How Fortune 500 Brands Dominate AI Search Results

Are your competitors already showing up in ChatGPT answers — and you’re not sure why?

You’re not imagining the shift. With around 800 million people using ChatGPT on a weekly basis — and that figure doesn’t account for Gemini, Perplexity, Claude, Grok, or Microsoft Copilot — AI search has moved from an emerging channel to mainstream buying behaviour.

The problem is that most marketing teams are still measuring it with the wrong tools, optimising it with the wrong tactics, and, in many cases, not tracking it at all.

In this webinar, Charlie Marchant, CEO here at Exposure Ninja, sat down with Josh Blyskal, strategist and researcher at Profound — an enterprise-level AI tracking platform — to dig into what Fortune 500 brands are actually doing to dominate AI search results. Josh’s findings are drawn from the analysis of 40 million AI search results, plus direct work with brands including Bosch, Expedia, and Chanel.

This article pulls out everything that matters: the structural differences between SEO and AI search, the content and PR moves that drive citations, the stats that will help you get board-level buy-in, and the three metrics you should be tracking right now. Whether you’re building a strategy from scratch or trying to make the case internally, there’s a clear place to start.

Executive Summary

• Fortune 500 brands have a head start — not because of domain authority, but because they have a large library of existing content that can be refactored for Answer Engine Optimisation (AEO), or AI Search Optimisation. Brands newer to content need to start building that context urgently.
• Traditional authority signals are collapsing — ChatGPT doesn’t have access to Google’s data, so backlinks and traffic volume carry far less weight than they do in organic search. Pure utility and precise, structured answers are what get cited.
• Listicle and comparative content dominates — analysis of 40 million AI search results found that listicle and comparative content accounts for the largest share of all citations, with blog content a distant second. Reddit and Wikipedia are both surging in citation frequency.
• Earned media makes up around 80% of what gets cited — but it turns over at roughly 50% per month. Your owned content is your fortress; everything outside it is a fast-moving, volatile ecosystem that rewards recency and agility.
• AI is already influencing purchasing decisions at scale — 90% of B2B buyers have consulted AI search during a buying journey, and 80% of shoppers who used AI to research a purchase attributed more than half their decision-making to it. Attribution tools just aren’t capturing it yet.

Do Fortune 500 Brands Have an Unfair Advantage in AI Search?

The short answer is yes — but not for the reasons you might expect. It has nothing to do with how much they spend on paid media, or even their current SEO performance. The advantage comes from volume and history.

“Fortune 500 brands have the opportunity to leverage a lot of the training data about them in answer engines,” Josh explained. “You kind of have this basis of information — it is a blessing. It can be a curse at times if you’re trying to rebrand or drastically change what your company’s known for.”

That last point matters more than it first appears. If your brand has been telling a consistent story for years, the accumulated web of articles, reviews, press coverage, and owned content works in your favour. AI models have more material to draw on and more consistency in that material. For brands that have been trying to shift positioning — rebranding, repricing, repositioning a product — older content can actively work against you. Josh noted that articles from 2018 still appear regularly in AI-generated answers for major brands.

The practical implication for mid-market and enterprise teams isn’t to panic about legacy content — it’s to recognise that a large content library is an asset if it’s actively maintained and optimised for how answer engines work today. The real opportunity is in what Josh calls “refactoring”: taking existing content and restructuring it to be hyper-specific and useful for AI search, rather than starting from scratch.

For brands with smaller content footprints, the starting point is different — but the path is clearer. The gate to AI search citation isn’t guarded by domain authority. It’s open to anyone who can provide the most precise, useful answer to a specific question.

How Is AI Search Different from Traditional SEO?

The single biggest shift, according to Josh’s research, is the collapse of traditional authority signals.

“You can go to Semrush*, look at the top ten cited pages in an industry — the number one cited source for fast food for a few months was some random guy who wrote an incredibly detailed blog of all the different fast food burgers he’d had,” he said. “Not well ranked, not well cited on Google. But the answer engines latched on and said, this content is absolutely incredible.”

