Google Discover's February 2026 shake up: which publishers quietly surged, which slipped

Google Discover update February 2026: early performance data

Executive snapshot

• In NewzDash's US Discover panel, unique publishers in the top 1,000 placements fell from 172 to 158 domains (-8 percent) after the February 2026 update, while unique content categories increased.[S1]
• In California, local articles in the top 100 positions rose from 10 to 16 posts (+60 percent), with state-focused outlets (for example, SFGate, LA Times) appearing in local rankings but not in the concurrent US-wide top 100.[S1]
• New York-focused publishers appeared about 5 times more often in the New York Discover feed than in California, and vice versa, indicating a measurable local layer on top of a shared national feed.[S1]
• Yahoo's representation in the US top 1,000 Discover articles fell from 11 to 6, with no Yahoo items in the US top 100 after the update, aligning with Google's stated push against "clickbait-style" content.[S1][S2]
• X.com content from institutional and media accounts increased from 3 to 13 items in the US top 100, and from 2 to 14 in New York's top 100, continuing a growth trend first observed in late 2025.[S1][S3]

Implication for marketers: Early 2026 Discover data points toward more local and topical depth but tighter domain concentration, favoring publishers with clear subject authority and regionally aligned news coverage.

Method and source notes

NewzDash DiscoverPulse analysis (primary data)

News SEO vendor NewzDash published an analysis based on its DiscoverPulse panel.[S1]

• What was measured: Frequency and ranking of domains, articles, and content categories in Google Discover feeds.
• When: Comparison of a pre-update window (January 25-31, 2026) versus a post-update window (February 8-14, 2026), during the rollout of Google's Discover-focused February 2026 core update.[S1][S2]
• Where: US-based Discover panel, with specific views for:
  • US-wide feed
  • California feed
  • New York feed[S1]
• How: A panel of "millions of US users" monitored via DiscoverPulse, aggregating which URLs and domains appear in top placements and their categories.[S1] NewzDash derived top-1,000 domains and top-1,000 articles for each geography and timeframe, then compared overlap, local skew, and category spread.[S1]

Google communications (policy intent and qualitative guidance)

• Google's announcement of the February 2026 core update with a Discover focus and revised Discover documentation.[S2]
• Stated goals: increase locally relevant content, reduce clickbait and sensational headlines, and emphasize in-depth and timely content from sites with topical expertise.[S2]

Context sources

• NewzDash prior analysis of X.com's emerging presence in Discover in 2025.[S3]
• Early analysis of the December 2025 core update suggesting brands and specialist sites gaining on generalist publishers.[S4]

Key limitations

• The vendor is commercially invested in Discover monitoring; data is proprietary and not independently audited.[S1]
• Observation windows are short and coincide with major events (Super Bowl, Winter Olympics, ICC Men's T20 World Cup), which can skew News and Sports category mix.[S1]
• The panel is US-only and English-language; Google has indicated a phased rollout to more locales and languages later in 2026.[S2]
• The analysis is descriptive; effects on traffic, click-through rate, or revenue are not directly measured.[S1]

Google Discover visibility and domain concentration after the February 2026 update

NewzDash's panel data suggests a notable shift in how Discover distributes visibility across domains in the US after the February 2026 update. While the number of unique content categories appearing in top placements increased across the US, California, and New York views, the number of unique publishers decreased in two of the three.[S1]

• US feed (top 1,000 articles):
  • Unique publishers fell from 172 pre-update to 158 post-update (-8.1 percent).[S1]
  • Unique content categories increased, described as a broader topical spread (exact counts not disclosed).[S1]
• California feed (top 1,000 articles):
  • Unique publishers fell from 187 to 177 (-5.3 percent).[S1]
  • Content categories again expanded.[S1]
• New York feed:
  • Unique publishers were relatively stable, with a smaller contraction than in the US and California views; categories still broadened.[S1]

Overall, Discover appears to cover more topic areas while routing that visibility to a somewhat smaller set of domains in key US views. The pattern mirrors earlier observations from the December 2025 core update, where specialist and brand-strong sites captured more share for "best of" and product-driven queries at the expense of generalist domains.[S4]

Because these shifts are measured as share of feed presence, they do not directly report on impressions or traffic. However, reduced domain diversity in a capped feed format usually implies that smaller or mid-tier publishers lose relative exposure while those aligned with Google's quality and relevance signals gain proportionally.

