In May 2025, Adobe Express sat down with 1,000 Americans — 800 consumers and 200 marketers and small business owners — and asked a simple question: are you using ChatGPT to search for things? The answer reshaped how a lot of people in digital marketing thought about the next twelve months.
Seventy-seven percent of ChatGPT users said yes. Nearly one in four said they now go to ChatGPT before they open Google. Three in ten said they trust ChatGPT more than traditional search engines. And 36% said they had discovered a new product or brand through ChatGPT — a figure that rose to 47% among Gen Z respondents.
These numbers landed at roughly the same moment that StatCounter data showed Google's global search share dipping below 90% for the first time since 2015. Both data points, taken together, mark something that search industry analysts have debated for two years: the moment when AI-native search behaviour crossed from early-adopter curiosity into mainstream use.
What the data does not show is Google dying. What it does show is that the definition of "search" is changing — and the businesses that adapt fastest to that change are likely to hold their visibility while others lose ground quietly.
What the Adobe Express Survey Actually Found
The survey's methodology matters before drawing conclusions. Adobe Express polled 1,000 U.S. respondents in May 2025, all of whom self-reported using ChatGPT as a search tool. That means the 77% figure refers specifically to people who already use ChatGPT — not 77% of all Americans. The distinction is important: the survey describes the behaviour of an existing ChatGPT-using population, not the general public as a whole.
Within that population, the findings are striking across every demographic, not just younger users. The breakdown by generation showed almost no significant gap in adoption rates:
The near-uniformity across age groups is one of the survey's most significant findings. AI search adoption is not a Gen Z phenomenon waiting to filter upward. It is already happening across the full adult demographic.
The Adobe survey asked respondents why they prefer ChatGPT for search. The top reasons: 54% said it summarises complex topics faster; 33% said it requires fewer clicks than Google; and 77% reported that results felt more personalised. These are friction-reduction motivations — the same behavioural driver behind any shift in tool adoption. Users are not abandoning Google out of ideology. They are choosing the path that gets them to an answer with less effort.
Google's Share: A Decline in Context
StatCounter's data through 2025 showed Google's global all-device search market share dipping to 89.57% in July 2025 — its lowest point in approximately ten years. For comparison, Google held roughly 92.9% share as recently as 2023. The decline is real, and the direction is clear.
But the absolute context matters. Google still processes an estimated 16.4 billion searches per day, and Sundar Pichai stated on Alphabet's Q4 2025 earnings call that Search had more usage in that quarter than ever before. The total search market — search engines plus AI tools combined — grew approximately 26% globally between 2024 and 2025. Google's share of a larger pie is shrinking, but the pie itself is expanding. AI tools are not stealing existing Google searches as much as they are capturing net-new query volume that would not have occurred on Google at all.
First Page Sage's Q2 2026 estimates place Google at roughly 80% of all digital queries when AI platforms are counted alongside traditional search, with ChatGPT accounting for approximately 17% and other platforms sharing the remainder. Those figures represent a dramatic redistribution in less than eighteen months.
How the Shift Happened: A Brief Timeline
The transition from novelty to mainstream behaviour did not happen overnight. It followed a predictable adoption curve, accelerated by a sequence of product decisions at both OpenAI and Google.
What This Looks Like for a Real Business
An SEO Agency That Tracked Both Channels — and Found the Gap
A mid-size UK digital marketing agency had spent three years building Google rankings for a portfolio of e-commerce clients. By Q3 2025, two clients in the same product category were seeing something unexpected: Google organic traffic was flat or declining, but overall revenue was holding steady. The traffic gap was not showing up in Google Analytics.
After running a brand mention audit across ChatGPT, Perplexity, and Gemini, the agency found that one client was being consistently cited in AI responses for product-category queries while the other was almost entirely absent. The cited client had a higher density of long-form guides, third-party review profiles (Trustpilot, G2), and earned media coverage from industry publishers — all formats that research consistently links to higher AI citation rates. The uncited client relied almost entirely on thin product pages optimised for Google's keyword model.
The agency restructured the uncited client's content around in-depth guides and FAQ formats, pursued earned media placement in three industry publications, and actively built out review profiles on third-party platforms. Within sixty days, the client began appearing in AI-generated responses for its primary product queries. Organic revenue — from both Google and AI-referred traffic — increased over the following quarter.
What This Means for Businesses and Marketers
The Adobe Express survey found that 76% of marketers and business owners already consider it essential for their brand to appear in ChatGPT results, and two thirds plan to increase investment in what is being called "AI visibility" in 2025 and beyond. That shift in marketer perception reflects a structural change in how brand discovery works.
The channels through which AI tools surface businesses differ significantly from the signals Google prioritises. Research from multiple sources — including McKinsey's September 2025 CMO survey and SE Ranking's citation analysis — points to a consistent picture of what drives AI search visibility.
Long-Form Guides and FAQs
The content format AI tools cite most often. In-depth how-to guides and structured FAQ pages are referenced far more than thin product pages or category descriptions.
57% marketer success rate
Third-Party Review Profiles
Presence on Trustpilot, G2, Capterra, Yelp, and similar platforms correlates with roughly three times higher AI citation odds compared to brands with no third-party review presence.
3× citation likelihood
Earned Media Coverage
Being cited in industry publications, blogs, and news sites boosts AI citation rates by up to 325% compared to brands that publish exclusively on their own sites.
