lightly modified competitor content underperforms.
Update evergreen explainers when the underlying facts change, and reflect that in a visible, accurate dateModified.
Cite primary sources inside the article, not just in a footnote. This also happens to be one of the strongest levers in generative-engine optimization: the Princeton and Georgia Tech GEO study found that adding statistics lifted visibility by up to 41%, adding quotations by up to 28%, and citing sources by up to 115% for pages that started with weak visibility (arxiv.org/abs/2311.09735).
Maintain topical consistency. Discover's personalization engine builds an interest graph per user; publishers with a clear, consistent topical footprint get surfaced to the right audience more reliably than generalist sites. This is the same logic behind topical authority building for core search, just applied to a feed instead of a query.

If you want a deeper foundation on the trust signals behind both Discover eligibility and AI-engine citation, our guide on building E-E-A-T signals that Google and AI engines actually trust walks through the author bio, sourcing, and editorial-review patterns that show up across both surfaces.

Discover vs. Top Stories: A Side-by-Side Comparison

Teams often conflate these two surfaces and end up optimizing for the wrong one. Here is how they actually differ.

Factor	Google Discover	Google News / Top Stories
Trigger	Personalized feed, no query	Query-triggered, or homepage carousel
Primary ranking driver	Interest match + freshness of relevance	Publish speed + structured data accuracy
Time sensitivity	Days to weeks (evergreen content can qualify)	Hours (first crawl is often the only shot)
Image requirement	1,200px+ width, `max-image-preview:large`	Standard `image` schema property, no minimum width mandate
Sitemap	Not required	News sitemap strongly recommended, 48-hour window
Content type	Evergreen explainers, guides, timely analysis	Breaking news, developing stories, wire-adjacent coverage
Schema impact on placement	None directly; helps AI extraction only	`NewsArticle`/`Article` completeness is a real gate
Audience discovery	Passive, algorithmic surfacing	Active search intent, plus a passive Top Stories carousel

The practical takeaway: if your content strategy is built around long-form guides and evergreen resources, you are playing the Discover game and should prioritize image quality, headline accuracy, and topical consistency. If you are running a newsroom or a fast-moving industry blog covering developing stories, you are playing the News game and should prioritize sitemap hygiene and publish speed.

Match your optimization priorities to your actual publishing cadence rather than chasing both surfaces evenly.

A Publish-Day Checklist for News and Discover Eligibility

Use this sequence for every article you want to compete for Top Stories placement or Discover surfacing. Skipping steps 3 through 6 is the most common reason technically sound articles never appear.

Confirm the featured image is at least 1,200px wide, 16:9, and free of logos or heavy text overlay.
Set max-image-preview:large in the robots meta tag (or confirm valid AMP if you still run it).
Verify NewsArticle or Article schema includes headline, datePublished, dateModified, author, and image, and that every timestamp carries an explicit timezone.
Push the URL into your news sitemap immediately on publish, not on the next scheduled crawl.
Submit the URL via Google Search Console or an IndexNow-compatible ping so Google's first crawl happens within minutes, not hours.
Re-check the news sitemap 48 hours later and prune the article if it has aged out.
Monitor Search Console's Discover and News performance reports for impression drops, which often signal a policy or quality issue before it becomes a visible ranking problem.

Common Mistakes That Get Content Excluded

Most Discover and News underperformance traces back to a small set of repeat offenders, not algorithmic mystery.

Missing or late news sitemap submission. If the article is not in the sitemap within the first hour, the narrow relevance window for Top Stories can close before Google even crawls it.
Ambiguous or missing timezones on datePublished. Google's crawler treats this as a data-quality signal, not a cosmetic detail.
Clickbait headlines that do not match the body content. Google's Discover policy calls this out directly, and repeated violations lead to reduced surfacing across the whole domain, not just the offending article.
Thin rewrites of wire content or competitor articles. Both Discover and Top Stories increasingly filter for originality; Google's February 2026 Discover core update specifically surfaced more in-depth, original, and timely content from sites with demonstrated topical expertise over repackaged coverage (Google Search Central).
Slow or JavaScript-heavy article templates. If the main content is not available quickly in the render, both eligibility and the user experience suffer.
Ignoring Preferred Sources. Since the April 2026 global expansion, users can pin specific outlets to prioritize in Top Stories in every supported language. Google has reported that users are roughly twice as likely to click through to a source they have marked as preferred. That is a compounding advantage worth actively building toward through consistent, recognizable branding and reader loyalty.
No internal linking or topical reinforcement. Isolated news articles with no connection to a broader content hub tend to underperform on both freshness signals and long-tail Discover surfacing. Structuring coverage into topic clusters helps Google understand the depth of your coverage on a given subject.

