Small Publishers Lost 60% of Their Google Traffic. I've Spent Two Weeks Figuring Out What to Do About It.

We sat with the Chartbeat numbers for a long time before writing a word of this. Not because they shocked us. We've watched small sites quietly bleed out for over a year now. It was seeing it all stacked up in one report, handed to Axios in March 2026, that made it land.

Small publishers, the sites doing 1,000 to 10,000 daily page views, lost 60% of their Google search referral traffic over two years. Medium publishers lost 47%. Large publishers lost 22%.

Read that again. If you run a small site, you lost roughly three times more than the big players did. Same algorithm. Same search engine. Wildly different outcome.

This is not a blip, and we'd be lying to you if we called it one. So we won't. Instead, here's what's actually happening, why small sites got the worst of it, and the playbook we'd run if our own site woke up tomorrow down 60%.

What does the data actually say?

Small publishers lost 60% of their Google search referral traffic over two years, while large publishers lost 22%, according to Chartbeat data covering more than 2,500 publisher websites. That's the headline, and it holds up.

The rest of the picture is grim in a more boring, structural way:

• Google Search page views fell 34% year on year between December 2024 and December 2025, per the same Chartbeat dataset reported by Axios.
• Google Discover, the other big referrer, dropped 15% over that period.
• Globally, Press Gazette reports Google search traffic to publishers was down 33% in the year to November 2025, and down 38% in the United States specifically.
• And the supposed replacement? ChatGPT referrals grew more than 200%, but AI chatbots still send less than 1% of all publisher page view referrals.

That last number is the one we keep coming back to. Everyone loves the story where AI search quietly takes over from Google. The data says AI isn't replacing the traffic. It's just removing it. The clicks are evaporating, not migrating.

Worth a small dose of perspective, though. Search Engine Land notes that total weekly page views across all publishers fell only about 6% from 2024 to 2025. The traffic isn't vanishing from the web. It's redistributing, and small sites are on the wrong end of the redistribution.

Why is this happening? Blame the AI summaries (mostly)

Here's the mechanism, and it's not a mystery. When Google answers the question for the user at the top of the page, the user has no reason to click through to you.

The cleanest evidence we've seen comes from the Pew Research Center, which tracked the real browsing behaviour of 900 US adults across 68,879 Google searches in March 2025. Their findings:

• Around 18% of searches produced an AI summary at the top of the results.
• When an AI summary appeared, users clicked a traditional search result just 8% of the time. Without one, they clicked nearly twice as often, 15%.
• Users almost never clicked the links inside the AI summary itself. That happened in just 1% of visits.
• People were more likely to end their browsing session entirely after seeing a summary (26%) than without one (16%).

So the AI answer doesn't redirect attention to a different link. It ends the journey. The user got what they came for and closed the tab. That's the zero-click problem in one tidy statistic, and it's exactly the dynamic we unpack in our breakdown of the AI Overviews click-through rate drop.

And these summaries are spreading fast. The Reuters Institute's 2026 trends report notes AI Overviews were appearing on roughly 10% of US search results at the time of writing and rolling out quickly elsewhere. We track the prevalence trend in detail in our piece on the rise in AI Overview appearances. The direction is one way.

Why small sites got hit three times harder

The gap between 60% and 22% isn't bad luck. It's structural, and once you see the four reasons you can't unsee them.

1. Brands have a moat. You don't (yet)

Google increasingly favours the sites people search for by name. Type "BBC recipe" or "Healthline migraine" and Google knows precisely what to serve. Type a generic query and it defaults to the recognisable, trusted names. A two-year-old niche site doesn't have that brand recall, so it sits in the part of the SERP that AI summaries are eating first.

2. AI summaries skim the answer off the top

If your whole page exists to answer a question that an AI summary can compress into two sentences, you're the most exposed site on the internet. Big publishers absorb this because they have brand loyalty and a wall of direct traffic. Small publishers had search as the lifeline, and that lifeline is the exact thing being severed.

3. Large sites can afford to diversify. Small ones can't

The big players are offsetting search losses through email, apps, social and direct visits, each with a dedicated team. A two-person site can barely keep one channel healthy. Search Engine Land put it bluntly: smaller sites now face real risk without a strong brand, direct audience relationships, repeat visitors or genuinely differentiated value. The cruel irony is that the sites that most need to diversify are the least equipped to.

4. Volume buys topical authority, and you're outgunned

Google rewards comprehensive coverage. A large publisher can run 50 articles on mortgage rates, each from a different angle. A small site might have three. In Google's eyes the large site is the authority, even when those three articles are flatly better. It's not fair. It's just how the topical authority maths shakes out, which is why we're so picky about internal linking patterns and how a small site connects what little it has.

The survival playbook we'd actually run

If we woke up tomorrow and a client's site had shed 60% of its traffic, here's where we'd put our hours. Not theory. Not a tidy list of "best practices". The actual plan.

1. Stop chasing head terms. Go embarrassingly narrow.

The days of a small site ranking for "best credit cards" or "how to lose weight" are gone. Those SERPs belong to brands with eight-figure budgets, and every hour you spend chasing them is an hour set on fire.

