Cecilia De La Viesca Every year someone declares that search is dead, and every year they are wrong. But in 2026, something has definitely changed, not in a speculative, ‘this is coming’ sense, but in a measurable, commercially significant sense that is already showing up in real client data.
In July 2026, we hosted two Future of Search breakfasts at Caravan, Covent Garden, bringing together over 80 marketing leaders from financial services, fintech, legal and professional services. We had six speakers across two events, and one consistent picture of where search is heading and what experienced marketing teams should do about it.
This article captures the most important insights from both sessions. Not as a summary, but as the practical briefing we would give any client facing the same challenges.
Readers who already know the event will want to jump straight to the slides: the full deck is linked here. Or stay here and read on for a full breakdown of everything we covered across both mornings.
Prefer to watch than read? The full highlights reel from the morning is on YouTube: two minutes covering everything we presented across both events. Watch the highlights on YouTube below.
The numbers that set the scene
Before the strategy, let’s give some context.
68% of Google searches now end without a click. That figure was 59% in 2025. The direction of travel is not ambiguous.
Google AI Overviews (the synthesised answer panels now appearing above organic results) are triggered on 47% of all queries. Coverage grew from 6.5% in January 2025 to 13.1% by March 2025 and 47% by March 2026. When an AI Overview appears, organic click-through rate drops 61%, from 1.76% to 0.61% (Seer Interactive, 2025).
At the same time, 37% of consumers now start their search with an AI tool rather than Google, with 62% doing so because they find it quicker, more helpful and more direct (Eightohtwo.com, 2026 AI Search Behaviour Study).
Google is not going anywhere
Google still processes 101 billion monthly visits. ChatGPT handles 5.63 billion. Gemini: 1.17 billion. Perplexity: 166 million. Google is not being replaced. Its behaviour is being changed from within.
The traffic AI does send is your highest-converting traffic
AI referrals convert up to 23x better than organic search. Similarweb’s 2026 data puts ChatGPT referral conversion at 7.1%, compared to 4.1% for organic and 3.2% for social.
Someone arriving from an AI tool has already done the research, comparison and shortlisting. They did not just find you; they were recommended to you.
AI is a discovery layer, not a replacement
85% of people who use AI for answers still verify them via traditional search (Eightohtwo.com, 2026). The customer journey now has two engines and brands need to show up in both.
Six insights from Future of Search 2026
1. Search has split into two distinct journeys
The discovery journey is increasingly happening through AI tools. The verification journey still happens in Google. The most important change required by any marketing leader is to stop treating SEO and AI search as competing priorities. They serve different moments in the same buyer journey.
A buyer in financial services might ask ChatGPT ‘what should I look for in an embedded payments provider?‘ a question they would never have typed into Google. They use AI to synthesise the landscape; they then Google specific brand names to confirm and verify, so your brand needs to appear in both moments.
2. Being citable by AI is a function of being credible
Suzannah Simpson, our Head of Organic Marketing, made a point that cuts through much of the tactical noise: being found by AI is not about gaming an algorithm, but about being credible, clear and citable across the web.
The GEO framework has three pillars:
- Content needs to be easy for LLMs to extract
- It needs to be super clear who you are and what you do
- You need to gain positive mentions outside your own website
Platform differences matter here. ChatGPT and Perplexity heavily favour Reddit content. Claude is completely different: 64% of its citations come from brand websites, and it does not cite Reddit at all. For financial services, B2B and professional services brands, website authority and LinkedIn content are some of the most valuable GEO signals.
Read more: Generative Engine Optimisation: A Complete Guide to Getting Found in AI Search
3. LinkedIn is now part of your search strategy
LinkedIn is now the second most cited domain in AI search responses, accounting for approximately 11% of AI answers. 95% of that cited content is original articles and posts written natively on the platform.
For senior marketing leaders who have treated LinkedIn as a brand-building channel separate from search: that distinction has dissolved. A 500 to 2,000-word LinkedIn article addressing the questions your buyers are asking AI tools is now a direct input into whether models cite your brand.
4. The phone call is not dying, it is becoming more intentional
The most counterintuitive data of both events came from Daniel Wilkinson, Chief Customer Experience Officer at Infinity, who presented findings from 41 million inbound calls tracked across Infinity’s customer base.
Call volume grew +24% in H1 2026 versus H1 2025. The visit-to-call rate grew +31% over 18 months, from 1.59% in January 2025 to 2.04% in June 2026 (Infinity QBR Machine v1.2). The step-changes correlate directly with major AI product launches.
