SEM Rush: If You Get In The “Top 10” For SEO, You Probably Will Be There in AI Search, Too. But Differently.

If SEO were a movie, this would be the scene where everyone yells, “Wait… the villain is actually a plot twist.” AI search hasn’t killed SEO. It has changed what “winning” looks like.

The short version: if your site consistently earns strong rankings in Google’s classic top 10, you’re still in a very good position for AI visibility. But AI search engines (and AI-powered search features) don’t always cite the exact same URL you ranked with. They often pull from the same domain, but from a different pagesometimes a deeper article, a help doc, a comparison page, or a niche FAQ that never got to page one.

That’s the “same, but different” part. And it matters a lot for publishers, SaaS teams, ecommerce brands, and any business that depends on organic discovery.

This article synthesizes insights from a mix of reputable U.S. sources and platforms, including SEMrush, Google Search Central, Google Search Help, Google’s official Search blog, Microsoft Advertising and Bing ecosystem guidance, Similarweb, Pew Research Center, SaaStr, Search Engine Land, and Search Engine Journal reporting and analysis. No fluff, no panic, no “SEO is dead” clickbait funeral slideshow. Just what changedand what to do next.

What the SEMrush Study Actually Suggests (and Why People Are Talking About It)

The headline idea comes from a practical observation that many marketers are now seeing in the wild: strong SEO performance still correlates with strong AI search visibility. SEMrush’s AI Mode comparison study gave that idea structure and numbers.

In that study, SEMrush analyzed 5,000 keywords across major intent categories and compared Google traditional search, Google AI experiences, and leading AI search platforms. The dataset included more than 150,000 citations. That’s large enough to move the conversation beyond “my friend’s site got fewer clicks last week, so the internet is over.”

The big takeaway wasn’t that AI search ignores classic rankings. It was that AI search partly overlaps with classic rankingsespecially at the domain levelwhile often citing different URLs.

The key shift: domain authority still matters, URL selection is more fluid

Think of classic SEO as winning shelf space in a supermarket aisle. AI search is more like being selected by a personal shopper. If your brand is trusted, you’ll likely make the cart more often. But the shopper may choose your smaller bottle, travel-size version, or “advanced formula” page instead of the exact product you expected.

In practice, this means:

• Your domain-level strength (trust, breadth, relevance, crawlability) still helps a lot.
• Your top-performing “money page” may not be the page AI systems cite.
• Deep, specific pages can suddenly become stars, even if they are not top-10 organic winners.

Why AI Search Feels Different Even When SEO Fundamentals Still Work

Google itself has been unusually clear on one point: the best practices for SEO still apply to AI features like AI Overviews and AI Mode. In other words, there isn’t a magical “AI schema” or secret file format you need to upload at midnight under a full moon.

Google’s documentation says there are no additional requirements to appear in AI Overviews or AI Mode beyond being indexed and eligible to appear in Search with a snippet. That’s important because it cuts through a lot of noise. If someone is trying to sell you an “AI-search-only metadata hack,” keep one hand on your wallet.

So why do the results look so different?

Because the retrieval behavior is different.

Google has explained that AI Mode may use a “query fan-out” approachbreaking a question into subtopics and running multiple related searches. That means AI systems are not always looking for one perfect page; they are assembling support from multiple pages and multiple angles. A classic rank #2 page might help, but so might your buried comparison chart, glossary entry, or troubleshooting guide.

This is where many teams get confused: they see lower click-through rates on some old SEO pages and assume their authority is gone. In reality, their authority may still be workingbut the traffic is being redistributed across pages and formats, and some of the answer consumption is now happening on the SERP itself.

Top 10 SEO Still MattersBut “Winning” in AI Search Is More Like Earning Citations

Classic SEO trained us to obsess over rank position. AI search adds another layer: citation eligibility and selection.

In ranked search, the user sees a stack of links and chooses one. In AI search, the system chooses which sources support the generated answer, and the user may click oneor none. That changes content strategy in three important ways:

1) Breadth beats one-page heroics

If AI tools cite different pages from the same domain, a site with broad, high-quality coverage often has an advantage over a site with one viral article and a lot of thin content around it. This aligns with the SaaStr-style interpretation of the SEMrush findings: classic SEO strength still matters, but breadth and depth matter more than many teams realized.

Translation: your content strategy cannot be “one big pillar page and vibes.”

