Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

Contact us

London

34 Bow Street, London​ WC2E 7AU

New York

hello@walk-in-media.com
+1 646-236-3405
71 Fifth Ave, 8th Floor, New York, NY 10003
Subscribe to newsletter

Subscribe to receive the latest blog posts to your inbox every week.

By subscribing you agree to with our Privacy Policy.
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

Search market update: How user behaviour and the search landscape are changing

‍

‍

‍

Summary

‍

Despite Google dominating the search landscape, accounting for ~93% of searches globally, there is a significant shift in user behaviour as a result of 1 in 10 global queries being AI-served. Users are also increasingly searching within the platform they are in, with the likes of TikTok seeing an emergence as a search alternative, especially with the younger audience. 

‍

With AI-generated results, be that within Google (AI Mode/AI overviews) or across alternative large language models (LLMs), users are receiving answers to their queries without having to navigate through search results, a shift that is trending upwards. Whilst Google still owns 89.66% of the market share of all search activity, followed by YouTube and then Bing, LLMs globally currently handle less than 2% of overall query volume, but this is growing. 

‍

Voice search is also on the rise, with 27% of searches in the Google App being done by voice, as well as Lens image searches increasing 4x since 2021. That being said, traditional text search still dominates the remaining 73% of searches.

‍

LLMs process content semantically, evaluating context, relationships between ideas and overall coherence, not just keyword placement or scraping. As a result of changes in user behaviour as mentioned above, visibility in AI-generated answers—and being included in those summaries—is increasingly critical. SEO is no longer keyword optimisation but must incorporate optimisations towards appearing in AI overviews and LLMs.

‍

Brands must develop a consistent, credible, and authoritative identity that these models can identify and reward. Whilst also considering alternative measurement KPIs as zero-click searches continue to rise.

 

‍

The Search Landscape and User Behaviour

‍

The search landscape shows an enormous gap between Google's 5+ trillion annual searches and all competitors combined. YouTube's 1.28 trillion annual searches make it a significant second player, while Bing's 36.5 billion searches represent a distant third. The most interesting trend is TikTok's emergence as a search alternative, particularly among younger users, though specific volume numbers aren't disclosed. 

 

Despite Google dominating ~90% of global searches, user behaviour is changing; users are increasingly searching on the likes of TikTok & Instagram, as well as going to communities or LLMs for inspiration, reviews and centralised content. 

‍

Looking deeper into user behaviour, it’s reported that:

-  10% of Gen Z prefer TikTok over Google 

-  41% of Americans have used TikTok as a search tool. 

-  Studies show 45% of Gen Z use social media (TikTok/Instagram) for search. 

‍

Multiple sources indicate that close to half of all product searches start on Amazon, while Google follows with 32%. So although Google remains the gateway for general web queries, search is increasingly happening within siloed platforms that offer relevance, visuals or trusted ‘advice’. 

‍

E-commerce is largely platform first – when shoppers move beyond awareness to transaction, they often start—and sometimes finish—their journey on Amazon, potentially bypassing traditional search altogether.

 

The future of search is assistant-driven and multi-modal. 

The next era of search is no longer about typing queries into a text box. Instead, it is assistant-driven and multimodal — where users interact with information through voice, images, video, and context-aware prompts, seamlessly integrated across devices. This shift is propelled by the rapid adoption of AI assistants, voice-activated devices, and visual discovery tools like Google Lens. 

‍

Search is evolving from being a reactive information gateway into a proactive companion that not only provides answers but also anticipates intent and executes actions. 

-         More than 1 in 3 consumers use voice assistantsregularly – including smart speakers and voice controls in cars 

-         Users prefer to ask full questions or moreconversational queries rather than short keywords 

-         Google Lens has surpassed 10 billion monthlysearches globally, whilst Pinterest Lens and TikTok/IG visual search havenormalised shopping via photos or videos. Google Lens only accounts for ~0.4%of all searches in the platform but has increased 4x since 2021 and isincreasingly popular with the younger demographic. 

-         With LLM-powered assistants like ChatGPT,Gemini, and Bing Copilot, search is becoming more personalised, contextual, andproactive. Instead of retrieving ten blue links, assistants can:

·      Interpret intent 

·      Cross-reference modalities (text, image,voice) 

·      Deliver an actionable outcome (book a table, buya product, generate a plan). 

This represents a shift from search engineto action engine. 

‍

‍

LLMs vs Traditional Search

‍

LLM usage is growing but still modest in volume. Google holds approx. 93% of search engine market share, whilst LLMs account for less than 0.2% of search traffic.

Globally, LLM-based tools are reported to handle under 2% of overall query volume today, whilst most searches still go through Google and other traditional engines. 

‍

AI Overviews are changing user query behaviour. 

-         On Google, AI-generated overviews now appear inover 13% of searches, up from around 6.5% at the start of 2025.

