Codrva Digital

Zero Click Search Hit 68 Percent: How to Prove SEO Value Without Traffic

Zero click search reached 68 percent in early 2026. Your rankings improve and the traffic line still falls, which makes every SEO retainer harder to defend. Here is what changed and how to prove value without leaning on sessions.

C
Codrva Team
Published Jul 30, 2026
9 min read
A search results page dominated by an AI Overview with the organic listings pushed below the fold

A search results page dominated by an AI Overview with the organic listings pushed below the fold

Zero click search reached 68 percent in early 2026, and it broke the way most people justify SEO. You can move a client from position nine to position two, watch the ranking chart turn green, and still open analytics to find sessions flat or falling. Nothing went wrong with the work. The result simply stopped arriving as a click.

If you sell SEO, or buy it, this is now the central problem. The metric everyone agreed to be judged on has quietly detached from the outcome it used to predict. Pretending otherwise is how retainers get cancelled.

What zero click search actually means now

The headline zero click search number understates it. Across all Google searches the zero click rate sits near 68 percent. When an AI Overview appears above the results, it climbs to about 83 percent. Inside AI Mode it reaches roughly 93 percent. Fewer than a third of Google searches now send a click to any website at all.

Users clicked a traditional organic result about 8 percent of the time when an AI Overview was present, against roughly 15 percent when it was not. That is close to half the click through rate, for the same position, on the same query. Publisher traffic from Google has already fallen by about a third.

The important detail is that this is not evenly spread. Informational queries got hit hardest, because a definition or a how to answer is exactly what a generated summary handles well. Commercial and transactional queries still send clicks, because someone comparing prices or ready to book wants to land on a real page.

Why your rankings still went up

Here is the part that confuses clients. Your ranking improvements were real. Position two is genuinely better than position nine. What changed is what sits above position two.

An AI Overview occupies the space your snippet used to own, and it frequently draws its answer from the same page you optimised. You earned the citation. You just did not earn the visit, because the searcher got what they needed without moving. In a zero click search environment the ranking is now an input, not an outcome.

This is why a report built on average position tells a story that quietly stopped being true. The number goes up, the business feels nothing, and the client concludes somebody is spinning them.

Read More -  WordPress Security: Hacked Through a Plugin You Forgot You Installed

The metrics that still mean something

Stop leading with sessions and position. Lead with things a business owner can feel:

  • Qualified leads and calls. Form fills, phone calls, bookings. If SEO is working, these move even when sessions do not.
  • Branded search volume. When more people search your client by name, awareness is compounding. AI Overviews often drive this without a click.
  • Citation presence. Whether the client is named inside AI Overviews, ChatGPT, and Perplexity answers for their key topics.
  • Conversion rate of the traffic that does arrive. Fewer, better visitors is a real outcome, not a consolation prize.
  • Commercial query performance, split out. Separate informational from commercial. The informational decline is structural. The commercial line is where your money is.

That last split is the single most useful change you can make. Reporting them together lets a structural industry decline mask genuine commercial progress, and you end up apologising for a trend nobody could have prevented.

What to do with the informational content you already own

Most sites are sitting on a library of explainer posts built for a search economy that zero click search has already replaced. Deleting it is the wrong reflex. That content still does three jobs, and only one of them was ever traffic.

It feeds the summaries. If a generated answer is going to be assembled from somewhere, you want it assembled from your pages, with your brand named as the source. A searcher who reads a summary citing your client three times has been marketed to, even though analytics recorded nothing.

It also serves the people already in your funnel. Prospects mid evaluation, existing customers looking something up, and sales teams sending links all use that library. None of that shows as new organic sessions and all of it affects revenue.

What it should stop doing is absorbing your production budget. If you are still publishing weekly explainers on questions a summary answers instantly, you are funding a channel that has structurally changed. Redirect that effort toward comparison pages, pricing, case studies, and tools, then keep the existing library accurate so it stays citation worthy. Our post on publishing frequency covers how to think about cadence once volume alone stops working.

The measurement problem nobody has solved

Be honest that zero click search is genuinely harder to measure than the old model, because pretending otherwise sets you up for an awkward quarter. There is no clean dashboard showing how many people saw a client named inside an AI Overview and later searched for them directly.

