AI Search
Google today launched its February 2026 Discover core update for English users in the US, promising to reduce clickbait while promoting locally relevant and expert content.
Google today released the February 2026 Discover core update, marking the first algorithm adjustment of the year specifically targeting the personalized content feed that has become the dominant traffic source for news publishers. The update introduces three major changes: prioritizing locally relevant content from domestic websites, reducing sensational and clickbait material, and surfacing deeper expertise from specialized sections within broader publications.
According to the announcement posted to the Google Search Central Blog on Thursday, February 5, 2026, the company’s internal testing demonstrates that users find the Discover experience “more useful and worthwhile with this update.” The changes apply exclusively to English language users in the United States at launch, with expansion to all countries and languages planned for the coming months.
The timing carries particular significance for publishers still recovering from December’s devastating core update. Google’s December 2025 core update triggered severe Discover traffic collapse within 48 hours, with some publishers reporting complete elimination of impressions after years of stable performance. Multiple website operators documented 70-85% declines in daily visitor counts during what should have been the most lucrative advertising period of the year.
The February update represents a shift in Google’s approach to content curation within Discover. Rather than modifying the underlying core ranking systems that affect traditional search results, this adjustment specifically targets the algorithmic recommendation mechanisms that surface articles in the mobile feed. John Mueller, Search Advocate at Google, wrote in the announcement that the update “will improve the experience in a few key ways.”
The first major component emphasizes locally relevant content from websites based in the user’s country. This geographical prioritization could fundamentally alter traffic patterns for international publishers who have built audiences across multiple markets. A news organization based in the United Kingdom covering United States politics, for instance, may find its content deprioritized for American Discover users in favor of domestic sources covering the same topics.
Google did not specify the technical mechanisms determining “locally relevant content.” The company’s systems must evaluate factors including domain registration location, hosting infrastructure geography, editorial headquarters, or byline attribution to implement this geographic preference. Publishers operating international bureaus with local reporting teams face uncertainty about whether their content qualifies as locally relevant despite foreign corporate ownership.
The local content emphasis arrives as research published in August 2025 revealed Google Discover accounts for two-thirds of referrals to news and media websites, having overtaken traditional search as the primary traffic source. This concentration means algorithmic adjustments to Discover distribution carry greater impact than equivalent changes to conventional search rankings.
Research from Chartbeat spanning July 2023 to April 2025 documented Google Discover traffic steadily increasing while traditional search referrals declined. The inflection point occurred in late October 2024, coinciding with Google’s AI Overviews rollout across more than 100 countries. Traditional Google Search traffic dropped from approximately 16% to 10% of total referrals during this period, making publishers increasingly dependent on Discover’s algorithmic curation.
The second component addresses sensational content and clickbait in Discover, though Google provided no technical definition of these terms or specific examples of prohibited practices. The announcement states simply that the update focuses on “reducing sensational content and clickbait in Discover” without elaborating on the distinction between legitimate curiosity-generating headlines and manipulative practices the algorithm now penalizes.
Publishers employing curiosity gap headlines – a longstanding journalistic technique that creates interest without revealing full story details – face uncertainty about whether their editorial practices now qualify as clickbait under Google’s undefined standard. A headline reading “What happened next shocked everyone” clearly employs clickbait tactics. Whether “Scientists discover unexpected solution to long-standing problem” crosses Google’s threshold remains unclear without specific guidance.
The clickbait targeting creates potential conflicts with established headline writing practices. News organizations routinely employ attention-capturing language to compete for reader attention in crowded information environments. The Financial Times, The Guardian, and other respected publications frequently use provocative headlines on legitimate news coverage. Google’s algorithm must distinguish between manipulative engagement tactics and effective headline writing – a subjective judgment that machine learning systems struggle to execute consistently.
Industry data demonstrates the stakes involved in Discover visibility. Analysis from Marfeel published in December 2025revealed that 51% of feed positions in the United States, Brazil, and Mexico now consist of AI Summaries rather than traditional publisher links. This AI content integration squeezes available publisher distribution even before accounting for clickbait penalties, creating intensified competition for remaining Discover placements.
The third major change introduces topic-specific expertise evaluation within publications that cover multiple subject areas. According to the announcement, Google’s systems now identify expertise “on a topic-by-topic basis” rather than evaluating entire domains uniformly. This granular assessment means a local news site with a dedicated gardening section could demonstrate expertise in gardening despite covering numerous other topics, while a movie review site publishing a single gardening article would not.
The expertise evaluation methodology represents a significant technical challenge for algorithmic systems. Google must determine whether specific website sections demonstrate genuine expertise through analysis of content depth, author credentials, publication frequency, source attribution, or other signals. The announcement provides no details about these technical mechanisms beyond stating that systems assess sites’ content to identify expertise.
