The question we keep hearing from senior leaders and marketing teams across every industry is almost identical: "Surely, eventually, we'll all just search using AI?"

It’s an understandable reaction to the pace of disruption, but that assumption rests on a fundamental misunderstanding: it assumes search is moving toward a single, new destination. What our latest data across both our flagship SearchPulse Q3 2026 Report and the SearchPulse Q3 2026 - Travel Edition reveals is the exact opposite.

People aren't simply swapping traditional search engines for AI tools; instead, they are building layered discovery habits, moving between Google, AI assistants, social platforms, niche communities, and brand websites depending on what they want to achieve and how confident they need to feel.

Search has become a much bigger conversation than SEO rankings or AI visibility. It is about understanding how human psychology drives people to discover, evaluate, verify, and ultimately choose.

Here is what our latest research tells us about how audience behaviour is shifting, and what your brand must do next to remain visible and trusted.

AI Has Plateaued into a Tool Stack, Not a Single Winner

For the past two years, brand strategy has been dominated by a rush to "show up in ChatGPT.", however, our data shows that AI search is no longer concentrated around a single platform.

While 47.1% of people regularly use at least one AI search tool, market leader ChatGPT’s regular usage has essentially plateaued quarter-on-quarter at 30.85%.

Meanwhile, secondary players are steadily expanding their footprint: Google Gemini holds 20.85% of regular users, and Claude has climbed to 8.25%. Rather than picking one "winner," users are quietly assembling their own personal technology stacks based on convenience and context.

77.95% vs 47.1%

GOOGLE SEARCH REGULAR USAGE (77.95%) VS OVERALL AI TOOL USAGE (47.1%)

From a behavioural standpoint, Gemini’s ongoing growth illustrates the Mere Exposure Effect, a psychological phenomenon where repeated exposure to an interface (such as default Google integration) increases familiarity and perceived trustworthiness over time.

What does this mean for marketers? A single "Generative Engine Optimisation (GEO)" score across all AI platforms is no longer sufficient [cite: 2]. As Mikkel Wegener Nielsen, Chief Revenue Officer at AccuRanker, highlights in our Q3 report:

ChatGPT, Gemini, and Claude have different models, data sources, and user intents... ChatGPT leans heavily on Bing's index; Gemini utilises Google's own index; while Claude tends to weigh authoritative, citation-worthy content more. Precise, helpful, and informative content performs across all LLMs, but tracking share of answer in each environment is critical. — Mikkel Wegener Nielsen, CRO at AccuRanker

If your strategy assumes your audience is only searching in one place, you are already leaving significant visibility on the table.

The Journey Shortens at Booking: Inspiration vs Consideration

One of the most striking patterns across our Q3 research, particularly within high-consideration sectors like travel, is how search behaviour alters as decision-making progresses.

In our travel study, 25.4% of travellers regularly turn to AI Overviews for initial inspiration, but by the time they reach the actual booking moment, 42.9% say they 'Never' use AI. Across the full search journey, the average number of search tools used steadily tapers from 2.38 tools at the inspiration stage down to 1.88 tools at booking.

2.38 → 1.88

AVERAGE NUMBER OF TOOLS USED PER PERSON FROM INSPIRATION (2.38) TO BOOKING (1.88)

This transition mirrors Daniel Kahneman’s dual-system framework of cognitive processing:

System 1 (Fast & Intuitive): Used during early discovery. People rapidly browse AI summaries, social vlogs, and inspiration hubs to gather ideas effortlessly. Here, the Availability Heuristic dictates that whatever surfaces easily shapes the shortlist.

System 2 (Slow & Deliberate): Triggered when financial commitment or risk enters the equation. Users narrow their focus back to a small pool of high-trust transactional channels (e.g., Google Search, OTAs like Booking.com, or direct brand sites) to verify details.

Understand System 1 and System 2 thinking from a SearchPulse perspective here.

George Shuttlewood, CMO at Global Touring (comprising Topdeck Travel, Back-Roads Touring, and Grasshopper Adventures), explains how this dynamic presents an opportunity for niche operators:

AI discovery rewards specificity. A small brand with real depth in one area can surface in places it would never outspend anyone to reach on generic brand search. The discovery shortlist is now the battleground... If AI doesn't recommend you at inspiration, you're not in the comparison at all. — George Shuttlewood, CMO at Global Touring

To win, brands must ensure they are present early in AI syntheses to make the shortlist, while providing bulletproof proof points on-site to convert deliberate System 2 evaluators.