2) Mid-ranking and deep pages can become AI citation magnets

SEMrush and related industry discussions point to a pattern many marketers are now noticing: AI systems often cite pages beyond the obvious top 10, especially when those pages answer a narrow question better. That means your “supporting” content is no longer just internal-link fuel. It can be front-line discovery content in AI search.

Examples of pages that often become unexpectedly useful in AI retrieval:

• Comparison pages (“X vs. Y for small teams”)
• Pricing explainers and ROI calculators
• How-to troubleshooting articles
• Glossaries and definitions with examples
• FAQ pages written in plain English
• Product documentation and setup guides

3) Brand trust plus specificity is a powerful combination

AI systems often need both trust and precision. A known domain with a page that directly matches a sub-question is a great combination. This helps explain why strong brands still appear frequently, but also why their specific cited pages may look unfamiliar to the SEO team watching only a handful of target keywords.

What This Means for Your SEO Strategy in 2026

Good news: you do not need to throw away your SEO playbook. Better news: you do need to upgrade it.

Keep the SEO fundamentals (seriously)

Google and Microsoft guidance both reinforce the same general truth: crawlability, metadata, internal linking, content quality, authority signals, and site structure still matter. AI systems cannot cite what they cannot reliably find, parse, or trust.

Your first checklist is still boringand that’s exactly why it works:

• Ensure important pages are indexable and not blocked accidentally.
• Improve internal linking to deep, high-value pages.
• Keep content current (especially stats, pricing, workflows, and product details).
• Use clear headings and chunked sections that are easy to parse.
• Make answers explicit before getting fancy.
• Support key claims with evidence and examples.

Build for citation selection, not just rank position

Ask a new question during content planning: “Which paragraph or section would an AI system quote or summarize?”

This changes how you write. Strong AI-citable content tends to have:

• Direct answers near the top of sections
• Clean comparisons and structured lists
• Specific numbers, ranges, caveats, and definitions
• Topic-focused pages instead of vague “everything pages”
• Context that explains when advice applies (and when it doesn’t)

In other words: write like a helpful expert, not like a keyword blender.

Invest in content coverage across the whole journey

AI search often handles multi-step, comparison-heavy, exploratory queries. That means your content should cover the full decision path, not just top-of-funnel basics.

For example, a B2B software company should not stop at “What is project management software?” It should also publish:

• “Best project management software for agencies under 20 employees”
• “Monday vs. Asana vs. ClickUp for client work”
• “How to migrate tasks from spreadsheets to a PM tool”
• “PM software pricing tiers explained (with hidden costs)”
• “Common implementation mistakes and how to avoid them”

Classic SEO may rank one or two of these. AI search may cite all five across different conversations.

Measurement Is Getting Trickier (But Not Impossible)

This is where many teams start stress-refreshing dashboards.

Google’s Search Console documentation and related industry reporting now clarify that AI Mode clicks, impressions, and positions are counted in Search Console performance reporting (within the Web search type). However, marketers still face a practical challenge: AI-generated exposure is not always neatly separated the way people want.

So yes, measurement is improving. And yes, it is still messy.

How to measure “AI search impact” without losing your mind

• Track blended organic performance: impressions, clicks, CTR, and conversions at page level.
• Watch deep-page growth: pages previously ignored may start getting meaningful traffic.
• Measure quality, not only volume: session depth, time on page, assisted conversions, and lead quality matter.
• Segment by query intent: informational pages may lose clicks while commercial pages gain better-qualified visits.
• Monitor citations manually for core topics: especially in Google AI Mode, AI Overviews, ChatGPT search, and Perplexity.

Google has also noted that clicks from AI-enhanced search experiences can be higher quality in some cases. That lines up with what a lot of practitioners are reporting: fewer clicks, but sometimes better clicks.

The Hard Truth: AI Search Can Reduce Clicks Even While It Increases Visibility

Here’s the part nobody loves hearing at the quarterly meeting: visibility and traffic are no longer the same thing.

Pew Research Center analysis found that users encountering Google AI summaries were less likely to click traditional links than users who did not see a summary. That doesn’t mean your brand lost visibility; it means some user intent is being satisfied before the click happens.

At the same time, the broader AI referral ecosystem is growing fast. Similarweb reported AI referral traffic rising sharply year over year, but still far smaller than Google Search’s total referral volume. So the market is changing quickly, but classic search remains enormous.