That’s the key difference. ChatGPT’s search component — a layer called WebGPT — doesn’t have access to Google’s signals. It doesn’t know which sites have higher domain authority, more backlinks, or better traffic. It takes a user’s query, restructures it into five to ten keywords, passes those to a search API, and then picks whichever source best answers the specific question. The result is a flight towards pure utility: the page that gives the clearest, most complete answer wins, regardless of whether it’s well-trafficked or commercially significant.

That said, Josh was clear that SEO and AEO aren’t separate disciplines. “At the core, we are using the same muscle groups — URL slugs, title tags, meta descriptions — but we’re doing different exercises with those same muscle groups now.” A few of the specific differences that change the game:

• JavaScript is invisible to answer engines. If your blog or product pages render content via JavaScript, answer engines see a blank page. On-page HTML is what gets read, parsed, and cited. This is a critical technical issue that many large sites haven’t addressed.
• Dates matter more than they do in SEO. Answer engines are biased towards recency. Including the year in the URL slug, the title tag, and the opening 100 characters of a page gives meaningful signals — even if it looks odd from a traditional SEO perspective.
• Semantic chunking changes how you structure content. Rather than narrative-led introductions, effective AEO content leads with a direct answer. “The average rent in San Francisco is £2,700 for a studio” outperforms “San Francisco has many wonderful neighbourhoods with a variety of rental options.” One answers the question; one doesn’t.
• FAQ schema and structured data are more important than ever. Answer engines reward content that has been broken into discrete, question-answerable units. Full question phrasing in H2s, HTML tables, and structured data all help answer engines parse context quickly.

For marketers already skilled in content and SEO, this is less a new discipline and more a recalibration. The fundamentals are familiar. The priorities are different.

What Are the Biggest Brands Actually Doing to Win AI Citations?

Josh’s work with Fortune 500 companies reveals a consistent pattern: the brands performing best in AI search aren’t doing something radically new. They’re doing familiar things — content, PR, structured data — with a much sharper focus on what answer engines actually need.

A few of the specific plays that are working:

• Enriching product and listing pages with machine-readable context. Most brands assume product pages are utilitarian — there for transactional queries, not informational ones. That assumption is wrong. When someone asks ChatGPT “what are the best noise-cancelling headphones for commuting,” it’s answering from product page content. Brands that have contextualised their products — use cases, comparisons, technical specifications, persona-relevant benefits — are appearing in those answers. Brands that haven’t are invisible.
• Restructuring informational content around specific questions. Josh described this as going further than generic FAQs: “Can we hypercontextualise these to be semantic prompts — using full questions in our H2s, thinking about what specific answer an engine needs for a specific kind of query?” The goal is one URL per distinct topic, with that URL being the most precise answer available for that specific question.
• Treating editorial coverage as citation-building rather than link-building. Chanel is a strong example here — third-party editorial coverage, owned heritage content, and PR all working together. The result is brand context appearing across a wide range of sources that answer engines draw from. For brands without Chanel’s heritage, the principle still applies: getting your brand story told, in detail, across as many relevant sources as possible is what builds AI search presence.

There’s an important nuance on content volume. The instinct to “write everything on the page” isn’t wrong in principle, but the way it tends to be executed usually is. Josh’s advice on using AI in content creation is worth taking seriously: “Your AI-written content should be like stringing beads onto a necklace. The beads are the value — the data, the comparisons, the pricing tiers, the research. AI’s job is to thread the string between them.”

The content that wins is built from proprietary data, first-hand comparisons, and specific facts. The writing is a vehicle for that value, not a substitute for it.

Why Does Reddit and Earned Media Matter More Than You Think

One of the more surprising findings from Josh’s research is the scale of earned media’s role in AI citations — and how volatile it is.

“About 80% of what gets cited is earned,” he said. “Your owned content is the fortress. But the sheer volume of what’s happening outside it means you can’t ignore the earned environment.”

Reddit’s rise in AI citations has been particularly sharp. “Reddit was up 80-something percent in the last month in terms of citations. Wikipedia was up 67%.” For many enterprise brands, Reddit has barely featured in the social or digital PR strategy. The forums exist, conversations happen about products and pricing, and until recently it largely didn’t matter whether those conversations were positive or accurate. Now it does.