Local content and regional publishers in Google Discover feeds

Dataset comparisons across US-wide, California, and New York Discover feeds show a clear local overlay added on top of a national baseline. NewzDash reports that while most of the feed inventory overlapped among the three views, each state-level feed contained local stories and publishers that did not appear in the others.[S1]

Quantitatively, state-specific publishers saw a strong skew in their home state's feed:[S1]

• New York-local domains appeared about 5 times more often in New York's Discover feed than in California's.
• California-local domains appeared about 5 times more often in California's feed than in New York's.

In the California top 100 Discover positions, local articles increased from 10 to 16 between the pre- and post-update windows (+60 percent).[S1] Outlets such as SFGate and the LA Times appeared in the California top 100 but did not show up in the concurrent national US top-100 list in the NewzDash dataset, underscoring that local prominence is sometimes constrained to state-level views.[S1]

This aligns with Google's stated goal for the February 2026 core update to raise locally relevant content in Discover, especially for users in specific regions.[S2] It is too early to know whether this local amplification will persist once the heavy news and sports cycle moderates, but the effect is already large enough (five-fold skews) to be material for regional newsrooms and local brands.

Clickbait reduction and topic-expert content in the Discover feed

One of Google's documented aims for the February 2026 update was to reduce the prominence of clickbait-style and sensational content in Discover while boosting timely, in-depth coverage from sites with clear topic specialism.[S2]

NewzDash attempted to quantify this by analyzing headline patterns it classifies as "templated curiosity-gap". It reported that such patterns lost visibility in post-update top lists, while stressing that headline pattern detection is not a definitive proxy for clickbait.[S1]

A more tangible data point comes from Yahoo's performance:[S1]

• In the US top 1,000 Discover articles, Yahoo's presence declined from 11 articles pre-update to 6 post-update (-45.5 percent).
• Yahoo had items in the US top 100 pre-update but had zero articles in the top 100 in the post-update window.

NewzDash positions this as consistent with a shift away from generalized, often templated or highly aggregated coverage toward more specialized outlets.[S1][S4] However, the analysis does not label Yahoo's content as clickbait; the causal link between Yahoo's decline and clickbait suppression remains unproven.

More broadly, the increase in unique categories combined with domain consolidation suggests that Google may be rewarding publishers who demonstrate authority across defined topical clusters, then using them to cover a wider range of related subjects.[S1][S4] Without page-level expertise metrics from Google, this remains inferred from output patterns rather than a confirmed mechanism.

X.com content growth in Google Discover feeds

NewzDash's February 2026 scorecard highlights a sizable jump in X.com content exposure in Discover, continuing a trend first tracked in late 2025.[S1][S3] The dataset focuses on X posts from institutional and media accounts that lead back to publisher content or hosted media.

Observed changes in the top 100 placements:[S1]

• US feed: X.com items increased from 3 to 13 (+333.3 percent).
• New York feed: X.com items increased from 2 to 14 (+600 percent).

In a separate November 2025 study, NewzDash tracked X.com's Discover growth, describing X as "quietly becoming a Discover powerhouse" with increasing inclusion for posts tied to major news events and live conversations.[S3] The February 2026 data indicates that the Discover-focused core update did not dampen this pattern and may have accelerated it.[S1]

The analysis is explicit that it cannot determine whether X posts are substituting for or complementing direct publisher links in Discover.[S1] Because X posts introduce an extra click layer between Discover and the underlying article (when that article is off X), net traffic to owned properties can rise or fall depending on user behavior, a factor not measured in the current panel.

