Google Discover Optimization: How to Get Consistent Traffic from the Feed

Most SEO practitioners spend their careers optimizing for search queries. They refine keywords, build backlinks, and tweak meta descriptions-all to intercept people actively searching for something. Google Discover flips that model entirely. Unlike traditional search, Google Discover is an AI-curated feed that surfaces articles, videos, and pages based on individual user interests and browsing behavior. Users never type a query. And here's why that matters right now: Google Web Search traffic to news publishers declined from 51% to 27% between 2023 and 2025, while the Discover feed climbed to 68% of Google referrals, creating volatility for content creators.

If your content strategy still treats Discover as a bonus channel you occasionally stumble into, you're leaving one of the most powerful distribution mechanisms on the table. On February 5, 2026, Google announced its first-ever Discover core update , signaling that this feed is no longer a side experiment-it's a platform Google is actively investing in and reshaping. The rules changed. The opportunities grew. But so did the complexity. This guide breaks down what actually works for earning consistent Discover traffic, based on Google's own documentation, practitioner data, and the post-update reality of 2026.

How the Discover Algorithm Actually Selects Content

Understanding Discover requires forgetting most of what you know about traditional SEO. The main feature is that Discover is not triggered by search queries, but it still analyzes content topics and relevance signals. Instead of matching keywords to intent, the algorithm uses machine learning to analyze each user's browsing history, app activity, location patterns, search history, and interaction signals-like what they follow-to build a real-time content feed tailored to that specific person.

Think of it as Google predicting what you'll want to read before you know you want to read it. The algorithm assigns relevance scores to pages mapped to known entities and topics , then matches those scores against each user's interest graph. What does this mean in practice? Once an article enters the system, a scoring and rescoring loop begins. Continuous CTR, impressions, interactions, and click quality feed models like Navboost to refine what gets shown. The algorithm distinguishes between good and bad clicks- valuable content is decided by the good vs. bad click ratio, and quality visits are used to measure lasting satisfaction and re-rank top performing content.

This is a critical insight most optimization guides miss: Discover doesn't just care that someone clicked. It cares what happened after the click. A high bounce rate or quick pogo-stick back to the feed signals dissatisfaction. A long, engaged read signals quality.

What the February 2026 Discover Core Update Changed

Google's first-ever dedicated Discover update wasn't incremental. In Google's own words, the update was designed to improve Discover by "showing users more locally relevant content from websites based in their country," "reducing sensational content and clickbait," and "showing more in-depth, original, and timely content from websites with expertise in a given area."

Three shifts stand out from the post-update landscape:

• Originality now outweighs presentation.

Original content is weighted more heavily, summarization-only content is being deprioritized, and E-E-A-T signals now play a larger role. Sites that rely on aggregating or lightly rewriting existing information saw their Discover impressions drop significantly.

• Geographic relevance became a ranking factor.

Discover now prioritizes content aligned with the user's location, which affects which articles are surfaced. Websites that fail to provide regionally relevant or localized content may experience lower visibility.

• Topic-level expertise replaced blanket authority.

Google's systems now identify expertise on a topic-by-topic basis. You don't get "authority" as a blanket label. You earn it in categories.

The February update also landed alongside another seismic shift. In July 2025, Google announced it would start testing AI summaries in Discover. By November, Marfeel reported that AI summaries made up 51% of the Discover feed in test markets. That's half the feed occupied by Google-generated summaries citing multiple sources-which means the remaining slots for traditional publisher content became even more competitive.

Building Topical Authority That Discover Recognizes

Discover doesn't rank pages. It recommends publishers. That distinction is everything.

Discover relies on topical authority to match content with interested users. Sites that cover dozens of unrelated topics dilute their authority signals and are less likely to be recommended for any individual topic. If you're a marketing agency writing about marketing, then publishing a random article about pet nutrition, that dilution hurts you. The path to Discover authority follows a specific pattern: Map your entity territory. The Discover algorithm selects content based on entity-driven interest modeling. If your content doesn't clearly relate to known entities, topics, or categories, it won't align with user behavior profiles. Use Google's NLP API, tools like InLinks or MarketMuse, and your own Search Console data to identify which entities your site already owns-and where gaps exist. Build pillar-and-cluster architectures. A pillar-and-cluster content architecture helps Google map your topical expertise and recommend your content to users interested in those specific areas. Each cluster page should target a distinct entity and link contextually to sibling pages and the pillar. Pages with no internal links to or from entity-related content are invisible as part of a cluster. Google's crawler follows links to discover relationships. If the link doesn't exist, neither does the relationship.

