AI search is all anyone's talking about at Cannes. We've been talking about it since Brighton SEO 2025.

FYI: This post is a follow up to our 2025 blog post on 'The Search Everywhere Trfecta'.

It's Cannes Lions week, and if you've been at the event (or like me you have been sat with envy reading posts here on LinkedIn) one conversation has clearly dominated: AI search.

From, How do brands get found inside ChatGPT, Perplexity, Gemini? through to chatter on Who owns it? Where does it live? AI is once again having a bit of a moment.

Here's the thing: it's the wrong starting point.

In 2026, the biggest challenge for digital teams isn't visibility, it's the silos stopping them from achieving it. SEO, brand, and social teams running their own agendas, chasing their own metrics, missing the compounding power they'd unlock working as one. AI search doesn't change that problem. It amplifies it.

At Deviation, we call the solution the Search Everywhere® Trifecta — treating Search, Brand, and Social not as separate disciplines, but as a unified discoverability engine. When these teams align, brands stop being merely seen and start being preferred.

And in 2026, there's a fourth dimension sitting across all three: AI. ChatGPT, Gemini, Perplexity, Claude, and AI Overviews are now primary discovery surfaces for millions of purchase decisions. Brands that earn citations — and more importantly, recommendations — inside AI responses will compound their advantage. The teams that figure out how to embed this into existing workflows will pull away from competitors still measuring keyword rankings in isolation.

The conversation at Cannes is the right one. This newsletter exists to go further than the panel.

Below is our 9-step checklist, built on our search everywhere framework, including an LLM discovery layer you can deploy right away.

The 9-step checklist

1: Audit current discovery signals

Map how your brand is being discovered across search engines, social platforms, and online communities. Where are people first hearing about you? Where are they searching before they buy? But now you need to ad one new audit dimension: run your category queries, ask key pains/needs/wants/desires through ChatGPT, Perplexity, and Google’s AI Overviews. Note which brands get cited and which don’t… that gap is an LLM opportunity.

LLM discovery angle: AI models surface brands based on the breadth and authority of their web presence - structured data (debated), editorial mentions, social mentions, news presence, and Q&A content all feed LLM training data. Audit these sources alongside your traditional search and social audit.

2: Identify siloed metrics

Before you can align, you need to see the walls.

List every metric each team is chasing. SEO rankings, PR coverage, influencer engagement, paid ROAS. Write them all down in one place, probably for the first time. Then ask the harder question: who gets rewarded for what?

Metrics are tied to performance reviews. Teams aren’t doing anything wrong, they’re responding rationally to the incentives they were given. The problem is that rational individual behaviour creates irrational collective outcomes. Social builds audience around topics SEO isn’t targeting. PR earns links to pages with no commercial intent. And nobody is producing the content that earns recommendations inside AI answers to the maximum potential - due to AI/LLMs leaning on all of the above.

LLM citation (but more importantly recommendation) frequency, how often your brand appears unprompted in ChatGPT, Perplexity, or Gemini, is a fast-emerging indicator of brand authority that currently belongs to no team (we’re gonna address this).

Mapping all of this before you align on shared KPIs isn’t a quirky admin task. It will give you the ability to diagnose your digital team. When people see where their work connects, competes, and falls through the cracks, alignment stops feeling like a reorganisation and starts feeling like an obvious fix.

3: Align on shared KPIs

Now you understand where your individual focuses are and where you have overlaps, its time to agree on joint measures of success. Instead of only tracking keyword positions or press mentions, focus on outcomes that cut across silos:

[Brand] Search & [Brand] + [Term] Search growth: more people typing your brand name directly into Google. A proxy for awareness converting to intent; someone has heard of you and wants to find you specifically.

Preference signals: Behavioural indicators that someone chose you over an alternative. Direct traffic, return visits, brand mentions in reviews and communities, unprompted recommendations. The difference between being known and being wanted.

Discovery mentions: Instances of your brand appearing in conversations you didn't initiate or pay for. Reddit threads, TikTok comments, forum posts, journalist roundups, AI citations. Where organic word-of-mouth lives.

Cultural relevance: Whether your brand is being referenced in the context of ideas, movements, or moments that matter to your audience. Not just ‘people know [BRAND]’ but [BRAND] becomes synonymous with a conversation… FYI, I would argue this is The hardest to manufacture, the most durable when earned.

LLM discovery angle: Add “AI recommendation share” to your shared KPI set, tracking how often your brand is mentioned unprompted inside AI-generated answers for category queries. Pairing citation/recommendation share with branded search volume then reveals whether LLM exposure is translating to real-world demand.

4: Build cross-channel dashboards

Nothing drives collaboration like a single source of truth. Create a dashboard that integrates SEO, Brand, and Social metrics so everyone sees how their actions amplify each other. Build an LLM layer into this dashboard from day one, even a simple tracker showing citation/recommendation frequency across major AI platforms tells a compelling story to leadership and earns buy-in for the broader framework. And if you can reflect AI/LLM site visits and conversions too - winner winner!

