The Clicks Are Gone. Now What?
Katie Morton and I just recorded the debut episode of Search Engine Journal's revamped podcast, and we kept circling back to one uncomfortable truth: clicks on the open web are dropping fast, and most of us still don't have a great answer for what to measure instead.
Here's the headline number that sets the table. Rand Fishkin pulled Similarweb clickstream data and found that roughly 68% of U.S. Google searches in the first four months of 2026 end with zero clicks at all [source: https://www.searchenginejournal.com/where-search-attention-is-going-how-to-measure-it/582142/]. Only about a quarter of searches actually send anyone to an external website. The rest either stop on Google, loop back into another search, or land on a paid result or Google-owned property.
That's not a rounding error. That's the new baseline.
So where is all that attention going? It's collecting in a handful of places most analytics tools don't track well. AI Overviews hold attention inside Google. Business Profiles, Maps listings, and newer surfaces like Ask Maps are pulling people toward specific businesses without ever generating a traditional click. Community platforms are showing up more often in results, too.
After May's core update, SE Ranking's tracking showed Reddit gaining top-three positions across all 20 niches it follows [source: https://www.searchenginejournal.com/where-search-attention-is-going-how-to-measure-it/582142/]. Google's partnership with Reddit has been live for a while, and they've leaned harder into forum results. I wouldn't say the deal alone explains the movement, but it's clearly part of the picture.
Transactional searches, local queries, and branded terms still send clicks to the open web. The content categories getting crushed are lifestyle pieces, wire-service journalism, and utility pages like weather forecasts and TV listings — the kind of thing an AI Overview can answer in three sentences without sending anyone anywhere.
And a lot of what does leave the open web lands on social platforms and video, where AI can't easily replicate the formats. Younger audiences especially are spending their time there instead of on traditional sites, and YouTube holds attention across age ranges.
The uncomfortable part: if your content lives in one of those hit-hard categories, the click decline isn't a temporary blip. It's structural.
The Content Bets That Stopped Paying Off
For years, the play was simple: cover every related topic with generic content and let Google index you into traffic. That playbook is broken now.
Blogs that tried to cover everything face stiffer competition from near-identical pages and AI summaries that answer directly in the search results. Volume alone doesn't cut it anymore, and Google's Danny Sullivan made this brutally clear at Search Central Live Toronto this spring [source: https://www.searchenginejournal.com/where-search-attention-is-going-how-to-measure-it/582142/].
Danny drew a line between commodity content and the kind of work only your business and your experience can produce. The listicle that could come from anywhere sits on one side. What only you can say sits on the other. The topic matters less than which side of that line your content falls on.
I think most people misread this. They assume it means "write weirder stuff." It doesn't. It means your content needs a point of view that couldn't be generated by scraping a dozen other pages and running it through a summarizer. If your piece could've been written by literally anyone, it's on the wrong side of Danny's line.
John Mueller made a related point on Reddit: for a new site, the audience work comes first, and search visibility may follow once you've built something worth finding [source: https://www.searchenginejournal.com/where-search-attention-is-going-how-to-measure-it/582142/]. Starting with "how do I rank" is the wrong place to start. Start with who you're serving and what they actually need.
Katie Morton tied this back to business fit in a way that's easy to nod at and hard to actually practice. SEJ doesn't take every client who waves money, because off-topic or low-quality sponsors can cost you your audience's trust. Her principle is straightforward:
It always comes down to authenticity. You want to provide value, you want to serve your audience. It has to be this mutually beneficial thing.
Old-school SEO principles don't die, and firsthand experience is the one thing AI genuinely can't replicate. The brands and publishers that survive this shift are the ones building something only they can build.
Measuring Visibility When The Click Never Comes
Here's where it gets messy. Raw traffic used to be a clear indicator of how well your content was performing. Today, your brand can appear throughout search results and still not show up in analytics at all if it didn't lead to a click.
Branded search is one signal worth monitoring alongside citation tracking. Repeated exposure in AI answers may lead to more branded queries over time, and Search Console's branded queries filter makes these changes easier to spot [source: https://www.searchenginejournal.com/where-search-attention-is-going-how-to-measure-it/582142/]. But branded search alone doesn't tell the whole story either.
Another shift involves moving from links to mentions. Unlinked mentions may matter in ways a typical link report can't detect. A brand might get referenced in a Reddit thread, which a chatbot later relies on for a recommendation — no hyperlink in the chain at all. If you only count links, you're missing half the picture.
Katie described this from the reader's side while shopping for merino wool travel clothes:
I'm seeing brand names, but I'm not always clicking during the research process. I'm seeing brand names mentioned over again on the SERP and Reddit and in Claude, my preferred AI chatbot … eventually it will trigger a branded search, and then I'll go directly to someone's website, and they would probably have no idea how I found them.
That's the attribution gap a lot of ecommerce is living in right now. Direct traffic they can't tie back to any source.
Her practical advice: track where you're being cited, keep a brand presence on Reddit, and make your content crawlable by AI [source: https://www.searchenginejournal.com/where-search-attention-is-going-how-to-measure-it/582142/].
No click doesn't mean no value. If your content helped form the answer someone got, you may be in the consideration set even when no referral follows. The measurement job is just harder now, and it demands more honesty about what your analytics tools can actually tell you.
Resetting The Scoreboard As A Content Leader
Katie emphasized that proving value as a content leader starts with identifying your north star. Revenue and the bottom line are central, but editorial and marketing need to work hand in hand with business strategy while still putting the audience first [source: https://www.searchenginejournal.com/where-search-attention-is-going-how-to-measure-it/582142/].
Part of the strategy involves niching down. SEJ improved its contributor program by becoming more selective about writers, prioritizing genuine expertise over self-promotion. That's a move I think more publishers need to make, even if it means fewer bylines in the short term.
My piece of this is the scoreboard. When clicks stop being the main number, you make the clicks you do get count for more. If the only metric you report is organic sessions, and Google's sending a smaller share of searches to the open web, that line is going to keep sliding.
Keep the declining sessions in view and set them alongside the growing signals — branded search queries, mentions, citation counts. A wider scoreboard can highlight value that a sessions-only report might miss.
It won't undo the revenue loss caused by traffic that doesn't return. But managing both aspects together — the declining clicks and the alternative signals — is truly the key to success now. The content teams that figure this out first will have a serious advantage over the ones still reporting to a scoreboard that no longer reflects reality.