Publish with consistency, not volume. To increase your chances of getting into the Discover feed, maintain a consistent publishing schedule. Once you start driving Discover traffic, the algorithm trains to crawl your site with the frequency you publish, so your content appears consistently.

Aim for at least 2–3 high-quality pieces per week rather than sporadic large batches. A site that publishes one thoughtful piece daily within its niche will outperform one that drops ten articles on Monday and goes silent for two weeks.

The Technical Checklist That Precedes Everything Else

Before worrying about content strategy, check the technical basics. Missing any one of these can quietly disqualify your content from the feed.

The max-image-preview:large Meta Tag

This is the single highest-ROI technical change for Discover visibility. Implementing this tag led to significant improvements for publishers, including a 79% CTR increase for Kirbie's Cravings and a 30% CTR increase and 332% click increase over six months for Istoé.

The implementation is a single line in your page header:

<meta name="robots" content="max-image-preview:large">

The Max Image Preview meta tag tells Google what size image to show in the preview when a publisher's web page is shown in Discover. This meta tag is a directive, which means Google is obligated to follow the instructions. Without it, your articles appear as small thumbnails in the feed-easy to scroll past.

Image Specifications

Google recommends including compelling, high-quality images that are relevant, especially large images. Specifications include high resolution of more than 300,000 total pixels -for example, a 1280×720 image at 16:9 aspect ratio. Each content snippet receives a huge visual placement, so clear, relevant, and well-lit photos at least 1200 pixels wide help attract attention. Don't use a logo as the primary image.

Equally important: generic stock photos are increasingly devalued by the Discover algorithm. Invest in original photography, custom graphics, or professionally designed visuals. Practitioners who study high-performing Discover content report that images with people- especially when the person is looking at the camera with a strong emotion-increase click-through rates significantly.

RSS Feeds and the Follow Feature

The Google Discover Follow feature is a component of Google Discover, a way to capture a steady stream of traffic. It works by allowing users to choose to receive updates about the latest content on a site they are interested in. The mechanism behind it is your RSS or Atom feed.

Google updated their guidelines for the Discover Feed feature to emphasize the importance of the feed <title> and <link> elements -ensure yours include both. For sites with multiple content categories, implement category-specific feeds so users can follow only the topics they care about. Link your feed in the <head> section of both your homepage and individual post pages:

<link rel="alternate" type="application/rss+xml" title="Your Site - Topic" href="/feeds/topic.xml" />

Keep your feed updated, just like updating your sitemap. Ensure your feed is not blocked in your robots.txt file.

Core Web Vitals for Mobile

Discover is a mobile-first experience. Core Web Vitals matter significantly: LCP under 2.5 seconds, INP under 200 milliseconds, and CLS under 0.1. Pages that load slowly or shift layout on mobile are less likely to be recommended.

You can improve these by minimizing JavaScript, compressing images, using more responsive formats like AVIF, and optimizing for mobile.

Crafting Headlines and Content That Earn Clicks Without Manipulation

Headline optimization for Discover walks a tightrope. Google's documentation explicitly warns against clickbait. Google recommends avoiding tactics that artificially inflate engagement by using misleading or exaggerated details in preview content, or by withholding crucial information required to understand what the content is about.

Yet practitioners observe a nuance. Research shows that somewhat clickbaity headlines perform way better in Google Discover than more dry factual headlines. In this sense, Google Discover is similar to Facebook. The distinction? Curiosity without deception. A headline that generates genuine interest while accurately representing the article succeeds. A headline that promises something the article doesn't deliver gets penalized-and after the February 2026 update, the algorithm now identifies headlines and articles that overpromise or provide minimal informational value, reducing their visibility.

Practical guidelines for Discover headlines:

• Lead with the emotional or surprising element without giving away the full answer
• Be specific and descriptive-vague titles don't compel clicks in a fast-scrolling feed
• A/B test when possible.

You can A/B test headlines and monitor CTR changes in Google Search Console's Discover report to find the most effective strategies.

• Avoid withholding key context. Google's policy specifically targets titles that make readers guess what the content is about

Content structure matters too. User behavior now plays a larger role. Metrics such as time on page, scroll depth, and return visits influence Discover visibility. Write opening paragraphs that immediately deliver value. Use short paragraphs, clear subheadings, and embedded visuals to maintain engagement. Discover rewards content people actually read, not just content they click.

Driving Early Engagement to Trigger the Algorithm

Here's what most Discover optimization guides understate: the algorithm evaluates early performance signals aggressively. Once an article is published and shared on social media, Chrome click data is stored and applied to the algorithm. The more quality click data and shares you can generate early in an article's lifecycle, the better your chance of success on Discover.

This means your Discover strategy isn't just about what you publish. It's about how you launch it.