This sectional expertise recognition could benefit regional publications with specialized beats maintained by dedicated reporters. A metropolitan newspaper employing a transportation reporter covering infrastructure developments for years might gain Discover visibility for those specific articles despite competing against national outlets for broader news coverage. Conversely, general interest publications producing occasional coverage across diverse topics without sustained expertise may find their Discover distribution reduced.
The expertise framework builds upon Google’s broader emphasis on E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness) evaluation across its search products. The June 2025 core update demonstrated partial recoveries for websites previously impacted by the September 2023 Helpful Content Update, with analysis from SEO expert Marie Haynes identifying comprehensive, expert content as the common characteristic among improved sites.
Google’s decision to launch exclusively for English language users in the United States creates temporary geographic market fragmentation. Publishers serving international audiences cannot predict when their markets receive the update or whether implementation details will vary across languages and regions. The phased rollout extending “in the months ahead” provides no specific timeline for global deployment.
This staggered implementation pattern mirrors Google’s October 2025 Discover feature launches, which introduced AI-powered brief previews in the United States, South Korea, and India before broader expansion. The selective deployment enables Google to monitor performance and adjust systems before affecting global publisher traffic, though it creates strategic challenges for content operations spanning multiple markets.
The United States-first approach reflects market priorities rather than technical constraints. English language content represents a substantial portion of Discover’s total distribution, and American publishers command significant advertising revenue. Successful implementation in this market provides proof of concept for subsequent international rollouts while minimizing disruption if adjustments prove necessary.
Publishers must adapt content strategies without complete information about implementation specifics or expansion timelines. Organizations operating across multiple countries face decisions about whether to modify editorial practices based on United States market changes that may not translate to other regions when the update eventually reaches those markets.
Google’s announcement provides no data quantifying expected traffic changes for affected publishers. The company stated that “this change may lead to fluctuations in Discover traffic” with some sites potentially seeing increases or decreases while many experience no change at all. This vague guidance offers publishers minimal insight for planning or performance forecasting.
The traffic uncertainty arrives as publishers navigate persistent volatility from previous algorithm adjustments. Google’s December 2025 core update required 18 days to complete after launching December 11, exceeding the typical 14-16 day implementation period observed in recent updates. The extended rollout created consecutive weekend disruptions on December 13 and December 20, with some publishers reporting traffic declines ranging from 70-85%.
Historical patterns suggest significant traffic shifts typically accompany major Discover algorithm modifications. The March 2025 core update lasted 14 days and concluded on March 27, 2025, creating what SEO experts described as “wild swings” in rankings. Glenn Gabe, a prominent algorithm analyst, noted that many sites experienced core update-like impact as early as March 6, roughly a week before Google’s official announcement.
Recovery prospects remain uncertain for publishers negatively affected by Discover algorithm changes. Google’s guidance consistently emphasizes that meaningful ranking improvements often require subsequent core update cycles rather than immediate content modifications. Historical data from previous updates shows limited recovery patterns even months after implementation, with sites in algorithmic “gray areas” facing continued volatility.
Google directed publishers seeking guidance to existing documentation about core updates and the “Get on Discover” help page. The company provided no new recommendations specific to the February 2026 update beyond these established resources, suggesting that fundamental content quality principles remain unchanged despite the algorithmic adjustment.
The existing core update guidance emphasizes creating helpful, reliable, people-first content without attempting to optimize for specific ranking factors. According to Google’s standard messaging, content creators focusing on these principles “will largely see updates come and go without notice, or they might find themselves performing better in search.” This advice offers minimal actionable guidance for publishers attempting to understand or respond to specific algorithm changes.
The “Get on Discover” documentation covers technical requirements for Discover eligibility including content policies, quality guidelines, and image specifications. Publishers must provide high-quality images, avoid manipulative practices, ensure content meets Google News policies, and maintain adequate page experience metrics. However, meeting these baseline requirements does not guarantee Discover distribution or protect against negative impacts from algorithm updates.
The limited guidance reflects Google’s consistent position that publishers should focus on content quality rather than algorithmic optimization. This philosophy creates challenges for media organizations attempting to understand traffic fluctuations or adjust strategies in response to specific update components like local content prioritization or clickbait reduction.
Google emphasized that the update maintains content personalization based on users’ creator and source preferences. According to the announcement, Discover “will continue to show content that’s personalized based on people’s creator and source preferences” alongside the new algorithmic adjustments. This statement suggests the update modifies content selection within personalized feeds rather than replacing personalization mechanisms entirely.