Distrust & The "Manual Checkpoint": Why Verification Drives Action

Despite AI’s rapid adoption, consumers are not surrendering critical evaluation. In fact, distrust remains the single largest barrier holding people back from AI search (31.8%). Right behind it, 24.6% of users explicitly state that they prefer checking original sources themselves.

Rather than replacing verification, AI search has altered where it takes place. AI handles the heavy lifting upfront, summarising options and eliminating research friction, leaving the user with a smaller, highly focused task: double-checking the source before buying.

+42%

HIGHER CONVERSION RATE FOR AI-REFERRED TRAFFIC COMPARED TO NON-AI ORGANIC TRAFFIC

This shifting habit explains why industry benchmarks, such as Adobe for Business data cited in our report, show that AI-referred traffic converts 42% better and generates 37% higher revenue per visit than non-AI traffic. By the time a visitor clicks through from an AI summary, they are already pre-qualified and searching for a reason to trust you.

Jae Hopkins, Marketing Director at Explore Worldwide, captures this operational reality well:

Our website content pages always used to bring in most of our organic search traffic, but now those people seem to be coming in more at 'trip page' level. We think AI tools are using info from our blog pages to answer higher-funnel questions, but as soon as it gets down to detailed info, people come to the trip pages to check for themselves. — Jae Hopkins, Director at Explore Worldwide.

If your landing pages are filled with generic marketing fluff rather than clear pricing, explicit trade-offs, and clear cancellation terms, you will fail the manual checkpoint.

The Authenticity Advantage & The Reddit Dilemma

Where do consumers turn when they want unvarnished truth? Community platforms.

Our Q3 data highlights that 51.9% of Reddit users cite 'information from real people' as their main reason for using the platform, beating LinkedIn (37.4%), ChatGPT (11.2%), and Google Search (8.9%) by a landslide.

Psychologically, this is explained by the Persuasion Knowledge Model - When consumers recognise an overt attempt to sell to them, their defensive filters activate. Independent, peer-led conversations bypass this scepticism because the motive to persuade is absent.

However, as brands scramble to capture Generative Engine Optimisation (GEO) citations by inserting themselves into community threads, a new risk emerges. As Reflect Digital’s Innovation & Automation Lead, Matt Greenwood-Wilkins, points out:

Reddit has become the poster child for non-brand sources in AI answers... But that reputation is turning it into a target. Brands are spinning up fake accounts en masse to seed threads. With Originality AI estimating around 15% of Reddit posts are already AI-generated, the harder that signal gets to trust, the sooner its prominence will erode. — Matt Greenwood-Wilkins, Innovation & Automation Lead at Reflect Digital [cite: 2]

Instead of trying to manufacture authenticity or game forums with undercover brand profiles, marketers should focus on understanding customer pain points through community social listening, and then address those exact questions transparently on their own platforms.

Key Takeaways for Marketers

Audit Multi-LLM Visibility: Stop treating AI visibility as a single metric. Look at benchmarking your commercial prompts across the AI search platforms independently.

Optimise for the 'Manual Checkpoint': Recognise that AI drives high-intent traffic looking to verify claims and ensure your product and pricing pages feature clear details.

Align Strategy to Cognitive Modes: Win System 1 inspiration and System 2 evaluation search behaviours.

Target Specificity Over Generics: AI tools reward depth and unique insights, so niche expertise and unique first-party data will outrank generic style content every time.

Protect Brand Credibility: Think about how you show up and communicate across community platforms; don’t go in too sales’y and create genuine connection.

Ready to audit your brand's multi-platform AI and SearchPulse readiness? Discover how your audience searches and convert intent into action with our human-behaviour-led search strategies. Book a meeting with our Strategy Team today or explore our AI & GEO Services.

MEET THE
AUTHOR.

LUCY TODD

Lucy works alongside the Marketing Director and Marketing Coordinator to oversee growth and delivery of the agency's marketing strategy. Overseeing and managing various channels including CRM, Content and Social, Lucy contributes to the agency's ongoing strategy through idea generation, research, campaign creation & activation, and project management. Alongside contributing to the agency's continuous growth, Lucy supports the agency's events including Roundtables, and contributes to the thought leadership programme.

More about Lucy