How Google AI Overviews Impact SEO: What You Need to Know in 2026
Google AI Overviews now sit above the results for a large share of searches, but the effect on your traffic depends entirely on what people are searching. Here is what I am seeing across the Search Console accounts I manage: which query types lost clicks, which ones held steady, what that means for a small business in India, and a 5-step way to adapt without guessing.
Somewhere in the last two years, the first thing most of your customers see on Google stopped being a blue link. For a large and growing share of searches, it is now a generated summary sitting above the results, built from a handful of pages Google's system decided were worth reading on your behalf. I manage Search Console for a mix of clinics, shops and service businesses across Pune, and I watch this play out differently on every single account. On some, impressions keep climbing and nothing in the click data looks unusual. On others, a handful of pages that used to pull in steady traffic have gone quiet, even while their ranking position barely moved.
This post is not another explainer on what AI Overviews are, I have covered that ground in what generative engine optimization is and in SEO vs GEO in 2026, and it is not a tactical how-to either, that is what how to appear in Google AI Overviews is for. This is specifically about impact: which kinds of searches lost clicks, which held steady, what that split looks like for a small business in India, and what I actually do about it on the accounts I manage. Everything below is framed as an observation from the properties I look at every week, not a claim about search behaviour everywhere, since AI Overview triggering varies by query, by device and by the month you happen to check it.
What actually changed in the results page
Google's AI Overviews read several ranking pages, generate a summary, and place it above the traditional results with citations back to the sources it drew from. The mechanism itself is not complicated. What matters for this post is simpler still: some searches now get answered on the results page itself, and some do not, and the split between those two groups is not random. I am not going to relitigate the full rollout history here, my honest take on whether SEO still works in 2026 covers that timeline in more depth and answers a broader question than this post is trying to. The question worth answering on its own is what this shift actually does to your click-through rate, and the honest answer is that it depends heavily on what someone typed into the search box in the first place.
I want to be upfront about one thing before going further: I am not going to hand you an industry-wide percentage for how much traffic AI Overviews have taken, and I would treat any source that gives you one, without naming exactly which accounts and which months it came from, with a healthy amount of scepticism. What follows is the pattern across the Search Console accounts I personally manage, checked by hand, month over month.
Click-through impact by query type
Informational queries
This is where the drop is real and, in the accounts I manage, the most visible. A search that wants a simple fact or a short definition, the kind of query a page could answer in one sentence, increasingly gets that sentence directly inside the AI Overview. I cannot hand you a precise percentage I would trust, but the pattern across the Search Console properties I look at every week is consistent: pages built around a single narrow definition are showing more impressions and fewer clicks than they used to, month over month, even while their average position holds steady.
Local and near-me queries
This is the category I have watched most closely because it is where most of my clients earn their revenue, and it has held up noticeably better. Someone searching for a service near a specific locality is usually not looking for a paragraph of general information, they want a specific business, with a phone number, hours and reviews, and that is still delivered through the map pack and organic listings rather than a generated summary. I have not seen a client account where genuine near-me traffic collapsed because of AI Overviews. Where a dip shows up, it tends to trace back to something else entirely, a seasonal lull or a technical issue, rather than the AI Overview itself.
Transactional queries
Searches with clear buying intent, comparing prices, checking availability, reading reviews before a booking, keep sending clicks in the accounts I track. A summary can tell someone that dental implants exist and roughly what they cost, but it cannot show them a specific clinic's actual patient photos, a real Google rating with hundreds of reviews attached, or the exact package a business offers. That evaluation step is still happening on a website, not inside a generated paragraph, because the decision itself has not gotten any less detailed just because the search got faster.
Brand and navigational queries
Once someone has already heard of a business, from a signboard, a referral or an Instagram reel, and types the name into Google, an AI Overview rarely appears at all. This is the one category I would call genuinely untouched in every account I manage. What shows up for that search, your website, your reviews, your social profiles, still depends entirely on the same SEO and reputation groundwork it always did before AI Overviews existed.
The Indian small-business lens
Search behaviour in the accounts I manage skews mobile, local and conversational in a way that happens to sit outside where AI Overviews do the most damage. A large share of Indian small-business queries are not research questions at all, the searcher already knows they want a dentist, an electrician or a CA, and they are trying to find one nearby they can call today, not read a summary about the profession in general. That kind of query was never going to be satisfied by a generated paragraph, and I have not seen it become one over the past year.
