Google Discover Core Update February 2026 Completed: What the Newzdash Data Reveals

22 days, three weeks of uncertainty, and countless nervous glances at the Search Console – it’s now officially over. Google declared the February 2026 Discover Core Update completed on February 27th. The brief confirmation on the Google Search Status Dashboard came at 5:02 AM ET: “The rollout was complete as of February 27, 2026.”

In my article at the start of the update, I analyzed what Google is planning and how to prepare in the DACH region. In the comprehensive Discover SEO Guide, you will find the complete technical checklist and workflow. I am dedicating this post to the two questions that really matter right now: What does the initial data show? And what do they mean in concrete terms – including for me at seo-kreativ.de?

Google had announced “up to two weeks” for the rollout. It actually took 22 days – from February 5th to February 27th, 2026. Barry Schwartz from Search Engine Roundtable had already suspected the delay in mid-February when no official confirmation came after the two weeks had passed.

What does this tell us? Not necessarily anything dramatic. It was similar with the December 2025 Core Update – which took 18 instead of the announced 21 days. Rollout duration and update intensity do not necessarily correlate. But: An extra week of work on a Discover-only update – the first of its kind ever – suggests that Google made careful adjustments here before putting its stamp on it.

Search Engine Land also confirmed the conclusion and emphasized that this was the first time Google explicitly designated a Core Update as a pure Discover update. Crucial for publishers: From now on, the effects can be analyzed cleanly without ongoing rollout fluctuations distorting the picture.

While most analyses on the Discover update rely on Google quotes and general recommendations, John Shehata provides real comparative data with the Newzdash scorecard. Pre-update data (January 25–31, 2026) was compared with post-update data (February 8–14, 2026) – across US Top 1,000 domains, US Top 1,000 articles, as well as separate exports for California and New York. The complete data analysis with all exports can be viewed directly at Newzdash.

What makes this analysis particularly valuable: The Newzdash DiscoverPulse data is based on a panel of millions of real US users, not on tool estimates. This is currently the most robust external data source available.

Local Relevance: The Domesticity Shift in Numbers

Google’s stated goal was to show users more content from their own country. According to Newzdash, the data confirms this at the domain level: The share of international publishers in the normalized total score dropped from 8.52% to 7.04%. At the same time, the US share rose from 88.86% to 89.94%.

The shift becomes even more tangible with individual publishers. UK media lost a massive amount of Discover visibility in some cases: According to Newzdash, The Sun lost about two-thirds of its score, The Independent about half, Reuters around a fifth, and even The Guardian recorded a double-digit decline. At the same time, US news brands like CBS News, NBC News, Axios, and AP News gained visibility.

But here is where it gets interesting: The BBC actually managed to gain in the post-update window. Newzdash interprets this cautiously – not as proof that Google treats the BBC as a US source, but as an indication that highly relevant international content can still break through. It’s not a hard geo-filter, but a scoring shift.

State-Level Personalization: Something Google Never Announced

One of the most exciting findings of the Newzdash analysis: In its announcement, Google only spoke of country-specific relevance. However, the data shows that Discover personalizes at the state level – meaning it differentiates between California, New York, and the national feed within the US.

A concrete example from the Newzdash scorecard: In California, the number of local articles in the Top 100 rose from 10 to 16 – an increase of 60%. In the overall Top 1,000 export, the share of Californian domains grew from 36 to 49 articles (+36%).

This is a strong signal. Because if Google is already differentiating by region within a country, there will likely also be differences within the DACH region when it rolls out internationally. An article about Munich real estate prices might then have different Discover chances in Bavaria than it would in Hamburg.

According to Newzdash, the anti-clickbait effect of the update can be pinpointed using concrete examples. And these examples are clear enough to derive a definitive message from them.

Autoevolution: From 5 to 0

Autoevolution had 5 articles in the US Top 1,000 in the pre-update window – all built on an almost identical “dramatic reveal” formula. Post-update: zero articles. Newzdash calls this one of the cleanest examples of how Discover apparently recognizes and suppresses repetitive sensationalist templates.