The personalization preservation creates complex interactions between user preferences and new algorithmic filters. A user who previously followed international publishers covering domestic news topics may find those sources deprioritized by local content emphasis regardless of explicit preference signals. The balance between user-expressed preferences and algorithmic content policies remains undefined in Google’s announcement.
Google introduced follow functionality for publishers and creators in September 2025, enabling users to subscribe directly within Discover. That update integrated content from X, Instagram, and YouTube Shorts alongside traditional web articles, creating a consolidated feed across multiple platforms. The February update’s local content emphasis and clickbait reduction apply to this expanded content universe including social media posts and video content.
The personalization approach distinguishes Discover from traditional search results, where explicit user queries determine content relevance. Discover’s algorithmic curation combines signals including browsing history, search activity, app usage, location data, and explicit follows to predict content interest. The February update modifies this prediction system to penalize certain content characteristics while favoring others, potentially overriding signals suggesting user interest in filtered material.
The update creates competitive advantages for specific publisher types while disadvantaging others. Local news organizations with established expertise in community coverage gain algorithmic preference over national outlets covering the same geographic areas. Publications maintaining dedicated subject matter sections staffed by specialist reporters benefit from topic-specific expertise recognition versus general interest sites producing occasional coverage.
International publishers face strategic decisions about content distribution and market targeting. Organizations operating bureaus in multiple countries must evaluate whether establishing distinct domains for different markets preserves Discover visibility better than maintaining unified international operations. The local content preference incentivizes geographic fragmentation of publishing operations that may conflict with editorial integration strategies.
Smaller specialized publications could gain relative advantage against larger generalist competitors through the sectional expertise evaluation. A focused technology analysis site demonstrating consistent depth across specific topics may outperform major news organizations producing broader but less detailed coverage. This dynamic could partially offset the brand authority advantages that have increasingly dominated search results since the September 2023 Helpful Content Update.
The clickbait reduction component forces editorial reconsideration of headline practices and content presentation. Publishers employing curiosity-generating techniques must determine whether their approach falls within Google’s undefined boundaries or risks algorithmic penalties. The uncertainty creates compliance challenges without clear violation examples or appeals mechanisms for wrongly penalized content.
Google provided no technical documentation explaining how its systems evaluate local relevance, identify clickbait, or assess topic-specific expertise. The announcement describes intended outcomes – more local content, less clickbait, deeper expertise – without revealing implementation mechanisms publishers could reference for optimization or compliance verification.
The local relevance determination likely involves multiple signals including domain registration data, hosting infrastructure location, editorial office addresses, byline attribution, content focus, and audience geography. However, the relative weighting of these factors and minimum thresholds for local classification remain undisclosed. Publishers cannot verify whether their content qualifies as locally relevant for specific markets without observing traffic patterns post-implementation.
Clickbait identification presents substantial technical challenges for automated systems. Machine learning models must distinguish manipulative engagement tactics from legitimate curiosity-generating headlines while accounting for cultural context, subject matter norms, and editorial voice. The subjective nature of clickbait evaluation creates potential for inconsistent enforcement and false positives affecting quality journalism.
Expertise assessment requires analyzing content characteristics including depth, source attribution, author credentials, publication consistency, and external recognition. Google’s systems must evaluate these signals at the section or topic level rather than domain-wide, creating computational complexity for sites covering numerous subjects. The granular analysis enables specialized section recognition but increases potential for incorrect categorization.
The Discover update arrives amid intensifying tensions between platforms and publishers over content distribution and monetization. Data from Marfeel revealed that AI Summaries now occupy 51% of Discover feed positions in test markets, with YouTube absorbing major shares of default exits. This platform prioritization reduces available distribution for traditional publisher links before accounting for algorithmic filtering.
The concentration pattern extends beyond Discover. Analysis of over 400 news publishers worldwide by NewzDash confirmed that Google Web Search traffic declined from 51% to 27% between 2023 and the fourth quarter of 2025, while Discover feed climbed to 68%. This fundamental redistribution means publishers now depend primarily on Discover’s personalized recommendation algorithms rather than traditional search optimization.
The dependency creates asymmetric power relationships between Google and content creators. Publishers must produce content suited to recommendation algorithms while having minimal influence over distribution decisions. This contrasts with traditional search, where content quality and relevance could improve rankings through established optimization practices documented in public guidance.
Samsung’s One UI 7 update created additional distribution challenges in April 2025 by defaulting Galaxy devices to Samsung News instead of Google Discover. The change affected millions of Samsung users, requiring multi-step reconfiguration to restore Google Discover access. This platform competition demonstrates the fragility of publisher distribution channels dependent on default settings controlled by hardware manufacturers.