5: Run search-informed experiments

Take one campaign and let SEO insights fuel brand and social work. Use trending queries to brief creators on relevant social platforms for your audience. Measure impact on both social engagement and search demand.

LLM discovery angle: Extend the experiment: publish a structured long-form piece (FAQ, comparison guide, or expert explainer) targeting the same conversation focus. LLMs heavily favour comprehensive, well-structured content that directly answers questions, and adds to the conversation. A piece that ranks well on Google and earns editorial links is also the piece most likely to be surfaced inside AI answers… one effort, compounding returns across three channels.

6: Share cultural insights widely

Search teams often sit on gold without realising it. Reddit threads, TikTok trends, and AI citation sources reveal where conversations are moving. Package these into insight packs for brand and social colleagues, it’s how SEO earns a seat at the strategy table rather than being handed a brief after the fact.

LLM discovery angle: Interrogate AI platforms directly in your insight-gathering process. Ask ChatGPT or Perplexity how they’d answer your target queries and which brands or sources they cite. The sources that LLMs draw from, authoritative publications, Reddit, Social content, YouTube, and industry wikis - tell you exactly where to build presence and earn links.

7: Celebrate compound wins

Don’t just report on your own lane. Document and circulate case studies where collaboration worked: a TikTok trend that spiked Google searches, or PR coverage that boosted branded search and AI citations simultaneously.

LLM discovery angle: The most powerful compound win to document in 2026: a PR story picked up by a major editorial outlet that earns a backlink/brand mention, facilitates a spike in branded search, and a new AI citation is achieved in the same week. When leadership sees that three-way flywheel in a single slide, search everywhere starts to look MEGA!

8: Formalise collaboration rituals

Make it structural. Set up monthly “Digital Team Syncs” and ‘Strategy Refresh Sessions’ for key times in the year where Search, Brand, and Social review each other’s priorities, campaigns, and results. Over time it becomes muscle memory, and the cross-pollination above happens naturally rather than by exception.

LLM discovery angle: Add a standing 10-minute “AI visibility review” to your monthly sync. What queries are competitors being cited for that you’re not? Which pieces of your content have earned new AI citations? This keeps LLM discoverability visible as a live competitive lever, not a quarterly report afterthought.

9: Make “preference” the north star

Visibility is no longer enough. The shared goal should be to become the preferred option. Whether it’s SEO content, a PR story, or a TikTok post, every tactic should point toward making your brand the default choice in the customer’s mind. This is what winning looks like in the future of search and its why its the goal of any ‘Search Everywhere’ approach.

LLM discovery angle: LLMs are preference engines. When a user asks “what’s the best technical outdoor brand for cold-weather expeditions?”, the model synthesises signals from across the web — editorial authority, community sentiment, structured product data, and brand narrative — to make a recommendation. Every piece of aligned, authoritative content your teams create is a vote cast for your brand inside that synthesis. Preference in 2026 isn’t just built in people’s minds; it’s built in the training data.

How Search Everywhere® directly powers LLM discovery

The Search Everywhere® Trifecta was built before LLMs became a primary discovery surface, but its logic maps almost perfectly onto what makes a brand discoverable by AI. Here’s why each pillar matters in the LLM era.

• Search signals feed training data: Well-structured, high-authority content that earns backlinks and engagement is disproportionately represented in LLM training corpora. Your SEO investment directly builds AI citation potential.
• Brand narrative shapes AI answers: PR, editorial coverage, and thought leadership create the consistent brand narrative that LLMs draw from when generating recommendations. An LLM can only describe your brand in the terms your own content and coverage provide.
• Social conversations inform sentiment: Reddit, YouTube, and TikTok are crawled by AI platforms. Community-level positive sentiment and authentic product discussion directly influence whether an LLM recommends your brand over a competitor.
• Structured content wins citations: FAQ pages, comparison guides, glossaries, and “best for” content are highly LLM-retrievable formats. Brief your content team to produce one structured asset per campaign alongside the hero piece.
• Schema markup aids AI parsing: Structured data — Product, FAQPage, HowTo, Organisation — makes it easier for LLMs to parse and surface your content accurately. An often-missed quick win for SEO and AI discoverability simultaneously. THIS POINT IS DEBATED, ITS MY OPINION THAT IT CAN’T HURT AT THE VERY LEAST!
• Consistency compounds across models: Different LLMs (ChatGPT, Gemini, Perplexity, Claude) pull from overlapping but distinct sources. A brand with consistent messaging across editorial, social, and owned channels is cited accurately across all of them.

Why this matters in 2026

Search isn’t just a platform anymore, it’s a behaviour. Consumers discover brands across TikTok, Reddit, YouTube, Discord, Google, and increasingly inside AI chat interfaces. If your teams aren’t connected, you’ll miss where demand is shifting.

The Search Everywhere® Trifecta is no longer a three-discipline framework... Instead, it has to now be 'Search. Brand. Social AND LLM.' Remove any one of them and the engine loses efficiency. Run all four in alignment and you don’t just get visibility: you become the brand an AI recommends before the customer even reaches your website.

Use this checklist to start breaking down silos, aligning signals, and building the kind of brand preference that machines and humans reach the same conclusion about.

‍