325% citation boost
Community Mentions
Brands with high mention density on Reddit, Quora, and similar community platforms have approximately four times higher AI citation odds — because AI tools weight user-generated discussion as a trust signal.
4× citation odds
Domain Authority and Link Volume
SE Ranking found that domains with 32,000 or more referring domains are 3.5 times more likely to be cited in AI search results — reinforcing that traditional SEO signals still matter for AI visibility.
3.5× more likely cited
Structured Data and Factual Clarity
AI tools prioritise content that states facts clearly, includes specific numbers, and structures information for machine readability. Ambiguous or vague content is consistently deprioritised in AI citations.
Precision wins
"Google's share is not collapsing. Search is expanding. The businesses at risk are those still optimising only for the old definition of the word."
Google SEO vs AI Visibility: How the Signals Differ
| Ranking Signal | Google SEO Weight | AI Citation Weight |
|---|---|---|
| Keyword-optimised page titles | High | Low |
| In-depth long-form guides | High | Very high |
| Third-party review platforms | Moderate | Very high |
| Earned media / press mentions | Moderate | Very high |
| Reddit / Quora mentions | Low | High |
| Domain authority (backlinks) | Very high | High |
| Page speed and Core Web Vitals | High | Not a factor |
| FAQ / structured answer formats | Moderate | Very high |
Track Every Channel Your Traffic Comes From
As search fragments across Google, ChatGPT, Perplexity, and Gemini, knowing which channel drives your clicks matters more than ever. Trimrly short links tell you exactly where your traffic originates — per link, per platform, per campaign.
What to Do — and What to Avoid — Right Now
- ✕
Abandoning Google optimisation to chase AI visibility exclusively. Google still processes approximately 80% of digital queries globally. Brands that deprioritise Google to pivot entirely toward AI search are trading a dominant channel for an emerging one before the numbers justify it. The correct posture is parallel optimisation, not substitution.
- ✕
Assuming AI search visibility is unpredictable or uninfluenceable. Research consistently shows that specific content formats, platform presence, and earned media coverage correlate strongly with AI citation rates. Brands that treat AI visibility as a black box are missing an optimisable system with documented inputs.
- ✕
Treating the Adobe survey's 77% figure as representing all Americans. The survey sampled existing ChatGPT users only. The broader US adult population has considerably lower ChatGPT penetration. The 77% reflects the behaviour of an already-converted population, which makes it significant as a directional signal but not as a description of the entire market.
- ✕
Ignoring product discovery in AI results. Thirty-six percent of survey respondents said they had discovered a new brand or product through ChatGPT, rising to 47% among Gen Z. For e-commerce and consumer brands, this is not a future trend — it is a current acquisition channel that most brands are not actively optimising for.
- ✓
Auditing your current AI visibility before investing in improvement. Search your brand name and primary product queries in ChatGPT, Perplexity, and Gemini. If you are not appearing, note which competitors are and what type of content they are being cited for. That gap analysis is your immediate action list.
- ✓
Building third-party review presence now, not later. Review profiles on Trustpilot, G2, Capterra, and similar platforms correlate with three times higher AI citation odds. This is one of the highest-leverage, lowest-cost actions available to most businesses — and it compounds over time as reviews accumulate.
- ✓
Tracking all your traffic sources with link-level granularity. As users arrive from an increasing number of entry points — Google, AI tools, social platforms, QR codes — understanding which channel drives which behaviour requires link-level tracking, not just session-level analytics. A Trimrly short link per campaign and per channel gives you that precision without touching your analytics setup.
Frequently Asked Questions
Not in absolute terms. Google's total search volume reached new highs in Q4 2025, according to Sundar Pichai on Alphabet's earnings call. What is changing is Google's share of an expanding total market. As overall search plus AI query volume grows, AI platforms are capturing a disproportionate share of new query growth, which causes Google's percentage to decline even while its absolute usage rises. The net effect is a more competitive landscape, not a shrinking one.
No. Adobe Express clarified its original report: the 77% refers to Americans who already use ChatGPT, not the entire adult population. All 1,000 survey respondents were self-selected ChatGPT users. The correct reading is that 77% of the ChatGPT-using population treats the tool as a search engine — which is a significant behavioural finding about that specific group, not a claim about the general public.
AI citation research points to several consistent drivers: publishing long-form, factually specific guides and FAQ pages; building a presence on third-party review platforms like Trustpilot and G2; earning media coverage in industry publications; generating community discussion on Reddit and Quora; and building high-authority backlinks. Brands with more referring domains, more earned media, and more third-party mentions consistently appear more often in AI-generated responses. There is no paid placement available; citations are earned through content relevance and authority signals.
The current data does not support a near-term replacement scenario. Google's infrastructure advantages — 16 billion daily searches, Android distribution, Chrome defaults, and deep integration with Maps, YouTube, and Shopping — are extremely difficult to displace at scale. The more likely near-term outcome is a hybrid search environment where different query types naturally route to different tools: conversational and summarisation queries to AI tools, navigational and transactional queries to Google. Gartner analysts suggest this fragmentation, rather than replacement, is the dominant near-term trajectory.
SEO strategy needs to expand its scope without abandoning its core. Google optimisation remains essential for the 80% of queries it handles. AI visibility optimisation — focusing on long-form content, third-party review presence, and earned media — should run alongside it. Ahrefs data from April 2026 showed Google's share experiencing its steepest single-month contraction on record, making diversification away from a single search channel the clearest strategic priority for any content or traffic-dependent business.