What you can do is triangulate. Track branded search volume over time, watch direct traffic alongside AI referral patterns, monitor citation presence across the main engines, and correlate all three against lead volume. No single number proves it on its own. The pattern across all four of them together does. Agencies that set this up early are the ones who can still defend a retainer in a zero click search world, and those that cannot will keep arguing about sessions until the client leaves.

AI referrals convert better than you expect

There is a real upside buried in this. Referral traffic from AI engines is growing quickly and converts at noticeably higher rates than classic organic. It makes sense: someone who reads a generated comparison and then deliberately clicks through to one named vendor is much further along than someone browsing ten blue links.

The catch is that this traffic is hard to see. Many AI referrals arrive with no referrer or get bucketed as direct, so the channel that is growing looks like noise in a standard report. If you are not deliberately measuring it, you will underreport the one thing that is going right.

Read More -  Your Team Loses Two Hours a Day to Manual Data Entry

How to get cited instead of summarised

You cannot opt out of AI Overviews, but you can change your odds of being the source they quote:

  1. Answer the question in the first hundred words. Generated summaries pull from passages that resolve the query directly. Long preambles get skipped.
  2. Use clear question style headings. An H2 phrased as the actual question makes the passage easy to lift and attribute.
  3. Add real structured data. FAQPage and other schema help machines understand which passage answers what.
  4. Publish things a summary cannot replace. Original data, pricing, first hand experience, and specifics. A model can rephrase generic advice; it cannot invent your numbers.
  5. Go after commercial intent harder. That is where clicks still live, so shift effort there rather than defending informational rankings that no longer pay.

The strategic shift underneath all five is accepting that the goal moved from "rank for the query" to "be the source the answer trusts." Traffic follows for the commercial subset, and brand awareness accrues from the rest.

Reporting on this needs tooling that tracks AI visibility alongside rankings and ads, which is exactly why we built AgencyReportr. If you need custom tracking or data pipelines behind it, our AI tools development team builds them.

Frequently Asked Questions

Is SEO still worth investing in with zero click search at 68 percent?

Yes, but the value has moved. Commercial and transactional queries still send clicks and still convert, and AI referral traffic converts better than classic organic. What no longer pays is chasing informational rankings purely for session volume. The investment case now rests on qualified leads and brand presence rather than raw traffic.

How do I explain falling traffic to a client whose rankings improved?

Show the mechanism, not the excuse. Put the AI Overview presence rate for their queries next to the click through rate change, and split informational from commercial performance. When a client can see that the informational decline is industry wide while their commercial line held or grew, the conversation shifts from blame to strategy.

What is the difference between zero click search and AI Overviews?

Zero click search is the broader behaviour where a search ends without a visit to any website, which also happens with featured snippets, knowledge panels, and local packs. AI Overviews are one specific driver, and the most aggressive one, pushing the zero click rate from roughly 68 percent overall to about 83 percent on queries where they appear.

Can I stop Google using my content in AI Overviews?

Only by blocking Google from indexing the content, which removes you from search entirely. There is no partial opt out that keeps rankings but excludes summaries. For nearly every business that trade is far worse than the click loss, so the practical strategy is optimising to be the cited source instead.

Which content types are least affected?

Commercial and transactional pages hold up best, along with anything requiring interaction, personalisation, or a real transaction. Content that has held value includes original research, pricing, tools, calculators, and genuine first hand experience. Generic explainer content has been hit hardest because it is the easiest thing to summarise.

How do I measure whether AI engines are citing my client?

You need monitoring that queries the AI engines for the client's key topics and records whether the brand or its pages appear in the generated answer. There is no position to track, only presence or absence, so it works more like share of voice than rank tracking. Very few agencies report on this yet.

Should I still report on rankings at all?

Keep them as a diagnostic, not the headline. Rankings still explain why other numbers moved and remain useful for spotting technical problems or competitor gains. They just no longer prove business value on their own, so they belong in the supporting detail rather than the summary slide.

If your traffic fell and you are not sure whether it was an algorithm change or the AI shift, our posts on recovering from a core update and thin AI content demotions cover how to tell the difference. Talk to our team if you want help rebuilding measurement around what still converts.

Share this post:
← Back to Blog

Comments (0)

Leave a Comment

Minimum 10 characters

No comments yet. Be the first to comment!

Chat with us