There is a strong correlation between the clicks, likes, shares, and comments an article gets on Facebook and Twitter and the article's success on Google Discover. Google tends to show articles from websites people have visited before. So maximizing engagement across channels-email newsletters, social distribution, push notifications-feeds back into the Discover algorithm. Think of the first 24–48 hours after publication as the audition period. Discover shows mostly fresh content that is relevant for one to three days. If your article gains traction in that window, the rescoring loop works in your favor. If it doesn't, the window closes. Concrete actions for launch day:

• Share to your email list within hours of publication
• Distribute across social channels with engagement-optimized messaging
• Embed the article in internal notification systems (push, in-app)
• Promote within relevant communities where genuine discussion happens

Generate as much interaction with your content as possible, including comments-there is a relationship between comment volume and Discover success.

Measuring What Discover Actually Sends You

One reason publishers chronically underestimate Discover traffic: many publishers are not giving optimizing for Google Discover the attention it deserves because they are underestimating the amount of traffic they get from Google Discover, as Google Analytics does not correctly track it.

Google Search Console is the authoritative source. The Performance report for Discover shows impressions, clicks, and CTR for any content that has appeared on Discover in the last 16 months, as long as your data reaches a minimum threshold of impressions. Filter by "Search type: Discover" to isolate the data. The February 2026 Discover Core Update created a meaningful separation between Discover performance and search performance. If you have access to Search Console, you can filter by search type to see the breakdown.

Google Analytics 4 requires manual configuration to properly attribute Discover traffic. In GA4, a significant portion of this traffic is recorded as direct because the system does not receive referrer data. Some transitions are classified as "Unassigned" or "Referral" because they are not classic HTTP referrers. The fix: create a custom dimension for page_referrer, look for com.google.android.googlequicksearchbox as the referrer value, then set up a custom channel grouping to properly classify it. Metrics to track weekly:

• Impressions (not just clicks)-

John Shehata, former VP of Audience Development at Condé Nast, emphasizes that impressions, not clicks, are the key metric to watch.

• CTR by article to identify which headline/image combinations resonate
• Discover vs. Search performance to understand if updates hit one surface differently
• Top-performing topics to double down on your entity strengths

Navigating the AI Summary Reality

The competitive landscape for Discover has permanently changed. Marfeel's Discover Monitoring shows the platform has transformed into an AI-powered ecosystem where YouTube videos and AI Summaries occupy space that once belonged to news organizations. In the United States, Brazil, and Mexico, 51% of the feed now consists of AI Summaries.

Google has been rewriting headlines and snippets of news articles in Discover's personalized content feed using AI, instead of displaying the ones supplied by publishers. By January 2026, Google reframed the update as a "feature" that "performs well for user satisfaction." AI headlines are likely here to stay. What does this mean for your optimization strategy? First, accept that some Discover impressions won't generate clicks. When Google summarizes your content in the feed, users may absorb the key points without visiting your site. The winning approach shifts from volume to depth: publish fewer, better assets that stand up to scrutiny. From traffic to visibility: measure where your brand shows up, not just who clicks.

Second, structure your content so it's valuable even when partially summarized. Lead paragraphs should convey your unique perspective immediately. If Google's AI is going to excerpt your content, make sure that excerpt includes your brand signal and a compelling reason to read further. Third, diversify. WebFX recommends treating AI-driven discovery as a catalyst to diversify traffic sources-newsletters, apps, social media, and owned communities. Each channel provides a direct connection to users, reducing dependence on any single source.

Building a Resilient Discover Strategy for 2026 and Beyond

Traffic from Discover is less predictable or dependable when compared to keyword-driven search visits. Google itself says you should consider traffic from Discover as supplemental to your keyword-driven search traffic. That's Google's own documentation. Accept it. But "supplemental" doesn't mean "ignorable." For sites in the Raptive network, 2025 data showed Google Discover traffic was down 4% year over year overall, but up 5% in the second half of 2025 -after they began actively optimizing. The difference between sites that experience Discover as random spikes and those that build consistent traffic comes down to treating it as a system, not a lottery.

Discover traffic is unpredictable and sometimes spiky. But you can smooth the spikes by doing what the algorithm rewards: publishing original, expert-level content on a consistent schedule, within clearly defined topical territories, with technically optimized pages and compelling visuals. Layer in early-engagement promotion across your owned channels. Monitor performance weekly in Search Console. Iterate on what works.

For publishers, the strategic question has shifted from "how do we win Discover" to "how do we remain resilient when Discover changes." That shift-from chasing visibility to designing for durability-may define the next phase of digital publishing. The practitioners building real audience relationships alongside their Discover strategy won't just survive the next algorithm update. They'll be positioned to benefit from it.