The other pattern worth naming is where the actual conversion happens. Almost every client I work with closes the sale over a phone call or a WhatsApp message, regardless of whether the discovery started on Google, Instagram or a referral. AI Overviews mostly touch the research phase of a search, the part where someone is still deciding what they want, and a meaningful share of Indian local search skips that phase because the intent was already specific and local before the search even happened. None of this means Indian businesses are immune, competitive city-wide keywords have gotten harder simply because more businesses are publishing content, but the underlying mechanism, a business ranking for the searches that lead to a call, is intact on every account I currently manage.
Where clicks fell and where they held
Laid out plainly, the pattern I keep seeing across client accounts looks like this: broad informational and definitional content has softened, sometimes noticeably, while local, transactional and branded searches have kept sending traffic at close to the same rate as before. It is worth repeating that this is an observation from the accounts I manage and the searches I check by hand, not a claim about search behaviour everywhere.
| Query type | What I am seeing |
|---|---|
| Broad informational ("what is X") | Clicks softening; impressions often stable or rising |
| Near me / service plus city | Little visible change; map pack still dominates |
| Transactional / comparison | Holding; users still click through to evaluate |
| Brand / navigational | Untouched in every account I track |
The businesses most exposed are the ones whose traffic leaned heavily on the first row of that table, generic definitional posts with no local or commercial angle. If that describes a meaningful share of your own content, the rest of this guide is about what to do next, and my separate piece on SEO KPIs worth tracking is a useful companion for deciding which numbers to watch while you make the change.
What to do about it
Be the cited source, not just a ranked one
Where an AI Overview is going to appear regardless of what you do, the next best outcome is being the page it cites. That is a tactical, page-level job, answering the exact question in the first few sentences, structuring sections so each one stands alone, adding schema that removes ambiguity about who wrote the page and when, and I have laid out the full process in how to appear in Google AI Overviews rather than repeating it here.
Own the map pack
Since local and near-me searches are the category holding up best, the single most valuable move for most Indian small businesses is making sure the map pack position is not in question. That comes down to the same fundamentals it always has, a complete and consistent local SEO setup and a Google Business Profile that is actually maintained, and I have written specifically about how this plays out for local businesses in Google AI Overviews for local business.
Capture the click with pricing and FAQs
A generated summary can state a fact, but it cannot show a specific price list, a real availability calendar or a detailed answer to an unusual question a customer actually has. Pages that lead with concrete pricing ranges and a genuine FAQ section, marked up with the schema covered in FAQ schema for Google AI Overviews, give a searcher a reason to click through even after reading a summary, because the summary was never going to contain that level of specific detail in the first place.
Build direct channels the AI cannot touch
Every WhatsApp broadcast list, email list or returning customer you build reduces how much of your business depends on a single search result, generated or otherwise. A customer who already has your number does not need to search for you at all, let alone read a summary about you, which makes this the most durable answer to a shift you ultimately do not control.
A 5-step adaptation plan
This is the process I actually run when a client asks what to do about falling clicks on a page that has not lost ranking. It takes a few hours spread across a week, not a rebuild.
Step 1: Segment your keywords by AI Overview exposure
Search your main target keywords in an incognito window and note which ones trigger an AI Overview and which do not. Rank trackers such as Semrush or Ahrefs flag this automatically if you are tracking more than a handful of terms. Sort the list into three buckets: no overview, a partial summary that leaves room for a click, and a full summary that answers the question completely.
Step 2: Protect the queries that still send clicks
For the near-me, transactional and branded terms from the previous section, resist the urge to touch what is already working. Confirm your Google Business Profile, pricing pages and review volume are current, since these are the pages carrying the account while the exposed pages get fixed.
Step 3: Rewrite exposed pages to lead with a direct answer
For pages sitting in the fully-summarised bucket, restructure the opening so it answers the core question in the first two or three sentences, then use question-phrased subheadings for the sections that follow. This will not necessarily win back the click on that exact query, but it improves your odds of being the cited source rather than an ignored one.
Step 4: Add schema that removes ambiguity
Article schema for authorship and dates, FAQPage schema for question-and-answer sections, and LocalBusiness schema if you serve a specific area. None of this guarantees a citation, but I have not seen a page earn one consistently without it either.
Step 5: Build a channel AI Overviews cannot touch
Pick one direct channel, a WhatsApp list, an email newsletter, a Google Business Profile posting habit, and commit to it weekly. This is the part of the plan that has nothing to do with what Google decides to summarise next, and it is the one most businesses skip because it does not feel like an SEO task.
