The New Rules of Marketing in a Zero-Click, AI-Mediated World
As AI reshapes how consumers search and decide, marketers must move beyond clicks and MQLs to focus on visibility within AI-generated answers, redefining performance in a zero-click world.
Topics
If you are still pulling up your analytics dashboard every morning, poring over your Cost-Per-Click, and popping champagne over an uptick in Marketing Qualified Leads, we need to have a very difficult and brutally honest conversation.
The harsh reality of our industry right now is that performance marketing is rapidly losing its actual performance, and we are collectively sitting in a state of comfortable denial about it.
For the better part of a decade, we have operated under the comfortable delusion that because a digital touchpoint is measurable, it must be meaningful.
We built massive, convoluted attribution models that reward the final, trackable action—a retargeted ad, a branded search, a desperate promotional email offering 10% off—and we pat ourselves on the back for “driving” a sale that was almost certainly going to happen anyway.
The truth is that the modern customer journey has fundamentally broken the performance marketing playbook. The metrics we obsess over are artefacts of a bygone era of the internet, and the longer we cling to them, the faster we will make our brands completely irrelevant.
The Lie of “Personalisation” and the Actual Death of the MQL
Let’s start with the most obvious casualty of this massive behavioural shift: the Marketing Qualified Lead. Now, I want to be incredibly clear here, because I do not throw around the word “death” lightly. If you have been in the marketing space for any length of time, you know that our industry absolutely loves to cry wolf.
We all sat through the agonising, years-long saga of the “death of third-party cookies,” a supposed apocalypse that was delayed so many times it practically became an industry-wide punchline. And we have all chuckled at the annual think pieces declaring the “death of email,” a channel that stubbornly remains one of the highest-converting assets most brands actually own.
So when I tell you that the MQL is dead, I am not trying to be sensationalist just to get a reaction. I am telling you that it is functionally, structurally, and fundamentally obsolete.
For years, we worshipped the linear funnel. The neat, tidy narrative went that a user clicks an ad, we track that click with a pixel, they fill out a gated ebook form, they cross an arbitrary lead-scoring threshold in our CRM to become an MQL, and we drop them into a six-part automated email sequence.
But that is not how human beings buy things, and frankly, it never really was. We only structured our marketing that way because previous technology severely limited what we could actually do. We couldn’t genuinely understand individual intent or track complex, multi-layered conversations, so we settled for the absolute next best thing: tracking linear clicks and dropping people into software-defined buckets.
It is actually profoundly sad that the marketing industry’s working definition of “personalisation” is still just grouping people into broad segments.
We take a complex, multidimensional human being with highly specific pain points and reduce them to a label like “Mid-30s Urban Female Who Clicked a Skincare Ad.” That isn’t personalisation; that is just digital stereotyping.
We track a linear set of actions because our software demands a linear input, completely ignoring the fact that the actual buyer journey is a chaotic, non-linear web of private Slack groups, text messages with friends, Reddit rabbit holes, and deep dives into untrackable YouTube videos.
By the time someone finally clicks your tracked link, they have already made their purchasing decision. We are measuring the very end of a journey we had almost zero influence over, and we have the audacity to call it a performance marketing success.
The Illusion of the “Sales Handoff”
Because the MQL is dead, we also need to rethink the traditional “sales handoff.” Think about how insulting the modern B2B buying process is.
A prospect spends weeks doing their own independent research, asking AI agents for tech stack comparisons, reading reviews on G2, and consulting peers in dark social channels. They have high intent, they know what they want, and they finally come to your website to request a demo.
What happens next? They are forced to talk to a 23-year-old Sales Development Rep whose only job is to ask them qualification questions that the buyer already answered in their own head three weeks ago.
We are treating highly educated, highly contextualised buyers as if they know nothing, simply because they are just now entering our trackable funnel. When the entire pre-purchase journey happens in the black box of AI and community forums, your sales team can no longer act as gatekeepers of information.
They have to act as specialised consultants who assume the buyer already knows your basic feature set because ChatGPT already summarised it for them.
Addicted to Clicks in a Zero-Click World
This obsessive, compulsive need to track every single movement has completely blinded us to the most massive behavioural shift of the decade. We are moving into a true zero-click world, and the terrifying part is that most marketers have absolutely no idea how to optimise for a journey that doesn’t involve a landing page.
Clicks used to be the lifeblood of the entire digital economy. They were the most exciting metric a marketer could track because a click meant intent, and intent meant a potential conversion that you could proudly show your executives in a quarterly review.
But recent data shows us that roughly 60% of all searches now end without a single click to an external website. The majority of the time, a consumer asks a question online, they get their synthesised answer directly on the search engine results page or from an AI overview, and they leave without ever crossing your carefully optimised digital threshold.
Brands are actively hurting themselves by continuing to optimise their entire budgets around Click-Through-Rate and driving traffic that is rapidly becoming nothing more than a vanity metric.
When you base your entire strategy on capturing clicks that are actively evaporating from the ecosystem, you end up spending increasingly more money to fight over a rapidly shrinking piece of the pie. We have been so intensely focused on getting the click that we completely forgot how to provide value without it, and we do not know how to build brand affinity when we can’t cookie the user.
Welcome to the Black Box: Marketing to the Machine
The reason these clicks are vanishing is that the way customers search has permanently changed, and the way intelligent agents search is moving right along with it. We are entering an era where Large Language Models act as the ultimate, entirely opaque middleman between your brand and the consumer.
Take the real-world example of my recent skincare overhaul.
