Google AI Overviews is a critical topic in SEO today. Google rolled out AI Overviews to all US users in May 2024. About a year later, the picture for publishers is clearer — and more nuanced than either the doom predictions or the “it’s fine” reassurances suggested.
Table of Contents
- Google AI Overviews: What AI Overviews actually do to organic traffic
- Which query types are safe — and which aren’t
- The citation opportunity most publishers are ignoring
- How to adjust your keyword strategy for an AI Overview world
- What hasn’t changed
- Frequently asked questions
- Related Articles
The short version: AI Overviews are real, they do suppress clicks for certain query types, and the traffic impact varies enormously depending on what kind of content you publish. Some publishers are down 20–40% on keywords where AI Overviews now dominate.
Others haven’t moved. A smaller group is actually getting more traffic because their content is being cited inside AI Overview boxes.
Here’s what actually happened — and what it means for an SEO blog being built right now. This is an important consideration when thinking about Google AI Overviews.
Google AI Overviews: What AI Overviews actually do to organic traffic
The core effect is simple: when Google generates an AI Overview for a query, it appears above organic results. Searchers who find what they need in the AI box don’t need to click through to any article. Click-through rates for the organic results below drop.
The data on how much CTR drops varies by study and query type. Research from Search Engine Land and Semrush found CTR drops ranging from 15% to over 60% for queries where AI Overviews appear. The widest drops happen for simple informational queries — definitional questions, factual lookups, basic how-to queries with short answers.
“What is keyword difficulty” is more vulnerable than “how to recover from a Google core update penalty” because the first has a self-contained answer and the second requires nuance. This is an important consideration when thinking about Google AI Overviews.
This means the keywords most threatened by AI Overviews are exactly the easy informational keywords that new SEO blogs used to chase first — low competition, clear answer, decent volume. Those still exist and still drive traffic, but the effective search volume is lower than the tools show. This is an important consideration when thinking about Google AI Overviews.
Which query types are safe — and which aren’t
After a year of data, a pattern has emerged around which content types hold traffic despite AI Overviews and which have lost ground.
Queries where AI Overviews hurt most
- Simple factual definitions (“what is X”)
- Basic calculations or conversions
- Single-step how-to queries (“how to add a meta description in WordPress”)
- Medical or legal symptoms with clear answers
- Short historical facts
For these queries, a short paragraph in an AI Overview answers the question completely. There’s no reason to click. Traffic drops are real here — not catastrophic for most sites since these were rarely your highest-value keywords anyway, but measurable.
When it comes to Google AI Overviews, these factors matter significantly.
Queries where clicks still happen
- Multi-step tutorials requiring sustained attention
- Tool comparisons and recommendations (commercial intent)
- Content requiring personal judgment or experience
- Anything time-sensitive or that changes frequently
- Queries with genuinely contested answers
- Specific troubleshooting (“why is my WordPress site slow after installing X”)
- Content with original data, surveys, or case studies
For these queries, the AI Overview often appears but can’t fully replace the article. A searcher who gets a three-sentence overview about how to do a content audit still needs to click through to see the actual process. The AI Overview becomes a preview — and if it’s good enough to intrigue rather than fully satisfy, it can actually increase click intent.
The citation opportunity most publishers are ignoring
Here’s what most analysis of AI Overviews misses: some publishers are getting more clicks because their content is being cited inside AI Overview boxes, not below them.
When it comes to Google AI Overviews, these factors matter significantly.
When Google generates an AI Overview, it pulls from specific sources and shows links to those sources inside the box — usually 3–5 articles listed in a sidebar or inline citation format. Being one of those cited sources delivers a different kind of traffic: users who’ve already had the topic introduced and are clicking specifically because they want depth.
The content that gets cited in AI Overviews shares characteristics with the March 2026 core update winners: original data, expert authorship, clear factual statements, and genuine depth. This isn’t a coincidence. Google’s AI systems and ranking systems both increasingly reward the same thing — content that contributes something genuinely new rather than restating what’s already available.
If you’re building an SEO blog right now, targeting AI Overview citations is a realistic goal. The strategy: publish content with specific statistics, clear definitions with nuanced caveats, and well-structured answers to complex questions. Content that looks like a reliable source — not just a well-optimized page — is what gets cited.
How to adjust your keyword strategy for an AI Overview world
The adjustment isn’t dramatic, but it’s real. Here’s what to do differently:
Check the SERP for AI Overview presence before committing to a keyword
Google the keyword. If an AI Overview appears and fully answers the query in 3–4 sentences, that keyword’s organic traffic potential is lower than the tool’s volume estimate suggests. Either move on to a related angle that doesn’t trigger an AI Overview, or reframe your content to go significantly deeper than what the AI Overview covers — making clicking through the obvious next step.
Prioritize query types that require decision-making
Queries where the searcher needs to make a choice — which tool, which approach, what to do after X happens — require judgment that an AI Overview can’t fully provide. “Best keyword research tool for a new site with no budget” is more defensible than “what is keyword research” in an AI Overview environment. The more specific the situation, the less likely an AI box fully covers it.
Build content that deserves to be cited
Structure every article to be a citable source. This means: include at least one specific, sourced statistic per major section. Write clear topic sentences that could stand as standalone definitions. Add a FAQ section with precise, well-bounded answers. Use schema markup (RankMath handles this automatically) so Google can parse your structure easily. Content built to be cited by AI systems and content built to rank on Google are increasingly the same thing.
What hasn’t changed
The fundamentals of SEO content haven’t changed. High-quality, specific, original content written for a defined audience with real expertise behind it still works — it’s just working in a more competitive environment where the floor for “good enough” has risen. The blogs that were already doing the right things before AI Overviews launched are the ones least affected by them.
One year in, AI Overviews are a feature of the landscape rather than an existential threat. Publishers who adjusted their keyword strategy to avoid fully-answered simple queries and invested in content depth have largely maintained or grown traffic. Publishers who kept producing thin content at volume have had a harder time — but that was already the direction the algorithm was heading before AI Overviews existed.
The lesson: AI Overviews accelerated an existing trend rather than creating a new one. Build content worth reading regardless of how it gets discovered, and the distribution channel — organic ranking, AI citation, social sharing — tends to take care of itself.
Frequently asked questions
Can I stop my content from appearing in AI Overviews?
Yes. Add the nosnippet meta robots tag to specific pages you don’t want cited in AI Overviews or featured snippets. Use RankMath’s per-post robots settings to add this. Note: using nosnippet may also remove your page from featured snippets, which can sometimes drive more traffic than AI Overview citations.
How do I know if AI Overviews are affecting my traffic?
Check Search Console for pages where impressions are holding steady but clicks and CTR are dropping. That pattern — visible in search results but getting fewer clicks — is the AI Overview footprint. Compare the current SERP for those keywords to what you see in older cached versions to confirm whether an AI Overview appeared.
Do AI Overviews appear for all searches?
No. Google is selective about when AI Overviews appear. They’re more common for informational queries and less common for commercial, navigational, and transactional queries. They also appear less frequently for time-sensitive topics, controversial subjects, and queries where Google’s systems aren’t confident enough in a synthesized answer.
For more information, see Google’s official AI Overviews documentation.