How This Fits Into a Broader AI Search Strategy

Discover and News do not exist in isolation from the shift toward AI-mediated discovery. Google's AI Overviews now reduce position-1 organic click-through rates by 58%, and the zero-click search share has risen from 54% to 72% as of Ahrefs' December 2025 analysis (ahrefs.com/blog/ai-overviews-reduce-clicks-update/). Google's AI Mode has separately surpassed roughly 1 billion users in 2026, and it shares only about 13.7% of cited URLs with AI Overviews, meaning the two surfaces pull from meaningfully different source pools (Ahrefs).

Against that backdrop, Discover and News matter more, not less. They remain two of the few remaining high-volume surfaces where a well-optimized article can still generate a genuine click rather than getting summarized away. The same underlying signals, original sourcing, clear attribution, fast technical delivery, also happen to be what AI systems reward when deciding what to cite. Reddit is now the most-cited domain across AI answer engines at roughly 40% of citations overall (about 24% of Perplexity citations as of January 2026 and roughly 12% of ChatGPT US citations), with Wikipedia close behind at around 13% of ChatGPT citations. There is no single dominant source type, which means original, well-attributed publisher content still has real headroom to compete for citations, not just Discover impressions.

This is also why publish-day technical execution and long-term editorial trust need to be managed together rather than as separate workstreams. A content operations platform that routes structured data, image specs, and sitemap submission through the same approval flow as your editorial review keeps both tracks moving without the manual handoffs that cause missed sitemap windows.

Measuring Discover and News Performance

Search Console separates Discover and Web performance into distinct reports, and treating them as one number hides real signal.

Discover report: tracks impressions and clicks specifically from the Discover feed. Watch for sudden impression drops, which usually indicate an image spec violation or a content-policy flag rather than a ranking change.
News/Top Stories visibility: not natively broken out in standard Search Console the way Discover is, so cross-reference query-level impressions for branded and topical terms immediately after publish to confirm Top Stories inclusion.
CTR benchmarks: Discover's average click-through rate runs around 11% across all pages, dropping to roughly 6% when news sites are excluded, according to Search Engine Journal's analysis of over 11,000 Discover URLs across 62 domains (Search Engine Journal), though this varies significantly by vertical and image quality. Compare your own Discover CTR against your standard organic CTR to see whether image and headline quality are underperforming.
Indexation lag: for News-focused publishers, track the gap between publish time and first Search Console impression. Anything beyond 15 to 20 minutes suggests a crawl-priority or sitemap-submission problem worth fixing before it costs you Top Stories placement on the next breaking story.

Per-URL indexation tracking tied to your GSC integration is one of the more underused diagnostics here. If you can see exactly when Google first crawled and indexed a news article relative to publish time, you can isolate whether a missed Top Stories window was a sitemap problem, a crawl-budget problem, or a content-quality suppression, three very different fixes.

Frequently Asked Questions

Does Google Discover require a news sitemap?

No. News sitemaps are for Google News and Top Stories eligibility. Discover works off your standard indexed content and personalization signals, with no separate sitemap requirement, though fast indexing still helps timely content qualify while it is still relevant.

Do HowTo and FAQ schema still help with Discover or News rankings?

Not for rich results. Google removed HowTo rich results in 2023 and FAQ rich results on May 7, 2026. The schema is still valid and can help AI systems extract and summarize your content accurately, but it no longer produces a SERP visual advantage.

How fast does an article need to be indexed to appear in Top Stories?

There is no published exact threshold, but practitioner consensus and Google's own guidance point to minutes, not hours. The first crawl is often the only opportunity a breaking story gets, so submitting through Search Console or an IndexNow ping immediately on publish is standard practice.

What image size does Google recommend for Discover?

At least 1,200 pixels wide, ideally in 16:9 landscape orientation with over 300,000 total pixels, plus the max-image-preview:large meta robots tag or valid AMP markup to unlock large image previews in the feed.

Can evergreen content appear in Google Discover?

Yes. Discover is not limited to breaking news. Evergreen explainers and guides regularly surface in Discover when they match a user's demonstrated interests, particularly after an update that refreshes the content's relevance signal.

What is Google's Preferred Sources feature and does it affect Discover?

Preferred Sources lets users manually pin specific outlets they want prioritized in Top Stories, and as of May 2026, in AI Overviews and AI Mode as well. It is primarily a News and AI-answer signal rather than a direct Discover ranking factor, but the branding and loyalty it rewards overlaps with what drives Discover performance too.

Why did my article disappear from Discover after initially performing well?

Discover traffic is inherently volatile because it tracks shifting user interest, not fixed rankings. A drop can also indicate a policy issue, most commonly a misleading headline or engagement-bait pattern, so check Search Console's Discover report for manual action notices before assuming it is purely algorithmic drift.

Should agencies manage Discover and News optimization differently for clients on tight publishing schedules?

Yes. News-focused clients need sitemap automation and sub-hour indexing checks built into the publish workflow itself, not a manual post-publish step. Discover-focused clients benefit more from image-spec QA and headline review before publish. Building both checks into a shared approval workflow prevents the technical steps from becoming optional under deadline pressure.