Google’s First Discover Core Update Is Rolling Out

Are your Discover clicks moving this week while your Search traffic stays the same? There is a likely reason. Google is rolling out the February 2026 Discover core update. It is the first core update aimed only at the Discover feed, not at Search results. It started on 5 February. For now it applies to English-language users in the US, and it is still in progress.

In this guide I explain what a Discover core update is and what Google says the feed will now reward. I also show you how I watch the numbers in Search Console without fooling myself, and what I’d leave alone until the rollout ends.

What Google Announced on 5 February

Google published a post on the Search Central blog called “Google’s February 2026 Discover core update”. It also logged the rollout on its Search Status Dashboard the same morning. The main facts are short:

• Start date: 5 February 2026.
• Scope: English-language users in the United States first. Google says it will expand to all countries and languages in the coming months. It gave no date.
• Rollout length: up to two weeks.
• What it changes: the systems that choose which articles appear in Discover. Google calls it a broad update to those systems.
• Google’s reasoning: its testing found people saw Discover as more useful and worthwhile with the update in place.

Google also says clearly that the change may cause ups and downs in Discover traffic. For most publishers, that is the honest summary of the next week or so.

What a Discover-Specific Core Update Means

Google runs several core updates a year. Until now, Google described them as changes to Search ranking systems. Discover traffic often moved during those updates too, but only as a side effect. This time the target is Discover itself.

This matters because Discover works differently from Search. Nobody types a query. Google guesses what a person will want to read from their interests, their activity and the sources they follow. Then it pushes articles into a feed in the Google app and on the mobile Google homepage. Your page is competing for a slot in someone’s feed, not for a spot on a results page.

So a Discover core update changes how Google picks articles for those feeds. It isn’t a penalty, and it isn’t aimed at specific sites. If your Discover traffic drops, it usually means other content now looks like a better fit for the same readers. It does not mean something on your site is “broken”.

Why Search and Discover Can Now Move Separately

The update targets Discover systems. So you can see a clear swing in Discover while your Search clicks stay flat. The reverse can happen too. A Search core update later in the year could hit you while Discover holds steady.

In my view, this is the most useful thing to remember while you read your reports. I always look at each surface on its own before I decide anything about the site as a whole.

What the Discover Core Update Rewards

Google listed three ways it wants to improve the feed. Each one points to a different kind of site that could gain or lose.

1. More Locally Relevant Content

Google says Discover will show more content from websites based in the user’s country. For a US reader, that means US-based publishers get more weight for the topics they cover.

Here is a simple example. A UK or Australian site that gets a big share of its Discover traffic from US readers could see some of it shrink, even if nothing on the site changed. A US regional news site, on the other hand, may gain visibility it didn’t have before.

2. Less Sensationalism and Clickbait

Google is openly cutting back sensational content and clickbait in the feed. This matches advice that has been in its Discover documentation for some time. Avoid misleading or exaggerated details in titles, snippets and images. Avoid tricks that use outrage, morbid curiosity or shock to get clicks.

Here is what this looks like in practice:

• A headline like “You won’t believe what this bank just did to savers” on an article about a 0.25% rate change is the kind Google wants to show less of.
• A headline like “Bank cuts easy-access savings rate to 3.5%: what it means for your account” tells the reader exactly what they will get.
• Thumbnails that promise something the article doesn’t give, like a dramatic crash photo on a routine traffic story, fall in the same group.

Honestly, I have never liked the curiosity-gap headline. It may win a click, but it rarely builds a reader who comes back.

3. In-Depth, Original, Timely Content From Sites With Expertise

The third goal is to show more in-depth, original and timely content from websites with expertise in a given area. The key detail is this: Google says it judges expertise topic by topic, not for the whole site.

Google gave an example along these lines. A local news site with its own gardening section can have built real expertise in gardening, even though it covers many other subjects. A movie review site that publishes a single gardening article probably hasn’t.

For publishers, this rewards depth inside a topic. One-off pieces that chase a trending subject you’ve never covered are likely to struggle more in Discover. Sections you’ve built up over time should do better.

What Stays the Same

Google says Discover will keep personalising the feed based on people’s creator and source preferences. If readers follow your site, or have chosen it as a source they want to see, that still counts. That is one reason I’d keep up the work of getting readers to pick your site as a preferred source alongside quality improvements.

The usual Discover basics still apply too. Your content must be indexed and follow Discover’s content policies. It should use clear titles. It should use large, high-quality images at least 1,200 pixels wide, turned on through the max-image-preview:large setting (a robots setting that lets Google show big image previews).

Who Is Likely to Notice the Discover Core Update First

The first wave covers only US English users, so the impact won’t be even. Here are some likely situations:

• US publishers with US audiences: the most direct exposure, up or down.
• Non-US sites with big US Discover audiences: watch the United States row in your country data closely. The local relevance change is aimed right at this case.
• Sites with mostly non-US Discover traffic: you may see little change for now. That doesn’t mean you’re safe. The update is due to reach other countries and languages later, so treat this as early warning.
• Sites that rely on trending, broad-topic articles: if your Discover wins come from topics outside your usual coverage, watch the topic-by-topic expertise signal.

How to Monitor the Update in Search Console’s Discover Report

The Discover performance report in Search Console is the only place that shows how Google’s feed treats your pages. It sits under Performance in the left menu, below the Search results report. If you don’t see it, your property hasn’t reached the minimum number of Discover impressions Google needs to show the report.