Overall, the data shows that X, as a content endpoint, is now present in a material share of high-ranking Discover slots around news and timely topics in the US and especially in New York.[S1][S3]

Interpretation and implications for SEO and content strategy

Domain consolidation likely favors clear topical authority

• Given the drop in unique publishers and expansion in categories, it is likely that Discover is relying more heavily on a smaller pool of sites perceived as authoritative per topic.[S1][S4]
• For publishers, this suggests focusing on distinct subject areas, depth of coverage, and consistent quality signals (E-E-A-T factors) rather than broad, shallow coverage.

Regional relevance is more important for news and local brands

• The five-fold increase in visibility for local domains within their home state feeds, plus the California top-100 local rise, indicates that precise regional relevance is increasingly rewarded.[S1]
• Local publishers and multi-location brands may benefit from state- or city-specific content strategies that align with local news, events, and needs rather than generic national templates.

Headline strategies should avoid curiosity-gap templates

• While NewzDash's detection of templated curiosity-gap patterns is an approximation, the reported drop in such headlines within top lists is consistent with Google's policy language.[S1][S2]
• Headline approaches that clearly describe the substance and outcome of a story, instead of withholding key information to generate clicks, are likely safer in Discover.

X.com distribution as a secondary Discover access point

• The sharp growth in X.com top-100 placements suggests that active, verified institutional accounts sharing timely content may gain additional Discover exposure via embedded X posts.[S1][S3]
• For publishers already investing in X, structured posting (clear linking, descriptive captions, and media usage) can increase the chance that these posts surface in Discover, although the traffic tradeoff between direct links and social embeds is not yet quantified.

Generalist portals may face ongoing share erosion

• Yahoo's relative drop, combined with December 2025 findings that generalist portals lost share on "best of" and product queries, hints at a medium-term rebalancing away from broad portals.[S1][S4]
• If this continues, non-specialist sites may need to segment and strengthen distinct vertical sections with dedicated editorial and UX patterns aligned to those niches.

Contradictions and gaps

Clickbait measurement

• Google's guidelines talk about clickbait-style reductions, but there is no shared, quantitative definition.[S2]
• NewzDash uses headline templates as a heuristic; this is informative but not definitive. A site could have sensational content with plain headlines or vice versa.[S1]

Traffic and revenue effects

• The panel measures share of feed inventory, not impressions, sessions, or revenue per visit.[S1]
• Publishers report varied anecdotal traffic changes from Discover around core updates, but systematically collected, large-sample traffic data linked to this specific update is not yet publicly available.

Event-driven skew

• The post-update window overlaps with the Super Bowl, Winter Olympics, and ICC Men's T20 World Cup, all of which increase demand for sports and live news.[S1]
• Part of the category expansion and X.com growth may be related to real-time coverage of these events, which could recede as the news cycle normalizes.

International scope

• Google has confirmed plans to expand the Discover-focused update beyond English-language US users but has not yet published clear timing or separate impact data for other markets.[S2]
• No equivalent large-scale panel data for non-US Discover feeds has been released, leaving international impacts largely unquantified.

Data appendix

Approximate key metrics from the NewzDash DiscoverPulse February 2026 scorecard.[S1]

Sources

• [S1] NewzDash, "Google Discover Feb 2026 Core Update Scorecard: Data Shows What Actually Changed", methodology and datasets for US, CA, NY Discover feeds, accessed February 2026.
• [S2] Google Search Central and public statements on the February 2026 core update and revised Discover documentation, including goals around local relevance, clickbait reduction, and topic expertise, accessed February 2026.
• [S3] NewzDash, "X.com Quietly Becoming a Google Discover Powerhouse in the US", longitudinal analysis of X content in Discover, accessed November 2025.
• [S4] Early December 2025 core update analyses (for example, Search Engine Journal coverage) highlighting specialized and brand sites gaining share on "best of" and similar queries, accessed February 2026.