Google Just Admitted It: Ranking #1 Doesn't Matter Anymore (Here's What Does)

Bruno Gavino - Codedesign.org
June 10, 2026

I almost didn't write about this one, because honestly, I've been writing some version of "AI is changing search" for two years now and I know how tired that sentence sounds. But on June 5th, Google did something it doesn't usually do: it updated its own Search Central documentation to officially name Generative Engine Optimization (GEO) and Answer Engine Optimization (AEO) as legitimate SEO disciplines.

Read that again. Google — the company that spent a decade telling us "just focus on quality content and rankings will follow" — just put it in writing that ranking and being chosen by AI are two different games now. Not "might become" different. Are different, today, in June 2026.

And the data backs it up in a way that genuinely stopped me mid-coffee this morning. Back in mid-2025, if you ranked in the top 10 on Google, you had a 76% chance of being cited inside an AI Overview for that query. By early 2026, that number had collapsed to roughly 38%. In less than a year, the link between "ranking well" and "being the source the AI quotes" went from a strong relationship to barely a coin flip.

That's not a tweak. That's a decoupling. And it's the topic I want to dig into properly today, because I think most Marketing Directors are still reading their analytics through a 2023 lens while the underlying machine has been quietly rewired underneath them.

Why this is a 1-3 month problem, not a "someday" problem

I'll be direct about why I picked this topic over the dozen other things I could have written about this week — ChatGPT opening ad inventory to any US business, Meta's ad revenue projected to overtake Google's for the first time ever, Google Marketing Live's agentic ads push. All fascinating. None of them will hit your traffic numbers as fast as this will.

AI Overviews are now showing up on something like 48% of all Google queries, up from 34.5% just six months ago. When an AI Overview appears above the organic results, the old #1 spot loses around 18% of its clicks on average. For purely informational queries — the "what is," "how to," "why does" content that most B2B and service businesses have spent years building — some sites are seeing 30-40% traffic drops on those pages.

If you run a content team, a blog, or a resource hub that was built around informational keyword targeting, this is already happening to you. You might just not have connected the dots yet because your rankings haven't moved. That's the trap — your position in Google Search Console can look completely stable while your actual clicks quietly bleed out, because the AI Overview is sitting above you eating the click before the user ever scrolls to position one.

A real example from one of our clients

I want to ground this in something real rather than just throwing statistics at you, because that's not how I like to write and it's definitely not how I like to read.

One of our clients — a mid-sized B2B software company in the compliance space, the kind of business where "what is SOC 2 compliance" or "how does GDPR apply to SaaS companies" are the bread-and-butter informational queries that have driven their top-of-funnel for years — came to us in April genuinely confused. Their core informational pages, the ones that used to bring in a steady stream of organic traffic, had dropped between 28% and 35% over about ten weeks. Rankings? Practically unchanged. Position 2, position 3, sometimes even position 1.

Their internal team had already ruled out the obvious culprits: no manual penalty, no technical issues, no major content changes on their end. Which is exactly the kind of "everything looks fine but the numbers are wrong" situation that drives marketing teams quietly insane.

When we pulled the queries and checked them manually, the pattern was obvious within about twenty minutes. Almost every query that had dropped now triggered an AI Overview — and our client's content was nowhere in the citations, even on queries where they ranked #1 or #2 in the traditional results below it. Meanwhile, three of their direct competitors — none of whom outranked them — were showing up consistently as cited sources inside those AI Overviews.

So what were those competitors doing differently? We went through their cited pages line by line, and the pattern was consistent: tightly structured FAQ sections with direct, self-contained answers near the top of the page, proper FAQ schema markup, and — this is the part that surprised even me — much more consistent factual framing across multiple pages on their own site. Same definitions, same numbers, same phrasing for key concepts, repeated consistently rather than rephrased "for SEO variety" the way we used to recommend.

It turns out that "say the same true thing the same way in multiple places" is now a ranking signal for AI citation, in a way it never was for traditional SEO. For years we taught writers to vary their phrasing to avoid looking repetitive. Generative engines apparently read that variation as inconsistency, and inconsistency reads as "this source isn't confident about its own facts."