What the Report Gives You

• Metrics: clicks, impressions and average CTR (click-through rate). An impression counts when your item scrolls into view in the feed. A click counts when someone taps through to your page.
• Dimensions: page, country, Discover appearance (such as Web Stories or videos) and date.
• No queries: Discover has no search terms, so you look at pages and countries instead.
• Data delay: Google says data is normally ready within two to three days, and the latest days may be preliminary.

That delay matters a lot right now. On 7 February, the report will only just show the first day or so of the rollout, and those numbers can still change. Anything you decide this weekend is a guess.

Set Up a Clean Comparison

Discover traffic is spiky by nature. One article can take off for a day and then vanish. To separate the update from normal noise, compare periods of equal length that cover the same days of the week.

What I do is set a baseline of the four weeks before 5 February. Then I compare it with the rollout period once it ends. If you already use longer views to smooth out daily noise, the same idea from my guide to Search Console’s weekly and monthly views applies here. Judge trends, not single days.

Filter to Where the Update Is Live

The first wave targets the US. So add a country filter for the United States and compare it with your other main countries. If US Discover clicks fall while UK or Canadian Discover clicks hold steady over the same period, that points to the update, not a site-wide issue.

Look at Pages, Not Just Totals

Switch the table to the Pages tab and sort by clicks for both periods. This is the step I spend the most time on. You’re looking for patterns:

• Do losing pages share a topic you rarely cover?
• Do they share a headline style that leans on curiosity or shock?
• Are gaining pages from sections where you publish often and in depth?

Those patterns match what Google said it is changing. They will tell you far more than the total line on the chart.

Check Search Separately

Open the Search results performance report for the same dates. If Search is stable and only Discover moved, the issue is specific to the feed. If both dropped, I’d look for a technical cause too before blaming the update. Think indexing problems, robots changes or a template change.

What Not to Do Mid-Rollout

A two-week rollout tempts people to react to every daily wobble. In my experience, most of those reactions waste effort or create new problems. Here’s what I’d hold off on until Google confirms the rollout is complete.

1. Don’t mass-rewrite headlines. Changing dozens of titles on old articles won’t put them back into feeds that have already moved on. It also muddies the data you’ll need later.
2. Don’t delete or noindex content in a panic. Removing pages because they lost Discover clicks can cost you Search traffic they still earn. Decide on pruning after the rollout, using full data.
3. Don’t fake freshness. Changing publish dates, or republishing old articles with small edits to look timely, is exactly the kind of trick Google’s quality systems are built to see through.
4. Don’t change image or robots settings on a hunch. Removing max-image-preview:large or changing robots rules can knock you out of the feed’s large-image format completely.
5. Don’t judge from one or two days. Data lags by two to three days, and Discover is jumpy even in quiet weeks. Wait for the full rollout plus a few days of settled data.
6. Don’t assume a drop is a penalty. A core update is not a manual action. If you had a policy issue, it would show in the Manual actions report in Search Console.

What you can do now is keep publishing as normal, to the standard you’d want Google to judge you on. I also keep a simple log of any big site changes with dates, so I can rule them out later.

What to Do Once the Rollout Finishes

Google will post on the Search Status Dashboard when the rollout is done. You can follow the dashboard entry for this update. When it’s done, give the data a few more days to settle. Then run your comparison properly.

If you lost Discover traffic, Google’s general core update advice applies. Look at your content honestly instead of hunting for a quick technical fix. For Discover, I ask these questions:

• Does each headline describe what the article really delivers?
• Is the article original, or does it repeat what everyone else published that morning?
• Does it go deeper than the competing stories on the same topic?
• Is it in a topic area where your site has a track record, with regular coverage and knowledgeable writers?
• Would a reader in your target country find it relevant?

Improvements based on those questions take time to show. Discover reacts faster than Search in some ways. But a full recovery after a core update can still take until the next update, which may be months away. I walked through a similar process in my review of what changed in the December 2025 core update, and the same patient, page-by-page approach works here.

Should You Change Your Discover Strategy?

For most sites, I’d sharpen the strategy, not replace it. Google’s three stated goals aren’t new ideas. They put into words what its Discover guidance has asked for over many years.

The bigger lesson is about depending on one source. Google’s own documentation says Discover traffic is less predictable than Search traffic and suggests treating it as an extra. If one feed update can move a big share of your visits, it’s worth building steady Search demand, email subscribers and direct readers alongside it.

A practical plan for the coming months could look like this:

1. Find the two or three topics where you already publish most often and with the most depth.
2. Put more of your Discover-friendly effort, such as timely, original reporting and strong images, into those topics.
3. Write headlines that are specific and accurate, then check whether they still earn clicks in Discover.
4. Cut back on one-off trending pieces outside your core areas, unless you can truly add something new.
5. Prepare now if your main audience is outside the US, since the update is expected to reach other countries and languages later.

Conclusion

The February 2026 Discover core update is the first time Google has aimed a core update at the Discover feed itself. It began on 5 February for US English users. It is still rolling out, and Google expects it to take up to two weeks.

Google has been clear about the direction: more locally relevant content, less clickbait and sensationalism, and more in-depth, original, timely work from sites with real expertise in a topic. My advice for this week is simple. Set up a clean comparison in the Discover report, filter by country and page, and avoid big changes until the rollout is confirmed complete.

Your next step: open Search Console today and note your Discover baseline for the four weeks before 5 February. Then set a reminder to check the Search Status Dashboard for the completion notice.