
Table of Contents
- What the Data Actually Shows
- Why Definition Queries Are Exposed
- What to Do About It
- What Not to Do
- Frequently Asked Questions
- Do AI Overviews reduce clicks to websites?
- Which kinds of pages lose the most traffic?
- Should I stop writing "what is" posts?
- What did Google change to send traffic back to publishers?
- How do I know if AI Overviews are affecting my site?
Google's AI Overviews are answering a growing share of searches before anyone clicks a result, and the click data shows it. An Ahrefs study published in February 2026 found that AI Overviews correlate with a 58% reduction in click-through rate for top-ranking pages, up from a 34.5% decline documented in April 2025. If your traffic comes from "what is X" queries, this is the change most likely to show up in your Search Console graph.
We publish plenty of definition-style posts ourselves, so this is not an outside observation: it applies to us too. Here is what the numbers say, what they do not say, and what to do about it.
What the Data Actually Shows
The 58% figure comes from an Ahrefs study reported by The Next Web in May 2026. It is a correlation, not a controlled experiment: the presence of an AI Overview goes along with a much lower click-through rate for the pages ranking at the top. The study does not prove that every site loses 58% of its traffic, and it does not break the effect down by query type.
A second data point points the same direction. Pew Research Center data cited in the same coverage found that users clicked a result 8% of the time when an AI Overview appeared, versus 15% when it did not.
Google's own response has been to add more ways for the summary to point outward. It announced five updates to AI Overviews and AI Mode: a "Further Exploration" links section, inline link context that shows a site or page title on hover, subscription labels for publications a user already pays for, previews from forums and social media with creator attribution, and product review cards for shopping queries. Whether those changes recover meaningful traffic is an open question, and nobody outside Google has published numbers on it yet.
Why Definition Queries Are Exposed
The data above does not split results by query type, so what follows is reasoning, not a measured finding. But the logic is hard to avoid. An AI summary can fully satisfy a query when the question has a short, stable, factual answer: what an acronym stands for, what a term means, what a feature is called. A query like "what does UGC stand for" is finished the moment the answer appears.
Queries with more at stake, such as comparing tools, choosing between strategies, or getting a template, give a summary less to work with. The searcher needs something to look at, download, or decide with, and a paragraph does not replace that.
If your content plan has leaned on definitional keywords because they are low competition, the same qualities that made them easy to rank for also make them the easiest for a summary to absorb. That does not make them worthless. It changes what they are for.
What to Do About It
1. Keep the direct answer, then give people a reason to stay. Answering the query in the first 50 words still helps you get cited and helps readers. But treat that answer as the entry point, not the whole post. Put something after it that a summary cannot reproduce: a worked example, a template, a comparison table, a decision framework, a screenshot of the real process.
2. Shift the mix toward queries with a next step. How-to guides, tool comparisons, templates, and "which should I choose" content give the reader something to act on. Our guides on how to do keyword research and the best keyword research tools in 2026 sit on that side of the line, and a definition post like what keyword research is works best when it leads into them.
3. Add first-hand evidence. AI summaries rewrite what is already on the web. Original numbers, real screenshots, and specific examples from your own work are the things they cannot fabricate and the things readers cannot get elsewhere. This is the same shift we covered in why authentic content is winning over AI-generated content.
4. Structure pages so they are easy to cite. Clear headings, a short answer near the top, and a proper FAQ section make it easier for a summary to attribute a claim to you. Being cited inside an AI Overview is worth something even when the click rate falls, because it puts your brand in front of the searcher at the moment they decide who to trust.
5. Measure more than clicks. If a page's impressions hold steady while its clicks fall, the page is being seen and summarized, not abandoned. Track impressions, click-through rate by query, and branded search volume together. Our post on what social media analytics actually means and the guide to metrics that boardrooms care about cover how to report that picture without overreacting to one number.
6. Build channels you own. Search is one door among several. An email list, an active social presence, and a community give you traffic that does not depend on whether an AI summary shows up. Our guide to social search in 2026 covers the platforms where people are increasingly searching instead of using Google.
Traffic Dropped and You Are Not Sure Why?
We audit which pages are losing clicks to AI summaries and which are holding steady, then rebuild your content plan around queries that still earn visits.
What Not to Do
Do not delete your definition posts. They still earn impressions, citations, and internal-link value, and they are often the first page a new reader meets. Do not chase every AI-feature change either: Google's summaries are still evolving and the five updates above may not be the last round.
The practical goal is a portfolio. Some pages exist to be cited and to establish that you know the terminology. Others exist to earn the click because they give the reader something a summary cannot. Knowing which is which is the whole job.
Frequently Asked Questions
Do AI Overviews reduce clicks to websites?
Studies suggest they do. An Ahrefs study published in February 2026 found AI Overviews correlate with a 58% reduction in click-through rate for top-ranking pages, and Pew Research Center data found an 8% click rate with an AI Overview versus 15% without. These are correlations across many queries, not guaranteed losses for any single site.
Which kinds of pages lose the most traffic?
The studies do not break results down by query type. Reasoning suggests short-answer queries, such as definitions and simple facts, are the most exposed because a summary can fully answer them, while comparisons, templates, and how-to content give the reader a reason to click through.
Should I stop writing "what is" posts?
No. Keep them for the citations, impressions, and internal-link value they still provide, but pair each one with deeper content that a summary cannot replace, and link the two together.
What did Google change to send traffic back to publishers?
Google announced five updates: a Further Exploration links section, inline link context, subscription labels, forum and social previews with creator attribution, and product review cards. There is no public data yet on how much traffic they recover.
How do I know if AI Overviews are affecting my site?
Compare impressions and clicks by page in Google Search Console over the past several months. Impressions holding steady while clicks fall is the typical pattern.
Want this executed for you?
Our agency team builds and runs social engines for B2B and SaaS brands.
Written by Mehran Shahmiri
B2B marketing strategist helping SaaS companies build revenue-generating social engines.
Follow on LinkedIn →
