Google February 2026 Discover Core Update Begins Rolling Out
On February 5, 2026 at 9:00 a.m. Pacific, Google launched what many publishers have been waiting for: a Discover-specific core update. This isn’t a broad Search core update. Google described it as a “broad update to our systems that surface articles in Discover.” The company announced the rollout on the Search Central Blog and logged it as an active incident on the Google Search Status Dashboard, with an estimated completion window of up to two weeks.
If Discover drives a meaningful share of your traffic (or your clients’), treat this as its own event—because it is. The early signal across data and community reporting is fairly consistent: Google is tightening what “quality” means in Discover, with an apparent preference for local relevance, real expertise, and original depth—and less tolerance for templated engagement tricks.
What makes this update different
This February 2026 change is widely seen as Google’s first core update aimed specifically at Discover, not traditional organic search rankings. That matters because Discover behaves differently:
- Discover is feed-based and personalized, so volatility can look spiky and inconsistent—even across similar sites.
- Search Console’s Discover report is your main source of truth, since rank tracking doesn’t map cleanly to feed visibility.
- The update changes what gets surfaced in Discover. It’s not framed as a “penalty,” but the traffic impact can feel the same when visibility drops overnight.
Google’s stated goals for the update center on three themes:
- More locally relevant content from country-based sites
- Less sensational/clickbait content
- More in-depth, original, timely content from sites that demonstrate topic-level expertise
That third point is the one to take seriously. “Topic-level” evaluation suggests you don’t need to be great at everything. You do need to earn trust and attention in the subjects you cover (for example, a local paper’s gardening coverage beating broader lifestyle content that’s thinner and more generic).
Timeline: what we know so far
Here’s how the rollout has played out publicly:
- Feb 5: Google announces the update and begins rollout, initially aimed at English-language users in the U.S.
- Feb 5–10: Heavy turbulence reported across forums (WebmasterWorld, Reddit) and social, including publishers describing 90–95% Discover drops overnight (often alongside broader swings across news/image surfaces).
- Around Feb 11–19: The initial U.S. rollout window appears to close based on stabilization patterns and tracking, while Google’s dashboard language remains broad. Google also signaled expansion to other countries/languages “in the months ahead.”
- Feb 20–23: Analyses (notably from NewzDash) start showing clearer patterns—especially more concentration and fewer domains appearing in top Discover visibility sets.
- Feb 24–25: Conversation shifts from “it’s still rolling out” to “it’s still uneven.” Volatility looks more stable but remains choppy, which matches Discover’s normal behavior—even without a major update.
Glenn Gabe captured the core challenge: Discover is already prone to surges and drops; an update can amplify that baseline instability and make attribution messy.
Early impact: volatility, consolidation, and “local-first” signals
The most useful early dataset comes from NewzDash, which analyzed visibility shifts using panel data across thousands of domains. A few patterns stand out.
1) Fewer domains are getting a seat at the table
In the U.S. Top 1000 Discover visibility set, unique domains reportedly fell from 172 to 158. That’s not just a stat—it’s fewer distinct publishers capturing most of the attention.
At the same time, the top 10 publishers’ share increased, pointing to greater concentration at the top.
2) Local relevance looks boosted
NewzDash also highlighted meaningful shifts in localized visibility (for example, California-local Top 100 content rising sharply). That lines up with Google’s stated direction: a Discover feed that feels more regionally useful to the person scrolling.
3) Category mix is shifting
Early scorecards suggest:
- News and Sports gained share
- Arts/Entertainment lost share
This doesn’t mean entertainment can’t perform. It does suggest that certain entertainment formats—especially commoditized aggregation—may be losing ground to coverage that feels immediate, reported, and grounded.
Winners and losers (and what that implies)
No third-party “winners and losers” list is definitive. Discover is personalized, and visibility isn’t evenly distributed. Still, the early patterns are hard to ignore.
Publishers that appear to be winning
- Major U.S. news brands (examples often cited include CBS News, NBC News, AP, Forbes)
- A noticeable bump in X.com visibility in some datasets
- International brands that remain highly relevant to U.S. readers (BBC is frequently cited as an exception)
NewzDash also noted an increase in editorial articles and a shift toward longer titles. That runs against the “short titles always win” assumption, but it tracks if Google is leaning toward clarity and context over pure curiosity.
Publishers/content types that appear to be losing
- Some international outlets showing significant drops in U.S. visibility (The Guardian, Reuters, Independent are frequently referenced)
- Heavily templated, commoditized content formats
- Overt clickbait patterns, including heavy reliance on question-style titles
John Shehata’s framing is the right mental model: the message isn’t “don’t write engaging headlines.” It’s “stop relying on low-value templates and curiosity-gap distribution.” Engagement still matters; manufactured engagement signals seem to matter less.
What I’m advising teams to do right now (without panicking)
Discover is uniquely stressful because traffic can drop fast—and rebound just as fast. While things settle, you can still make smart, low-regret moves.
1) Separate Discover diagnosis from Search diagnosis
A Discover drop is not automatically a sitewide quality problem. Start here:
- Search Console → Discover performance report
- Segment by page, topic, and date ranges around Feb 5 onward
- Look for clusters: specific templates, headline patterns, or content types that slipped
2) Audit headlines for clarity and value (not just CTR)
If your titles lean on:
- vague curiosity (“You won’t believe…”, “This is why…”, “What happened next…”)
- excessive question formats
- repeated, templated phrasing across many pages
Start rewriting there. The goal isn’t to be less compelling—it’s to make the reader benefit obvious.
3) Build topic clusters that prove depth
Jason Davis’s “depth beats breadth” observation matches what this update appears to reward. Discover increasingly seems to favor publishers that earn repeat visibility in a topic, not those that pop up once with a viral post.
- Strengthen internal linking between related articles
- Be consistent with author/topic coverage
- Publish work that shows lived experience, original reporting, and original analysis (E-E-A-T in practice, not as a checkbox)
4) Re-check Page Experience basics
Discover has always leaned toward good user experience. Post-update, Core Web Vitals and mobile UX feel even more like baseline requirements.
- Ensure mobile layout stability
- Avoid intrusive interstitials
- Keep performance clean on top Discover landing pages
5) Diversify discovery channels
Relying on Discover as the primary growth lever is risky. If this update increases concentration among top brands, smaller publishers should treat that as a cue to build resilience through newsletters, direct traffic, partnerships, and search growth that isn’t feed-dependent.
What to watch next
Two things are worth tracking closely:
- Global expansion: Google says this begins with U.S. English and expands to other countries/languages in the months ahead. If you operate outside the U.S., use the time now to run content and UX audits before the rollout reaches your market.
- Publisher consolidation: If fewer domains continue to dominate the top sets, Discover may become increasingly “winner-takes-more.” Brand authority, topical consistency, and repeat user satisfaction will matter even more.
Conclusion: my bottom line on the February 2026 Discover update
Google’s February 2026 Discover Core Update is a clear signal that Discover is now its own optimization discipline—with its own cadence, thresholds, and distribution logic. The sites that win won’t be the ones that publish the most. They’ll be the ones that publish with clear reader value, strong topical credibility, better regional relevance, and a genuinely solid on-page experience.
If you’re trying to make sense of the volatility, focus on clean diagnostics and turn the signals into a repeatable editorial system. If you want a platform that helps operationalize content quality and authority-building instead of guessing, AIuthority helps teams plan, evaluate, and scale expert-led content that holds up when Discover’s rules shift.