# Zero-click search is 69% of queries. What that means for a launch

> About 69% of searches now end without a click. What zero-click search means for a launch week, and where the first hundred visitors come from instead.

Published 2026-09-05 · 6 min read · zero-click-search, launches, ai-overviews, indie-makers
Canonical: https://bidsurvivor.space/blog/zero-click-search-and-your-launch

---

Roughly **69%** of Google searches now end without a click on anything, according to the figures [Search Engine Land](https://searchengineland.com/google-ai-overviews-cut-search-clicks-report-471497) summarised this month. That is the number behind the phrase zero-click search, and if you shipped something this week and are asking where the users are supposed to come from, it is the number that answers you: mostly, not from the results page. Not this week, and probably not this year.

That is not a reason to panic. A launch never ran on search anyway. It is a reason to stop pretending the search plan is a launch plan, and to spend the first week on the channels that still send a person to a page.

## What zero-click actually measures

A zero-click search is a query where the person got what they wanted from the results page itself and left without visiting a site. Some of that is old: the weather, a currency conversion, the opening hours of a shop. What changed is the AI Overview, which now sits above the results for a large share of informational queries and answers them in a paragraph, with a few citations most people never open.

The reported effect is stark. On queries where an Overview appears, click-through to the pages beneath it is down **61%**. Across the publishers [Press Gazette](https://pressgazette.co.uk/publishers/search-isnt-dead-its-fragmenting-how-to-manage-google-traffic-decline/) spoke to, search clicks are down 42% from their baseline. And the share of all queries that end with no click at all is now around 69%.

Two caveats worth carrying. These are publisher numbers, measured mostly on informational queries, which is where Overviews are densest. A search for your product's name still sends a click, because there is nothing for a model to summarise instead of you. And "zero-click" counts an impression that never became a visit, so it overstates the loss for anyone who never had the impression in the first place — which is every site launched this week.

## Why it barely touches launch week, and why it wrecks month three

Here is the part people get backwards. Zero-click search is nearly irrelevant to the launch itself, and very relevant to what you were planning to do afterwards.

A launch runs on channels that were never search: the people who already know you, the communities you posted in, a directory listing or two, whatever you paid for. Our [launch-day plan](/blog/how-to-get-traffic-on-launch-day) stacks those into a single day, and none of them involve ranking for anything. A new domain does not rank in its first week regardless of what Google does with Overviews. So the 69% costs your launch nothing it ever had.

What it costs is the plan you had for month three. The usual indie sequence is: launch, get a bump, write twenty articles, wait for organic search to take over. That third step used to be the reliable one — slow, but compounding. Now the articles you write will be read by a model and summarised above the fold, and a large part of the click you would have earned goes to the summary. The compounding still happens; it compounds at a much lower rate. The channel you were going to coast on has become a channel you can still use but cannot rely on. The [post on where small sites get traffic now](/blog/where-small-sites-get-traffic-after-ai-overviews) goes through the alternatives channel by channel.

## The first hundred visitors, without a blue link

So the question for a launch week is the older one: where do the first hundred people come from when nobody is searching for you yet? We have written about the [five sources that reliably produce them](/blog/get-your-first-100-website-visitors), and the ranking has not changed much. What has changed is how much weight the last two deserve.

**People you can name.** Everyone you have told, everyone who replied to a build-in-public post, the twelve people who joined the waitlist. This is the highest-converting channel a launch will ever have, and the smallest. Use all of it on day one.

**The rooms your buyer is already in.** A forum thread, a Discord, a subreddit, a newsletter that covers your niche. Answering a real question with a link is still the best free channel there is, and it is untouched by Overviews, because it was never search traffic.

**Directories and launch sites.** They send a spike and then nothing. Take the spike; do not build a plan on it.

**Bought placement, priced by the hour.** This is the one whose weight has gone up, because it is the one that does not depend on a results page at all. If a search click costs what it did but arrives 61% less often on the queries you wanted, a visitor from a placement you paid a fixed price for got relatively cheaper without changing price. The arithmetic is in the [paid-versus-organic post](/blog/paid-vs-organic-when-search-sends-fewer-clicks).

**Your own list.** Whatever you collect this week is the only channel nobody else can change the terms of. Start it now, even if it is nine addresses.

## The literal answer to "where do users come from"

If you posted this week that you shipped and have no idea where users come from, here is a literal, unglamorous answer, including one option that is ours.

Post in three places where the question your product answers gets asked. Email everyone who ever expressed interest. Add the site to two directories. And if you want a bought placement to compare against those, the cheapest one we know of is the one we run: [BidSurvivor](/) sells two twelve-hour slots a day on a single page, and the first brand into an empty slot pays **$0**. No account, no card, one free claim per bidder per day. Taking a held slot means beating the standing bid, from $1.00 up to a $5,000 ceiling. Every visitor starts with $100 of house credit before signing up, and a bid that fails to beat the standing bid costs nothing. If somebody outbids you later you lose the slot and the money — no refund. The slot's opens and click-throughs are published on the front page and on your brand's own page, so you can see what it sent before you decide whether to try again.

We should be plain about the scale. The board is young and its traffic is small; the numbers are live on the front page, and you should read them before you assume anything. It is not a replacement for search. It is one honest, priced channel that does not care whether an AI Overview exists, alongside four free ones that do not either.

The 69% is real. It just is not your problem yet. Your problem is the first hundred people, and the ways to reach them have been the same for years: talk to people directly, show up where they already are, and pay a known price for a known slot when you want a number to compare against. The results page was always the slow channel. It has just got slower.