-         These AI overviews are associated withzero-click rates hovering around 58–60%, meaning users often get answerswithout visiting external websites.

-         Consequently, 20–40% drops in organic click-throughrates are reported for searches with AI summaries. Heavily impacting reportingon traditional SEO metrics.

‍

Consumer behaviour and LLM adoption trends 

-         A Gartner forecast suggests over 25% of searchinteractions could occur through AI chatbots by 2026.

-         Estimates from Trackerly.ai suggest that, combining standalone LLM usage(like ChatGPT) and Google’s AI overviews, around 9.7% of total searches mayalready be answered directly by AI systems—without redirecting to traditionalsearch results.

-         Reddit discussions echo this shift:

“1 in 10 Americans now turn to generative AI first for online search.” 


“Zero-click journeys are accelerating: 58.5% of US searches and 59.7% in Europe end without clicks.” 

 

 

Implications for Advertisers

‍

Brands and advertisers must develop a consistent, credible, and authoritative identity that LLMs can identify and reward, whilst continuing to build on a best-in-class SEO and paid advertising strategy as well as maintaining the brilliant basics. Brands must also consider alternative measurement KPIs as zero-click searches continue to rise, moving away from traffic-based metrics to visibility and association.

‍

Key recommendations for advertisers to get ahead and build strong foundations for LLMs and AI results include:

-         Build a strong technical SEO strategy. Includingexpert content on site increases brand mentions and link building.

-         Using Share of Model tools to gain anunderstanding of your visibility against relevant queries/prompts vscompetitors. These tools can break insights down by each major LLM as well asthe qualitative sentiment that brands are mostly associated with.

-         Recent research has indicated that Wikipedia issourced in around 50% of ChatGPTresponses, whilst Reddit is consistentlyreferenced across a number of them and YouTube is prevalent for Gemini, sothese are rich environments to start testing.

-         To influence LLMs, AI overviews and ultimatelyconsumers, brands need to adopt a multi-modal approach that uses variouscontent types across text, image, video and audio across multiple platformsfrom web to social and video/audio platforms that are used to build LLM data.

‍

Another key area to consider is optimising for voice queries within paid search, using campaign types such as broad match, PMax and AI Max, which allows brands to appear for more conversational keywords based on users' intent.

‍

Although AI overviews and LLM results are not yet being monetised (in the UK), this is something we expect to be coming very soon, and we therefore need to be prepared.For Google (Gemini/AI Mode), this is testing campaign types mentioned above; this is also the case for Microsoft Bing (Co-pilot) with their equivalent campaign types.

‍

As mentioned previously, advertisers must also rethink measurement. Going beyond clicks to measure:

-         Brand presence in AI answers (share ofvoice) 

-         Citation frequency acrossReddit/Quora/Wikipedia 

-         Engagement and conversions from social searchqueries 

-         Use AI search audits to understand how yourbrand appears across different LLMs. 

‍

The shift away from traditional search engines is not just a technological evolution; it’s a fundamental change in consumer behaviour that demands corresponding shifts in marketing: from persuasion to precision, from keyword to advice, and from market share to problem-share. ​

Do it right, and brands can establish themselves as essential participants in the algorithmic conversations that increasingly shape consumer decisions.

‍

Written by Donna Stanley

‍

‍

‍

References:

Sourceson Search Market Share & Usage

Sources on Social Search &Discovery Platforms

Sources on LLM & AI SearchImpact

‍

Share this post
The shift away from traditional search engines is not just a technological evolution; it’s a fundamental change in consumer behaviour that demands corresponding shifts in marketing
Read More >
Search market update: How user behaviour and the search landscape are changing
October 9, 2025
Over the years, Google have implemented many AI powered tools that aim to ease the efforts of campaign building, monitoring and optimising for those in marketing.
Read More >
The implications of AI driven analytics
August 22, 2024
I am currently a Product Assistant at Walk-In Media completing a Level 3 Digital Marketing apprenticeship. This involves an 18-month long process where....
Read More >
A week as an apprentice at Walk-In Media!
July 17, 2024
As a media agency, we’re often briefed on campaigns where we are asked to provide the measurement KPIs as part of our response. More...
Read More >
The Measurement Quandary
January 1, 2024
Usually, the most challenging time to find a job is straight after university. Many graduates may have limited, or even zero, experience...
Read More >
4 Benefits of Doing a Placement Year
January 1, 2024
Introduction I joined Walk-In Media as a Planning and Strategy Assistant on a Professional Placement year. This was a massive change to...
Read More >
Working at Walk-In Media on a placement year.
February 1, 2024
Who we are
Who we are
Our work
Our work
Work for us
Work for us
Contact
Contact
Who we are
Our work
Work for us
Contact
© 2024 Walk-In Media. An MSQ company. All rights reserved.
Privacy PolicyTerms of Service