The broader picture is one of constant flux. At a page-by-page level, roughly 50% of citations turn over from one month to the next — sources that were driving AI mentions for a brand one month may have been displaced by something newer and more specific the following month. That’s partly because recency is one of the most underrated ranking factors in AI search. Answer engines are biased towards the most recently published answer to a specific question.

What this means practically:

• You can’t treat earned media as a “set it and forget it” channel for AI search. The sources that matter shift regularly.
• Knowing which sources are being cited for your category — and actively pursuing coverage on those sources — is more valuable than a broad scattergun PR approach.
• Reddit conversations about your brand, products, and pricing are now effectively part of your AI search profile. If you’re not monitoring them, you don’t know what AI is saying about you.
• A tool like Profound, or even a simple spreadsheet of manual prompt tests, lets you identify which sources are driving citations in your category right now — and gives you a PR targeting list to work from.

The good news is that this volatility cuts both ways. It means a well-placed piece of content, or a strong piece of editorial coverage, can move the needle faster than almost anything in traditional SEO.

How Do You Make the Case for AI Search Investment Internally?

For marketing directors and VPs trying to get budget and buy-in for AI search, the challenge isn’t usually understanding its importance — it’s proving it to people who aren’t yet convinced. Josh’s research gives you the numbers to do that.

• 80% of people who purchased something having used AI in their research attributed more than half of their decision-making to AI search.
• 63% of shoppers using AI search self-reported that AI models were a key influence on their decisions.
• 90% of B2B buyers have consulted AI search at some point during their buying journey.
• 6 in 10 consumers are using AI search. A Forrester study supports this at scale.

“Everyone is attributing this as direct traffic,” Josh noted. “You find the thing you want to buy in ChatGPT, you copy the name, you open a new tab, you paste it into Google, you go and buy the thing. And it shows up in analytics as search or direct.”

This is one of the clearest symptoms that AI search is already affecting your business — it’s just not showing up in your attribution model yet. Adding “ChatGPT” (and other AI platforms) to your “how did you hear about us” form is a low-effort first step that tends to produce surprising results. Profound found that 10–15% of their own inbound leads now cite ChatGPT.

For teams facing internal resistance to experimentation on the main site, there’s a tactical workaround worth knowing about. You can create content that’s indexed only by answer engine crawlers — de-indexed from Google, but accessible to Perplexity, Gemini, or Copilot — to run a contained experiment without touching the main site’s SEO profile. It’s not a long-term solution, but it can generate the early proof points that get bigger experiments approved.

As Josh put it: “The number one factor in whether someone is going to be successful is that first experiment. Once you do your first experiment and get a tangible gain in visibility, you want to do another.” The goal is to build momentum, not to have a definitive strategy before you start.

If you want a clearer view of what your AI search presence looks like right now — and what prompts you should be tracking before you start testing — request a free marketing review from Exposure Ninja and we’ll help you map the starting point.

What Should You Be Tracking in AI Search

One of the most common questions from marketers new to AI search is simply: what do I measure? Traditional search analytics don’t capture it, and the attribution loop is largely broken. Josh’s answer centres on three metrics.

Referral traffic from AI sources. This is your baseline barometer. Even in a world where click-through rates from AI search are low, referral traffic tells you how often your site is appearing as an inline citation — the small linked chicklet within an AI answer. “Even one referral click from AI search means you’re showing up in a bunch of different prompts and questions around the world,” Josh said. Track the trend, not the absolute number, and resist the temptation to compare it directly to organic traffic volume.

Citation share. This measures how often your site is being cited in response to a specific set of queries — typically branded or category-level questions. It’s the right metric when you want to track ownership of a particular topic or product area. If you want to know whether your help centre content is authoritatively answering “how do I set up [your product],” citation share tells you that.

Brand visibility. This is the broader measure: how often your brand name appears in AI answers across unbranded, category-level queries. “Where do I get the best noise-cancelling headphones?” — does your brand come up? This is your equivalent of organic visibility in traditional SEO, and it’s the most strategic of the three metrics to track over time.

AI traffic as a percentage of total traffic currently sits at 2–5% for most brands. For more tech-forward audiences — companies selling cloud infrastructure, developer tools, or AI products — it can reach 10% or more. For traditional industrials and local services, it may be closer to 0.2%. The percentage matters less than the direction of travel, which is up, consistently, across almost every sector.