Honest limits: what nobody can promise
I am not going to tell you there is a formula that guarantees a citation, because I have not found one, and I would be sceptical of anyone who claims otherwise. Two pages with near-identical structure and quality can get treated differently by the same AI Overview, and a page that was cited last month can quietly stop being cited this month without anything on it changing. Google's own system decides this at the moment of the search, not once and permanently, which is a hard thing to plan around but an honest thing to say out loud.
There is also no reliable way to predict how long a fix takes to show results. On the Pune dental clinic account I mention across this site, ranking moved from position #59 to the top five in about two months, and I have seen AI citation follow a broadly similar timeline elsewhere, appearing within weeks of a page reaching page one. I have also seen pages sit on page one for months without ever being cited for a specific query. Treat every timeline anyone gives you, including mine, as an observation from past cases, not a schedule you can plan a quarter around. If you want a wider view of what SEO investment does and does not return right now, is SEO worth it in 2026 covers that question directly.
Mistakes I see most often
Auditing accounts that feel like they are losing ground to AI Overviews, the same handful of problems show up repeatedly, and most of them predate AI Overviews entirely.
- Blaming AI Overviews for a drop that is actually seasonal or technical. Check Search Console for the full picture before assuming an AI summary is the cause, sometimes a page simply slipped in ranking for an unrelated reason.
- Rewriting pages that were never exposed in the first place. Spending effort on a near-me or branded page that never lost clicks, while an exposed definitional page sits untouched.
- Chasing a citation instead of the underlying ranking. Schema and structure help, but none of it matters on a page that is not already on page one.
- Treating every AI Overview appearance as a loss. A citation with your brand name attached is closer to a trust signal than a threat, and some of these pages see steadier, more qualified traffic afterward.
- Abandoning content entirely instead of redirecting it. The fix for a softened definitional post is usually to add pricing, local detail or a genuine FAQ, not to stop publishing altogether.
Frequently asked questions
Do AI Overviews reduce clicks for every kind of search?
No. In the Search Console accounts I manage, the drop is concentrated in broad informational queries, the kind that ask for a simple fact or definition. Local, transactional and branded searches have held up far better in the same accounts, because those searchers need to compare, verify or simply find contact details that a short AI summary cannot fully replace. Treat the impact as uneven by query type rather than a flat tax on all organic traffic.
Which searches still send full click traffic despite AI Overviews?
From what I track across client accounts, near me and service-plus-city searches, branded searches for a business by name, and comparison or pricing questions still convert into visits at close to the same rate as before. These are queries where the searcher wants to evaluate a specific business, not just receive a fact, so an AI summary functions as a preview rather than a replacement for the click.
How can a small business in India get cited in an AI Overview?
Start with a page that already ranks on page one for the question, since AI Overviews draw almost entirely from that pool. Answer the exact question in the first few sentences, add FAQ and LocalBusiness schema, and keep your Google Business Profile details identical to your website. I cover the full tactical process in my guide on how to appear in Google AI Overviews, this post focuses on what the shift means for your traffic rather than the mechanics of citation.
Should I stop investing in blog content because of AI Overviews?
Not the content that answers a real buying question, but I would rethink posts written only to catch a generic definition search, since those are the pages losing the most clicks in the accounts I manage. Redirect that effort toward pages with pricing, local specifics, FAQs and detail a summary cannot substitute for. That kind of page tends to hold its traffic and still has a reasonable shot at being the cited source.
Is there a way to guarantee my page gets cited in an AI Overview?
No, and I would be cautious of anyone who tells you otherwise. Citation depends on Google's own evaluation of trust, structure and relevance at the moment someone searches, and it can change from week to week even without you touching the page. What you can control is ranking well, answering clearly and keeping your information consistent, which improves your odds without ever guaranteeing the outcome.
Related reading
- Google AI Mode in India: What It Is and What It Changes
- Google AI Mode vs AI Overviews: Differences That Matter
- How to Rank in Google AI Mode: What Gets You Cited (2026)
- How to appear in Google AI Overviews: the step-by-step tactics this post deliberately does not repeat.
- SEO vs GEO in 2026: how optimising for AI citations differs from and overlaps with traditional SEO.
- Does SEO still work in 2026?: the broader, numbers-first answer to whether organic search is still worth the effort.
- What is GEO (Generative Engine Optimization)?: the full definition, for readers who want the concept before the impact data.
- What is AEO (Answer Engine Optimization)?: the broader term covering every kind of direct-answer search surface.
- Google AI Overviews for local business: the GBP and schema signals that matter specifically for local queries.
- Free Website SEO Checker: run 40+ on-page checks before you decide which pages to rewrite.
Get new guides in your Google Search & Discover feed.