Go the other way. Hunt the weird, hyper-specific, long-tail queries the big publishers can't be bothered with. Not "best running shoes". Try "best running shoes for flat feet after plantar fasciitis surgery". Lower volume, sure. But near-zero competition and conversion intent through the roof. This is the heart of how we approach keyword optimisation in 2026: depth over breadth, intent over volume.

2. Build an email list like the business depends on it. Because it does.

Search traffic is rented. Email subscribers are owned. Every visitor who lands and leaves without giving you an address is a relationship you'll probably never get back.

Put a content upgrade on every high-traffic page. Not a limp "subscribe to our newsletter" pop-up. Something specific and genuinely useful, tied to what they just read. A checklist. A template. A PDF of the data you cited.

Aim for 3 to 5% of visitors converting. At 1,000 daily visitors that's 30 to 50 subscribers a day. In six months you've built a direct line to 5,000-plus people that no algorithm update can touch.

3. Be the primary source, not the summariser

AI summaries can rephrase existing information all day long. What they cannot do is run an experiment, gather original data, or share genuine first-hand experience. That's the moat.

Run a gardening site? Don't write "10 best tomato varieties" like everyone else. Plant all ten, photograph them weekly for three months, and publish what actually happened. That original data gets cited by AI, linked to by other sites, and can't be reproduced by anyone with a ChatGPT tab open.

This isn't us being romantic about "quality content". It's straight from Google's own people-first content guidance, which asks whether your page offers "original information, reporting, research, or analysis" and demonstrable first-hand experience. Google says trust is the single most important part of E-E-A-T. Original work is how a small site earns it. We've written more on turning that into a habit in our guide to using original research to win AI search visibility.

4. Own your local patch

If you serve a geographic area, go local and go hard. Local SEO is one of the last places where small genuinely beats big. Google needs local businesses for local results, and no national publisher is ever going to write about the best plumber in Ealing.

Get the Google Business Profile right, build local citations, and create neighbourhood-specific content. The local graph is one of the few arenas where being small is an actual advantage rather than a handicap.

5. Show up where small still reads as authentic

Reddit, TikTok and YouTube reward specificity and a real human voice over brand size. A solo creator with genuine expertise can out-engage a 50-person content team that all sounds the same.

Younger users in particular are starting their searches in more places than Google, often on TikTok, YouTube and Reddit before a traditional search engine ever loads. We dug into that shift in our look at how Gen Z uses TikTok as a search engine. Being present where your audience actually is, on platforms that reward depth and personality, gives a small site a real fighting chance.

6. Build relationships, not just a content calendar

Here's the one thing small publishers can do that big ones genuinely struggle with: build real community. Reply to every comment. Answer every email. Turn up in niche forums and be useful without pitching.

The small sites we've watched survive these drops all share one trait. They built a relationship with their audience that outlasts any single traffic source. When people come back because they trust you, the algorithm stops being the only thing keeping you alive.

What we'd stop doing immediately

The "stop" list matters as much as the "start" list. Maybe more.

1. Stop pumping out high-volume, low-depth content. If a machine can write it in 30 seconds, it's not worth publishing, and every thin page drags down your whole site's quality signal. That's not our opinion, it's how Google's site-wide helpfulness signal works.
2. Stop worshipping keyword difficulty scores. The old game of finding "easy" keywords with decent volume is broken. Compete where you have real expertise instead.
3. Stop ignoring Search Console. Open it right now. Find what's actually driving traffic, double down on those topics, and prune the dead weight. If you've never run a clear-out, our content pruning playbook walks through exactly how we decide what stays and what goes.
4. Stop waiting for things to "go back to normal". They won't. The 60% drop isn't a dip, it's a structural shift, and the sooner you accept that, the sooner you start building something that works in the world we're actually in.

Is it really that bad? An honest answer

We're not going to sugarcoat it. Some small publishers won't make it through this. The ones that leaned entirely on Google search for revenue and never built a single alternative channel are in genuine trouble.

And the people running these sites already feel it coming. The Reuters Institute surveyed 280 media leaders across 51 countries for its 2026 report, and on average they expect search traffic to fall by more than 40% over the next three years. Around a fifth of them expect to lose more than 75%. That's not doom-mongering from the sidelines. That's the people running newsrooms telling you what they see.

But we've also watched small, focused sites ride out every algorithm update by doing one stubborn thing: being genuinely, irreplaceably useful to a specific group of people. Not trying to be everything to everyone. Not chasing traffic for its own sake. Just solving real problems for real people and earning trust one visitor at a time.

The sites that come through this won't be the ones with the slickest SEO tactics. They'll be the ones that built something worth returning to no matter how people find them. That's slower to build, and it doesn't fit in a quarterly dashboard. But it's the only strategy we'd genuinely stake a business on.

If you're staring at a Search Console graph that's pointing the wrong way and you want a second pair of eyes, that's the work we do every day. Our SEO team can help you find the narrow queries worth winning and the channels worth building, and you can tell us what's happening to your traffic and we'll tell you, honestly, whether we think we can help. No pretending the 60% isn't real.