The smart takeaway is not “SEO is dead” or “AI will replace everything by Tuesday.” The smart takeaway is:

• SEO remains foundational.
• AI search visibility is increasingly layered on top of SEO authority.
• Traffic patterns will shift, fragment, and become more page-specific.
• Brands that publish broad, trustworthy, structured content will adapt faster.

A Practical Playbook for Ranking in SEO and Showing Up in AI Search

1) Protect your technical foundation

Fix crawl blocks, indexation issues, broken internal links, and slow pages. AI retrieval cannot rescue a broken site architecture.

2) Expand topic coverage by intent and depth

Create content clusters that cover beginner, comparison, implementation, troubleshooting, and decision-stage questionsnot just broad definitions.

3) Write “answer-first” sections

Use crisp subheadings and direct answers in the first 1–3 sentences of each section. Then add detail, examples, and caveats.

4) Build citation-friendly assets

Original research, benchmark summaries, calculators, glossaries, expert explainers, templates, and documentation-style pages are increasingly valuable.

5) Update more, rewrite less

AI systems and users both reward freshness. Updating strong pages with current stats, examples, screenshots, and FAQs often beats publishing ten thin new posts.

6) Broaden your off-site visibility

SEMrush-style findings and industry observations show user-generated and community platforms (especially discussion-heavy ones) play a bigger role in AI citation ecosystems. If your brand has no credible presence where your audience discusses problems, you are leaving visibility on the table.

Experience Notes from the Field (500+ Words): What Teams Are Actually Seeing

The most common experience I hear from content and SEO teams right now sounds something like this: “Our rankings are mostly okay, but traffic is weird.” That sentence is doing a lot of work.

What “weird” usually means is not one dramatic crash across the entire site. It means a patchwork of changes. A few high-volume informational pages lose clicks. A handful of deep pages suddenly pick up traffic and conversions. Brand queries stay strong. Comparison pages become more valuable. Stakeholders panic for two days, then calm down when lead volume remains stable.

For example, a SaaS team might see its old “What is CRM?” article flatten out while “CRM for real estate teams,” “CRM migration checklist,” and “HubSpot vs. Pipedrive for small teams” start punching above their historical weight. In classic SEO reporting, that can look like a mixed bag. In AI search terms, it often means the site is being used more as a source of specific support rather than a destination for broad definitions.

Ecommerce and affiliate-heavy publishers are reporting another pattern: roundup pages still matter, but supporting content matters more than before. A buyer’s guide may remain the front door, while AI systems cite the “how to choose,” “size guide,” “care instructions,” or “best for [use case]” pages during follow-up questions. If those supporting pages are thin, outdated, or generic, the site loses a chance to stay in the conversation.

There is also a measurement psychology issue happening inside teams. Many organizations still evaluate content success with a page-one mindset: “Did this article rank #1 for the head term?” In AI search, the more useful question is often, “Did our domain show up across the decision journey, and did the user eventually convert?” The path can be less linear now. A user may see your brand in an AI summary, click later from a comparison query, return through branded search, and convert on a product page. If your reporting model only rewards the first visible click, your team may under-invest in the exact content that is building AI visibility.

Another real-world experience: content teams that write in dense, abstract language struggle more than they expect. Pages that sound smart but fail to answer concrete questions get skipped. Meanwhile, “boring but clear” pages win citations. This is not a downgrade in qualityit is a reminder that clarity is a competitive advantage. The internet’s new valedictorian may not be the most poetic page; it may be the page that defines the term, compares the options, lists the tradeoffs, and explains the next step without making the reader feel like they need a decoder ring.

Finally, the teams adapting fastest are usually the ones that stopped arguing about whether to call it SEO, GEO, AIO, or “AI search optimization” and just got practical. They audit what pages are being cited, improve internal links to deep content, refresh outdated guides, add better examples, and publish more decision-stage pages. The terminology debate is fun on social media. Revenue prefers execution.

Conclusion

If you’re strong in classic SEO, you are not starting from zero in AI search. You’re starting from an advantage. The catch is that AI search doesn’t reward only your top-ranking URLs; it often rewards your domain’s overall credibility and the depth of your content library.

So yes: if you get in the top 10 for SEO, you probably will show up in AI search, too. But differently. Less like a ranking trophy, more like a citation economy. Less “one keyword, one page,” more “one trusted domain, many useful answers.”

That’s not the end of SEO. It’s the next version of it.