Google’s algorithm update frequency throughout 2025 established patterns likely continuing into 2026. The company deployed three core updates during 2025 – in March, June, and December – alongside continuous smaller refinements and specialized adjustments like the February Discover update. This deployment cadence suggests publishers should expect persistent ranking volatility rather than extended stability periods.
The March 2025 core update began March 13 and required 14 days to complete. The June 2025 update launched June 30 and needed 16 days for implementation. The December update started December 11 and concluded after 18 days on December 29. These timelines demonstrate substantial deployment periods requiring ongoing monitoring rather than single-day adjustments.
Between major core updates, Google implements continuous algorithmic refinements affecting search results. Documentation updated December 9, 2025 – two days before the December core update announcement – clarified that ranking improvements can occur without waiting for confirmed updates. Google stated that “we’re continually making updates to our search algorithms, including smaller core updates. These updates are not announced because they aren’t widely noticeable.”
The continuous update approach creates persistent uncertainty for publishers attempting to attribute traffic changes to specific algorithm modifications versus natural fluctuation. The February Discover update represents an announced specialized adjustment, but undisclosed refinements to other systems continue simultaneously, complicating performance analysis and strategic planning.
Publishers face substantial challenges measuring the February update’s specific impact on their traffic. Discover traffic appears in Google Analytics and Search Console alongside traditional search referrals, but distinguishing update effects from normal volatility requires careful analysis of timing, content type patterns, and geographic segmentation.
The United States-only initial deployment enables some geographic isolation for impact measurement. Publishers serving multiple markets can compare United States Discover traffic changes against other regions not yet receiving the update. However, this analysis requires sufficient traffic volume in each market and assumes comparable content consumption patterns across geographies.
The local content preference, clickbait reduction, and expertise evaluation operate simultaneously, making attribution of specific traffic changes to individual update components nearly impossible. A publisher experiencing Discover traffic decline cannot determine whether the reduction stems from geographic deprioritization, perceived clickbait characteristics, insufficient demonstrated expertise, or combination effects.
External tracking tools monitoring search volatility lack visibility into Discover’s algorithmic mechanisms. Services including Semrush, Mozcast, and Sistrix measure traditional search ranking fluctuations but cannot detect Discover-specific changes with equivalent precision. Publishers must rely primarily on their own analytics data for update impact assessment.
Google’s announcement that the update “will expand it to all countries and languages in the months ahead” provides minimal timeline specificity. The phased rollout could extend weeks or months depending on monitoring requirements, adjustment needs, and engineering resources. Historical patterns from previous feature launches suggest three to six month windows for full global deployment.
The geographic expansion creates ongoing uncertainty for international publishers. Organizations must monitor United States market performance for indicators about likely impacts in their regions while recognizing that implementation details may vary across languages and cultural contexts. A clickbait definition calibrated for American content consumption patterns may require adjustment for European or Asian markets.
Language-specific rollouts introduce additional complexity beyond geographic targeting. Content evaluation systems trained primarily on English language material must adapt to linguistic nuances, cultural norms, and editorial conventions varying across languages. The expertise assessment proving effective for English content may require recalibration for publications in German, Japanese, or Arabic.
The multi-month expansion timeline means publishers will operate in heterogeneous environments where some markets receive updated algorithms while others maintain previous systems. This fragmentation complicates content strategy, editorial standards, and performance benchmarking across international operations.
Who: Google implemented the February 2026 Discover core update affecting how content appears in the personalized mobile feed. The update impacts publishers, content creators, and users accessing Discover on mobile devices and the Google app.
What: The algorithmic adjustment introduces three major changes to Discover’s content curation: prioritizing locally relevant content from websites based in users’ countries, reducing sensational content and clickbait, and identifying topic-specific expertise within publications covering multiple subjects. Google stated its testing shows users find the experience “more useful and worthwhile with this update.”
When: Google released the update on Thursday, February 5, 2026. The deployment began immediately for English language users in the United States, with expansion to all countries and languages planned for the months ahead. No specific timeline was provided for global rollout completion.
Where: The initial implementation targets exclusively English language users in the United States. The geographic and linguistic restriction creates temporary market fragmentation as publishers serving international audiences cannot predict when their regions receive the update or whether implementation details will vary across markets.
Why: The update responds to user experience concerns about content quality and relevance within Discover. Google aims to surface more locally relevant information, reduce manipulative engagement tactics, and better identify genuine expertise across diverse publication types. The changes arrive as Discover has become the dominant traffic source for news publishers, accounting for two-thirds of Google referrals according to August 2025 research.
Your go-to source for digital marketing news. Get the latest updates from Google, Meta, Amazon, and The Trade Desk. Stay informed on ad tech innovations, programmatic trends, and policy changes.