The explanation: AI is pre-qualifying the caller. Someone who picks up the phone in 2026 has typically already researched on ChatGPT or Perplexity. They understand the product. They are calling because they are ready to commit.
When Infinity tracked 16,381 AI-attributed calls (direct click-throughs from AI platforms), those callers had a 73.6% long-call rate, which is equal to organic search. Their Smart Match conversion rate was 1.46 times higher than paid search. AI is not only sending browsing traffic, but also sending buyers.
5. The friction point has shifted downstream
In Q2 2025, calls failed primarily because buyers were not yet ready decision-making uncertainty was the top barrier at 20.8%, and product understanding difficulty was 17.7%.
By Q2 2026, the picture had changed substantially. Product understanding as a barrier fell 55%, from 17.7% to 7.9%. Price and budget barriers rose 77%, from 17.8% to 31.6%. Availability issues appeared as a new significant barrier at 18.3% (Infinity Smart Outcomes, Q2 2025 vs Q2 2026).
AI search did the whole education process. Buyers arrive in 2026 knowing what they want. The sales conversation has changed from “informing the customer” to matching and closing the deal.
6. Paid search is working. You are not measuring it fully.
PPC visitors call at 3.08% — 1.75 times the rate of organic search visitors. When they call, 71% stay on for more than a minute, the same quality signal as organic search. Display generates calls at 6.7 times less the rate of PPC. Social at 17 times less (Infinity Internal Metrics).
The goal is no longer to maximise clicks, but to maximise influence, and then measure it across multiple signals.
GEO in practice: How Splitit ranked #3 ahead of Stripe, PayPal and Affirm
To ground all of this in real outcomes, Robyn de Wet, our Senior Account Manager, and Robin Olsson, Director of Digital Marketing at Splitit, presented the results of a GEO programme we built together.
Splitit is an embedded payments provider competing in a space dominated by Stripe, PayPal and Affirm. The question was whether a challenger brand with less name recognition could compete in AI search, and whether doing so would drive commercial outcomes.
The approach started with a Deep Research (DRX) audit, using AI tools to investigate how ChatGPT, Perplexity and Gemini perceived Splitit. Content was restructured to be answer-first and AI-readable. A thought leadership programme followed, including LinkedIn-native articles, editorial placements and data-backed original research.
The results:
- #3 AI Overview ranking across tracked keywords — ahead of Stripe, PayPal and Affirm
- +229% LLM sessions year on year (89 to 293 monthly sessions)
- +300% qualified leads year on year
- Website lead to closed-won conversion: 0.5% to 2%
- -84% cost per click versus paid social account average
- +2,217% click-through rate on thought leadership content
The Splitit results demonstrate something important for any challenger brand: the brand’s AI tools recommend are not necessarily the biggest. But being the ones most clearly, credibly and consistently described across the web helps a lot.
Five practical actions for marketing teams
- Run a GEO audit before anything else. Understand your current AI search visibility before changing any content or strategy. Ask ChatGPT, Perplexity and Google’s AI Mode the questions your buyers are asking. See whether your brand appears, and how it is described when it does. Our free GEO audit does this for you.
- Structure content for extraction, not just reading. Stop the waffle. Answer the question in the first paragraph, not the fourth. Use H2 headings that match how buyers phrase queries in AI tools. Include statistics with named sources. This applies to existing content as much as new.
- Check your AI referrer data this week. Open your analytics platform and search for chatgpt.com, perplexity.ai and gemini.google.com in your referrer data. For most businesses, AI-referred traffic is already there and untracked. Segmenting it is a ten-minute task.
- Rethink what LinkedIn is for. LinkedIn articles on the questions your buyers research in AI tools are now a GEO signal. For B2B and professional services brands, this is recommendedof the most cost-effective GEO activities available.
- Build attribution from multiple signals. Combine web conversions with AI referral traffic, brand search demand trends, assisted conversions and, where relevant, call conversion data. You do not need a single model. You need a more complete picture than any one metric provides and look at all of them more holistically and as a trend.
The shape of the next 12 months
The central tension for marketing leaders in 2026 is between two things that are simultaneously true. AI is creating more complexity in measurement and attribution, and AI-referred traffic is your highest-converting traffic. The challenge is building the operational capability to capture the opportunity while managing the uncertainty.
We help clients build the kind of credibility that lasts, not algorithmic shortcuts, but a clear, consistent presence that earns trust from real buyers and from the AI tools those buyers are increasingly turning to.
Want to see how your brand currently appears in AI search?
Our free GEO audit shows you where you appear, how you are described, and where your competitors have an edge. Get in touch with us here.