We rebuilt the client's top 15 informational pages around this — restructured for direct-answer passages, added FAQ schema across the board, and did a full consistency pass on how we define and describe their core concepts site-wide. It's been five weeks. Citations in AI Overviews for their target queries have gone from essentially zero to appearing on about 40% of the queries we tracked, and organic clicks on those pages have recovered roughly two-thirds of what they'd lost. Not fully back, but trending the right way, and trending fast for an SEO change.

What I've been reading, watching, and listening to on this

I don't think any single source has this fully figured out yet — including us, to be honest — but a few things I've consumed recently shaped how I'm thinking about it.

I went back and reread our own piece from earlier this year, "The March 2026 Google Update Changed Everything: Why Your AI-Generated Content Strategy Might Be Failing". We wrote that one before this June documentation update, and rereading it now, it almost reads like a prediction. The core argument — that volume of AI-generated content was becoming a liability rather than an asset — looks even more true today than it did then.

I'd also point you to "Google Didn't Kill AI Content—You're Just Using It Wrong", which is the more optimistic companion piece. The short version: AI content isn't the problem, undifferentiated AI content that says the same generic thing as everyone else's AI content is the problem. That's even more relevant now that "saying things consistently" is apparently a citation signal — consistency only helps you if what you're being consistent about is actually distinct from your competitors.

On the podcast side, I caught up on a couple of episodes of Voice of Experts — if you haven't subscribed yet, this is genuinely where I get a lot of my early signal on this stuff, partly because, full disclosure, our team contributes to it, but mostly because the guest lineup skews toward people actually running campaigns rather than people selling courses about running campaigns. The recurring theme across the recent episodes has been exactly this GEO/AEO shift — multiple guests independently describing the same "rankings stable, traffic down" symptom my client experienced, which told me this isn't an isolated case, it's a pattern hitting a lot of businesses at once, right now, in roughly the same window.

I also listened to a good chunk of a recent episode of Lenny's Podcast that touched on how product and growth teams at larger companies are restructuring their content operations around "answer-first" formats rather than article-first formats — basically treating every page as if it needs to survive being chopped down to a two-sentence AI summary. That framing stuck with me: write every page as if the AI Overview is going to extract exactly one paragraph from it, and ask whether that paragraph, on its own, with no other context, is actually a good, accurate, complete-feeling answer.

What I'd actually do about this in the next 90 days

I'm not going to pretend there's a silver bullet, and anyone promising you a guaranteed citation count is, as Google itself basically said in that same June 5th update, making a claim nobody can actually back up from outside their systems. But here's the realistic, prioritized version of what we're doing for clients right now:

• First, audit your top 20-30 informational pages specifically for AI Overview presence on their target queries — not just rankings. This takes a few hours manually or you can use one of the newer GEO tracking tools, but either way, do it this month. You need a baseline before you can know if changes are working.
• Second, restructure your highest-traffic informational pages so the first 100-150 words after the H1 directly and completely answer the core question, written as if it will be lifted out and shown with zero surrounding context. Save the nuance, caveats, and depth for below that.
• Third, add FAQ schema markup site-wide where you have genuine FAQ content — and if you don't have FAQ sections on your key pages, this is the month to add them.
• Fourth — and this is the one most teams will skip because it's tedious — do a consistency audit across your site. Pick your 10-15 most important concepts, definitions, or claims, and make sure you're describing them the same way everywhere, with the same numbers and the same framing. Kill the "vary it for SEO" instinct. It's now working against you.

None of this is exotic. It's not some secret GEO hack. It's mostly just good information architecture applied with a new lens. But the businesses that do it in the next quarter are going to be the ones showing up in AI Overviews when their competitors search for the same questions their customers are asking — and the businesses that wait will be explaining flat traffic with stable rankings to their boards sometime around Q4, wondering what changed.

One question for you

Here's what I'd genuinely like to know: have you checked whether your own informational pages are showing up inside AI Overviews for your target queries — not ranking, showing up as a cited source? If you haven't looked, pick your three most important "what is" or "how to" pages right now and search those exact queries. I'd be curious to hear what you find, especially if it surprises you the way it surprised my client.

And if you want a second pair of eyes on what you find, you know where to find us at Codedesign — or drop into a future episode discussion over at Voice of Experts, because I suspect this conversation is only getting started.

---