AI News

Google rolls out a Preferred Sources button: a chance to win back traffic

Google’s Preferred Sources button lets readers mark favorite publishers and see them more in Search, Discover, and Google News. Discover and audio briefings gain new controls.

2026-08-21 ·Hai Anton

Website traffic keeps shrinking as AI answers grow inside search results. In response, Google introduced an interactive Preferred Sources button that publishers can embed on their pages. It lets readers mark a publisher as a favorite so its links surface more often in Search, Discover, and Google News. The company is also expanding personalization: natural‑language tuning for the Discover feed and customization of audio daily briefings in the Google News app on Android. What does this mean for publishers and readers right now?

What is the Preferred Sources button and why does it matter?

Preferred Sources is a new interactive button that online publishers can embed on their sites. When readers press it, they mark a publisher as a favorite, and Google highlights its links more often across Search, Discover, and Google News. The idea is straightforward: speed up access to trusted sources when people search for information or learn about a topic through Google’s AI experiences.

The button arrives amid a long‑running shift in user behavior, where AI summaries take a larger share of attention. For publishers, that means fewer clicks and tougher competition in results. The “favorite” signal gives readers a direct way to say, “Show this more.” It does not rewrite the entire algorithm, but it adds another meaningful cue.

The launch extends Google’s move toward user‑directed controls. Preferred Sources was previously available in Top Stories. In May, the company brought it to AI experiences, including AI Mode and AI Overviews. Now publishers get a hands‑on element that moves the action from settings to a one‑tap choice on their own site.

The goal is clear: make trusted content more visible where people already spend time — in Search, the Discover feed, and Google News. If users keep relying on short AI overviews, the ability to spotlight familiar brands becomes critical. This is an attempt to return control to both readers and publishers.

How do readers favorite sites, and where does it apply?

There are two simple paths. First, a reader taps the Preferred Sources button directly on a publisher’s website. Second, they can visit Google’s source preferences page and find a publisher by name or by website. After adding a favorite, Google highlights links from that site more often across Search, Discover, and Google News.

This approach demands little effort from readers. A click on a familiar page or a quick lookup in preferences, and the impact appears across multiple surfaces. That matters: users do not need to re‑filter content every time. The signal persists and works wherever they interact with search and news.

As of the May launch, Google said people had already selected over 345,000 unique sources through this method. That suggests users are willing to make a one‑time choice to see a known brand more often. For publishers, it is an opening: the sooner the button appears on a site, the sooner loyal readers can send their signal.

It is also notable how broad the coverage is: Google applies favorite markings not only in classic search, but in Discover and Google News. Additionally, in May the option arrived in AI Mode and AI Overviews. That widens the moments when a “favorite” brand can win the click as users read summaries or look for background.

Can this actually bring traffic back to publishers?

Google states plainly that becoming a preferred source can drive more traffic. In earlier studies, it found that people are twice as likely to click through when a preferred source is available. The added tool is meant to soften the damage from the rapid growth of AI‑powered search features for traffic‑dependent businesses.

For newsrooms, this is not a silver bullet, but it is a controllable lever. The button meets readers where loyalty already exists. If they are willing to favorite a publication once, Google will highlight its links more often. When competition intensifies in feeds and overviews, even doubling click‑through odds can add up over time.

The logic is simple: less friction yields more action. The embedded button is clear, fast, and contextual. It sits next to content the reader already values. That increases the chance the “favorite” signal happens now, without extra steps or detours into settings.

At the same time, Google does not hide the broader shift: AI search changes user behavior and eats a share of clicks. The response is to offer tools that work inside the new reality. Preferred Sources helps build a habit of seeing known brands more often. If you are a publisher, place the button prominently and explain its purpose.

What other personalization tools is Google adding to Discover and News?

Alongside the button, readers will soon be able to customize the Discover feed in Google’s app “in their own words.” Tap any three‑dot menu in the feed, then tell Google which topics you want more or less of — using natural‑language commands. These signals help Google refine the feed in real time.

This lowers the barrier to control. No need to sift through category lists or rate dozens of cards. You can phrase your request the way you would tell a friend: “Show more analysis on this topic” or “Less of quick digests.” The system processes the intent and adjusts the feed to your expectations immediately.

In addition, Google says Android users will be able to customize their audio daily briefings in the Google News app. For people who listen on the go, that means more control over what lands in the daily rundown. Combined with Preferred Sources, this forms a broader toolkit for shaping a personal news experience.

The search giant is not alone here. In recent months, a number of top social apps launched user‑controlled algorithms that let people fine‑tune what is recommended to them. The trend is clear: more control for audiences, and more chances for publishers to be noticed where users spend time and seek answers.

Based on the original source.

Ready to automate your store?

We'll analyze your workflows, find the bottlenecks, and propose a concrete automation plan. First consultation is free.

Message us on Telegram →
Hai Anton
Hai Anton

Founder of HAIQ — AI Automation Agency. Founder of HAIQ. I build automations and AI solutions for Ukrainian e-commerce on n8n. I write about automation, chatbots, and AI for business.