Next Steps

This Week

• Add “ChatGPT” and other AI platforms to your “how did you hear about us” form, if you have one. You’re likely already getting AI-attributed leads that aren’t showing up in your data.
• Run ten to fifteen prompts relevant to your category and products across ChatGPT and Perplexity. Note which sources get cited, which brands appear, and whether you’re visible. Do this manually if you don’t have a tracking tool — a Google Sheet works fine to start.
• Check your site for JavaScript-rendered content. If key pages — product listings, blog content, service pages — are rendering via JavaScript, answer engines are seeing blank pages. This is often a quick fix with significant impact.

Next 30 Days

• Identify the two or three queries in your category where a competitor or a third-party source is consistently cited instead of you. Build one piece of content specifically designed to answer those questions — structured with direct answers, relevant data, and question-phrased H2s.
• Audit your product or service pages for contextual depth. Are you describing who your product is for, how it compares to alternatives, and what problem it solves? If not, that’s your first on-site optimisation priority.
• Map the sources that are being cited in your category. These are your PR targets. If you can get editorial coverage on those sources — or create something that performs similarly — you’re building citation presence directly.
• Look at what Reddit threads exist about your brand and products. If there are negative threads influencing AI sentiment (pricing complaints, poor reviews), decide whether there’s an action to take — addressing the underlying issue often updates AI perception within a few weeks.

Next 90 Days

• Build a tracking framework. Whether you use Profound or a manual spreadsheet, you need a consistent set of prompts you’re running regularly — so you can measure movement over time rather than taking one-off snapshots.
• Integrate AI search into your content planning process. Before any piece of content is commissioned, run the target query in the major AI engines. If someone else is already answering it well, your brief needs to be more specific, more data-led, or targeting a more precise sub-question.
• Develop an earned media strategy with AI citations in mind. Rather than a broad PR push, focus on the specific sources that are being cited for your category — and build relationships or pitches directed at those publications.
• Present your first experiment results internally. Even a small visibility gain from a single piece of content is the starting point for a bigger conversation about resource and investment.

If you want expert support building and executing this plan, request a free marketing review from Exposure Ninja. We’ll assess your current AI search visibility, identify the highest-impact starting points, and help you build a roadmap that’s right for your business.

Resources Mentioned

• Profound — Enterprise AI tracking platform that monitors citation share, brand visibility, and referral traffic across major AI search engines. Includes prompt volume data — the equivalent of keyword research, but for AI search queries.
• Semrush — SEO and search analytics platform referenced for analysing top-cited pages within an industry vertical.
• Zero Click 2025 — Profound’s in-person AI search event in New York on 8 October 2025. Described as the first AI search-focused industry event of its kind. Limited seats; no virtual option or recording.
• Forrester study on B2B buyer behaviour — Cited for the statistic that 90% of B2B buyers have consulted AI search at some point during the buying journey.
• Josh Blyskal on LinkedIn — Josh shares research findings and raw data from Profound’s analysis on LinkedIn. Recommended for anyone tracking the AI search space closely.
• BrightonSEO — Industry SEO conference where Josh presented findings from his 40-million-result AI search analysis. Josh will be speaking at the US edition (San Diego) and the UK edition in April.

In Conclusion

The brands winning in AI search aren’t doing so by accident, and they’re not doing it with tactics that are fundamentally alien to good marketers. They’re creating highly specific, well-structured content that directly answers questions. They’re getting their brand story told across the sources that AI engines actually draw from. And they’re moving quickly enough to stay relevant in an environment where 50% of citations turn over every month.

What they have that most mid-market teams don’t yet is a tracking infrastructure that tells them where they stand, what’s working, and where to focus next. That’s the gap to close first — even if your initial tracking is manual. You can’t improve what you can’t measure, and right now most businesses are flying blind on a channel that’s already influencing how their customers buy.

The attribution data will catch up eventually. The businesses that start building AI search presence now will be the ones with the advantage when it does. If you’d like help working out where to start, request a free marketing review from Exposure Ninja.

Is your marketing underperforming?

Request a website and marketing review and our team will tell you how to improve your marketing.

Request a Free Review

*Disclaimer: Exposure Ninja may get a commission through the marked links above, at no cost to you