68.01 percent of Google searches ended without a click in early 2026, according to SparkToro’s latest analysis of U.S. search behavior. That number will travel fast, and for good reason, it sounds like a verdict on SEO.
It is not a verdict. It is a reset.
For early-stage founders, the more useful question is not whether search still works. It is where search belongs in the first-customer plan, when you have little traffic, little brand, and a short runway.
What the data says
Rand Fishkin’s June 2026 report, built with Similarweb data, shows that 68.01 percent of Google searches ended without a click between January and April 2026. In 2024, that figure was 60.45 percent. The direction is clear, more searches now resolve inside Google, not on the open web.
The report also says AI Overviews now appear on more than 20 percent of searches, and when they do, click-through rates can fall by roughly 60 percent. Google’s own AI Mode is still a small share of total searches, about 0.34 percent, but Google says it already has roughly 1 billion monthly users.
The trendline matters more than any one number. Search is becoming less of a referral engine, more of an answer surface.
Search is still useful. It is simply less often a handoff to your site.
What it does not say
The headline does not mean SEO is dead. It means the old expectation, rank, then receive traffic, is weaker than before.
It also does not mean content is worthless. In fact, SparkToro’s own argument is that AI answers are still built from ranked, crawlable content. If your work is not visible in the corpus search systems draw from, you are not contributing to the answer layer at all.
And traffic decline is not the same as revenue decline. Many searches were never going to convert. A higher zero-click rate can coexist with stronger buyer intent on the searches that still do produce visits.
Why founders should care
Most SEO commentary is written for companies that already have an audience. They have pages to protect, rankings to defend, and traffic to mourn.
That is not your situation if you are pre-scale.
If you have zero customers or only a handful, then zero-click search changes your starting point. It does not ask you to preserve an existing funnel. It asks you to build a reputation in places where people are already reading, even if they never visit your site.
That is a different problem.
For a startup with no traffic, the objective is not, “How do we maximize organic sessions?” The objective is, “How do we become legible to buyers, and to the systems that summarize the market?”
Start here
The early-stage playbook is not complicated. It is just less flattering than the old one.
Publish for citation
Write pieces that are specific enough to be quoted, not just skimmed. Original data, a point of view, a framework, a clear comparison, these are the assets that can earn citations in AI answers and still educate a prospect who finds you later.
This is where Better’s broader thesis matters, trust compounds when you publish useful work. A click may disappear, but the influence can remain.
Borrow audiences
Zero-click marketing is not only a search story. It is a distribution story.
In practice, that means founder-led posting on LinkedIn, niche communities, podcasts, partner newsletters, and customer-adjacent channels where your buyer already spends time. SparkToro has argued for audience research and borrowed attention for years, and that advice is even more relevant when search no longer guarantees a visit.
- Post the argument where your buyers already read
- Turn one strong idea into several formats
- Use replies, comments, and shares as signal, not vanity
- Send the deepest version to your own site only after the idea has earned attention
Keep SEO selective
Not all search is equal. Some query types still deserve investment, especially branded, local, and high-intent transactional searches. Cyrus Shepard’s work on what still wins in Google points to the same conclusion, broad informational SEO is weaker, but intent-rich search still matters.
That means you should be selective.
- Protect branded search, so people who already know you can verify you quickly
- Own bottom-of-funnel pages, where buying intent is explicit
- Use local or category-specific pages if your market depends on them
- Deprioritize generic top-of-funnel content that is unlikely to drive action
What to stop doing
If you are early, stop writing for volume first.
Do not build a content calendar around vague educational queries that AI can answer instantly. Do not chase keyword counts as a proxy for market understanding. Do not spend the first six months of your marketing effort manufacturing awareness that you cannot connect to a buyer.
Also stop asking content to do every job.
One article can build trust, another can support sales, another can train AI systems on your language. These are related outcomes, but they are not identical. Treating them as one thing leads to weak content and weak measurement.
How to measure
This is the uncomfortable part. If traffic is no longer the main event, your dashboard has to change.
For first-marketing leaders, the better set of signals is smaller and more honest:
- Branded search growth
- Direct traffic trend
- Referral quality from walled gardens
- Sales conversations started from specific content
- Mentions, citations, and inbound replies
- Assisted pipeline from founder-led distribution
That does not mean abandoning analytics. It means accepting that early content is often a compounding asset, not a last-click channel.
SEO is dead
No. But the old version of SEO is.
The old version assumed that ranking created a reliable visit, and that visit could be treated as the main unit of value. Zero-click search breaks that assumption. Search now behaves more like a visibility layer, part search engine, part answer engine, part recommender system.
If your business is young, that is not bad news. It simply means your first content strategy should optimize for three things at once, citation, credibility, and distribution.
First customers
When you have zero customers, the goal is not to dominate Google. The goal is to be found by the right ten people, and to be remembered by them.
That usually means a mix of:
- One strong point of view on your market
- One or two proof pieces that show you understand the problem deeply
- One distribution channel you can repeat weekly
- One search page that captures high-intent demand
That is enough to start.
Better marketing for an early-stage company is not about volume. It is about trust, clarity, and repetition in the few places that still matter.
Takeaway
The new zero-click reality does not punish founders who publish. It punishes founders who publish without a distribution plan.
If you are early, treat search as one part of a broader visibility system. Publish so you can be cited. Borrow audiences so you can be seen. Keep SEO where intent is real. Measure less by traffic, more by market response.
That is the practical answer to the 2026 data, SEO is not dead, but it is no longer the center of the plan.