Yahoo: Out of the Top 100

Yahoo’s presence in the US Top 1,000 shrank from 11 to 6 articles. Even more telling: Yahoo had several entries in the Top 100 before the update – and not a single one afterward. This also fits the picture: Curiosity-gap formats are losing ground in the top feed positions.

Psychological Listicles: A clear crash

A listicle from Geediting in the style of “psychology says…” dropped in ranking from around position 14 to position 153. This supports a broader pattern: Format-driven content that relies on psychological triggers rather than substantive content is measurably losing Discover visibility.

What matters instead?

Newzdash summarizes the winning traits as follows: The biggest winners are not always the biggest brands, but often publishers who combine a clear thematic focus, usefulness, and up-to-date reporting in their core categories. This aligns exactly with Google’s “topic-by-topic” framing.

This reminds me of what I have been observing in classic search for a while: Google is increasingly focusing on thematic depth instead of breadth. Discover is now following suit – with its own, even stricter quality filters.

Writing Discover tips is one thing – appearing in Discover yourself is another. When I open the Discover report for seo-kreativ.de in the Google Search Console, I see: little to nothing. A German-language SEO specialist blog featuring deep-dive articles on E-E-A-T or the Semrush Sensor serves a clearly defined niche – but not necessarily the broad interests that typically dominate Discover feeds.

Nevertheless, this update gives me hope. It shifts the logic in three directions that are favorable for specialized blogs: The topic-specific expertise rating means that seo-kreativ.de does not have to become a general-interest magazine – the blog must become an authority in SEO. The local component shrinks the competition down to the German-speaking SEO scene, where I have a solid starting position with over 12 years of experience. And the anti-clickbait wave finally rewards what I do anyway: substantive content instead of sensational headlines.

Specifically, I am implementing three things that I will report on with real GSC data in the coming months: First, the image strategy – systematically switching to at least 1,200 px width, visually appealing motifs, and max-image-preview:large. Second, I am increasing the topic frequency for current SEO developments because Discover favors fresh content from expert sources. Third, I am expanding Discover monitoring – GSC annotations, weekly checks, correlation with publication dates. I don’t know if that will be enough. But for the first time, I feel like the rules of the game are tilting in my favor.

What distinguishes this article from the other Discover posts on seo-kreativ.de?

The article on the start of the update analyzes the three goals, the DACH assessment, and preparation. The Discover SEO Guide is the complete manual with a technical checklist and workflow. This article here has a different focus: Data. What do the Newzdash analyses show after the rollout is complete? Who won, who lost, and what can be deduced from that?

Is the Newzdash data reliable?

Newzdash uses a panel of millions of real US users and tracks Discover in real-time. This is the best external data source currently available. But: It is a panel, not a full census. Newzdash itself emphasizes that the data can show patterns and shifts – not absolute numbers for the entire Discover ecosystem. In addition, the post-update window overlaps with an active news phase, which can distort individual results.

Can a small specialized blog benefit from Discover at all?

The Newzdash data provides an encouraging signal for this: The biggest winners were not always the biggest brands, but publishers with a clear thematic focus and usefulness. Google’s own example – the local news site with gardening expertise – underlines this. Whether this also works in practice for German-language niche blogs remains to be seen – but the signals are more positive than ever before.

The February 2026 Discover Core Update is not just finished – it has been measured. Thanks to the Newzdash scorecard, for the first time in a Discover update, there is concrete before-and-after data that goes beyond anecdotes.

The core message is clear: Google’s three announced goals – local relevance, anti-clickbait, expertise – are not empty promises. The data shows that clickbait templates are surgically removed, that international publishers lose visibility in US feeds, and that there are even signs of regional personalization at the state level that Google never announced.

The update is not yet live in the DACH region – but the window to prepare is getting smaller. You can find concrete measures in my two other Discover articles: the update start post for background and DACH assessment, and the Discover SEO Guide for the technical checklist and workflow.

What personally resonates with me the most: This update rewards exactly what I have been working on at seo-kreativ.de for years – thematic depth, honest content, genuine expertise. Whether that is enough for Discover, the next few months will show. But for the first time, I feel like the rules of the game are tilting in my favor.