SEO for News Websites

TL;DR — News sites have extra SEO surfaces beyond normal search: Google News and Top Stories, the Discover feed, and special handling for paywalls and syndicated (reprinted) articles. This hub gathers the parts that are specific to publishers.

What’s different about news SEO

A news website still needs the same technical foundation as any site, but it plays on surfaces most sites never touch.

Google News and Top Stories surface fresh, newsworthy articles — speed of discovery and trust matter. Google Discover is different again: it’s a feed Google pushes to users based on interests, not a response to a query, so you can’t “rank” for it the way you rank in search. Publishers also have two recurring technical problems: letting Google read paywalled articles without it counting as showing one thing to Google and another to readers, and deciding how to handle syndicated content when the same article runs on multiple sites.

Eligibility on one surface never carries over to another — being in the News tab doesn’t get you into Discover, and ranking in Search doesn’t guarantee either. Each surface has its own eligibility rules, its own ranking factors, and (mostly) its own Search Console reporting, so treat them as separate systems that happen to share a site, not one score you’re optimizing.

This hub is a map, not a manual — the mechanics for each surface live in their own deep dives: Google News SEO, Google Discover SEO, paywalls and SEO, syndicated content, and news sitemaps.

TL;DR — News SEO layers publisher-specific surfaces on top of normal SEO, and two of them changed recently. Google News and Top Stories are now fully automatic — there’s no application. Paywalled content stays indexable without cloaking only if you mark it up with isAccessibleForFree. When a syndicated copy should not compete in Google Search, require the partner to use noindex; Google does not recommend canonical as the control for syndicated copies. Speakable is still a limited beta.

Google News and Top Stories are automatic now

There is no Google News application anymore. As of April 2024, Google stopped accepting new publications in Publisher Center and moved to algorithmically surfacing any site that follows its news content policies. Eligibility for the News tab and Top Stories is now driven by producing policy-compliant, high-quality news — being in Publisher Center neither guarantees inclusion nor is required for it. What still drives ranking within that pool is relevance, prominence, authoritativeness, freshness, and location.

Full walkthrough — policies, ranking factors, and the Publisher Center transition timeline — in Google News SEO.

Google Discover: a feed you can’t target with keywords

Discover isn’t search. Google pushes it to users based on interests, so there’s no query to rank for and no keyword to target. Eligibility is driven by content quality, content freshness, strong unique imagery (large, high-quality images that Google is licensed to display large), Core Web Vitals, and mobile experience. Discover traffic is famously spiky and non-recurring, so it’s a supplement to search demand, not a substitute for it.

Full walkthrough — eligibility levers, image specs, and the Discover-specific core update — in Google Discover SEO.

Paywalled content: flexible sampling without cloaking

Showing Googlebot the full text of an article that readers must pay for would normally be cloaking. Google carves out an exception called flexible sampling, but only if you declare the paywall in structured data so it can tell a legitimate subscription wall from a spam one. Set isAccessibleForFree to false on the article and point a cssSelector at the wrapper class around the gated section:

{
  "@context": "https://schema.org",
  "@type": "NewsArticle",
  "isAccessibleForFree": false,
  "hasPart": {
    "@type": "WebPageElement",
    "isAccessibleForFree": false,
    "cssSelector": ".paywall"
  }
}

The cssSelector must be a class selector (.paywall) — IDs and other selector types aren’t accepted — and the class has to wrap exactly the paywalled markup. Done right, the gated article is indexed on its full content without being treated as cloaking.

Full walkthrough — the JavaScript-paywall trap and metering strategy — in Paywalls and SEO.

Syndication: control competing copies with noindex, not canonical alone

When the same article runs on partner sites, Google may choose either version for search — canonical is a signal it can override, not a directive, so it’s not a reliable way to force one copy out of contention. If a partner copy must not compete, require noindex on that copy; Google’s current guidance recommends noindex over canonical for syndicated content specifically, because syndicated pages often differ enough from the original that Google won’t honor the canonical anyway. That means syndication needs a governance step as much as a technical one: agree the indexing treatment (canonical, noindex, or neither) with each partner in the syndication agreement before the piece runs, not after a partner copy has already started outranking you.

Full walkthrough — all three options, Bing’s different preference, and the canonicalization mechanics — in Syndicated content and SEO.

News sitemaps and the freshness window

A news sitemap is not a smaller XML sitemap. It’s a separate news:-namespaced file that should contain only articles from roughly the last two days, with articles removed from it once they age out. Its job is fast discovery of breaking content, so stale URLs in it work against you. Standard XML sitemaps still cover the rest of the site.

Full walkthrough — the exact news: tags, the 1,000-URL cap, and dead tags to stop copying — in News sitemaps.

Measure each surface separately

Don’t roll Search, News and Discover numbers into one traffic figure — they’re different reports measuring different things, and blending them hides which surface actually moved. Search Console’s Performance report covers standard Search results; Discover gets its own filter inside that same report, which only appears once a site has recorded enough Discover impressions. There’s no comparable per-query performance report for the News tab or Top Stories itself, so News visibility is mostly judged by referral traffic and Search Console’s general indexing/coverage data rather than a dedicated dashboard. Third-party keyword-volume estimates are useful for prioritizing which surfaces and topics to invest in, but they’re not a substitute for these first-party reports, and a spike or drop in one surface’s report doesn’t imply anything about the others.

Speakable schema is still a limited beta

Speakable structured data marks the sections of an article best suited to text-to-speech playback. It remains a beta limited to news content and to Google Assistant / Google Home devices in US English. It’s worth adding if you publish English-language US news and want that surface, but not something to over-invest in given the narrow, still-changing support.

News/publisher SEO adds publisher-specific surfaces on top of normal SEO: Google News / Top Stories (freshness, originality, trust), Google Discover (a pushed feed, not query-driven — driven by quality, freshness, imagery, Core Web Vitals, mobile), news sitemaps (fast discovery of recent articles), paywall handling (show crawlers gated content via the supported flexible-sampling/structured-data approach so it isn’t cloaking), and syndicated content (noindex on partner copies that should not compete, agreed with the partner in advance rather than left to canonical alone). Eligibility on one surface never transfers to another, and each is measured separately in Search Console rather than as one blended traffic number. This hub links to each deep dive: Google News, Google Discover, paywalls, syndicated content, and news sitemaps.

Changelog