In the past, a shopper in their mid-30s looking to revamp their routine might have Googled “best anti-ageing serums for 30s,” clicked on three different sponsored listicles, read some reviews, and maybe clicked a retargeting ad a week later so some brand could gleefully claim them as an MQL.
Today, that journey looks entirely different. That same shopper opens ChatGPT, types out a highly specific prompt about their exact skin concerns, their age, and their preferred budget, and asks the AI to act as a personal dermatologist.
The AI instantly generates a highly individualised routine. The shopper then goes directly to the Amazon and Sephora apps on their phone, types in the exact product names the AI recommended, and buys everything in one fell swoop.
Where is the tracked “click” in that journey? How does a performance marketer attribute that massive cart value to a specific campaign?
They don’t. It is a complete black box. The AI synthesised the entire internet, made a recommendation based on its massive training data, and completely bypassed the traditional performance marketing funnel.
A Mandatory Mindset Shift: From Share of Search to Share of Model
If performance marketing is becoming less performant, and the customer journey is no longer a trackable, linear path of clicks, what should marketers actually be planning for?
It requires a total mindset shift. You need to stop exclusively worrying about your Share of Voice or your Share of Search, and start heavily factoring in your Share of Model.
Share of Model is not a silver bullet, but it is absolutely the new operational baseline you must understand. It measures how often, and how favourably, your brand is recommended when a consumer asks an AI agent a relevant question.
Traditional search engines were somewhat forgiving; if you weren’t the number one result, you could still scrape some decent traffic from the bottom of page one or the top of page two. But AI models are entirely ruthless.
There is no “page two” in a conversational interface. If the LLM does not recommend your product in its single, synthesised answer, your brand practically does not exist to that consumer in that moment.
To win Share of Model, you have to realise that context matters exponentially more than clicks, for both customers and agents. You are no longer just marketing to a human shopper; you are marketing to the AI agent that is advising them.
LLMs do not care about your retargeting pixels, your clever ad copy, or your gated ebooks that require a business email. They care about structure, authority, and whether your brand is consistently cited as the definitive, trusted answer to highly specific queries across the wider web.
The New KPIs: What We Actually Need to Measure Now
So if we are moving away from MQLs and vanity clicks, how do we actually measure success in an AI-mediated world? You need to start looking at entirely new frameworks that assess your brand’s authority within these black boxes.
First, you need to track your Inclusion Rate (or Mention Rate).
This is simply the percentage of relevant, high-intent prompts where the AI explicitly brings up your brand by name. If a user asks Claude to build a project management tech stack for a remote team of fifty people, does your software even make the list? If your Inclusion Rate is low, it means the AI doesn’t see you as a category leader.
Next, you have to look at your Citation Rate.
It is one thing for an AI to mention you; it is another entirely for the AI to link to your owned assets as the source of truth. If your Citation Rate is low but your Inclusion Rate is high, it means the AI knows who you are based on third-party reviews and forum chatter, but it absolutely does not trust your own website’s content enough to use it as a primary reference.
Finally, you need to audit for Resolution and Brand Sentiment.
When the AI talks about you, is it just spitting out vague marketing fluff, or is it providing a “high resolution” answer filled with specific data points, statistics, and verified use cases? LLMs love what researchers call “high-entropy” content—dense, factual, and highly contextual information. If the AI associates your brand with negative adjectives or superficial claims, you are losing the Share of Model battle before the customer even reads the response.
Who Is Actually Doing This Right? (The Best Buy Comeback)
If you want to see who is actually winning in this new environment, you need to look at brands that have successfully translated their inventory and reputation into deep, highly structured digital signals that AI can easily read.
Take a look at Best Buy. They are staging an absolute masterclass of a comeback in the AI era. According to recent BrightEdge data analysed by Search Engine Land, ecommerce results show a clear winner for AI engine citations: Best Buy completely dominates both Google’s AI Overviews and ChatGPT.
Why are LLMs prioritising Best Buy over so many other massive retailers? Because Best Buy isn’t just treating their website like a digital catalogue; they are treating it like an information architecture powerhouse. They provide highly structured product data, massive repositories of verified user reviews, incredibly detailed technical specifications, and easy-to-parse comparison charts.
When a user asks ChatGPT, “What is the best 65-inch OLED TV for a bright living room under $2,000?”, the AI needs high-entropy, contextual data to formulate an answer. Best Buy serves that data up on a silver platter, making them the ultimate authoritative citation for the machine.
Nike is another fantastic example. They dominate the AI fitness space across multiple models because they have cultivated massive amounts of user-generated content on platforms like Reddit, Strava, and running blogs.
They don’t just talk about “running shoes” in broad strokes; their digital footprint is deeply contextualised around highly specific use cases, like “best carbon-plated shoes for marathon training under a six-minute pace.” LLMs eat that granular context for breakfast.
Or look at the travel industry. Booking.com, which I had really never used previously, has consistently been dominating AI travel recommendations across different models because they have optimised for the exact characteristics that LLMs prioritise when answering complex travel queries: convenience, variety, and structured spatial data.
They don’t just sell hotel rooms; they provide a massive database of interconnected, highly structured information that the AI can confidently synthesise for a user planning a multi-city European vacation.
GNC is a surprise “comeback brand” as well!
What Else Do Marketers Need to Start Thinking About Today?
If the middle of the funnel is becoming an untrackable black box, marketers have to completely rethink their overarching philosophy. Here is what else you need to be deeply considering as you plan for the next iteration of the internet.
Information Architecture is the New Creative. We spend so much time obsessing over the colour of a call-to-action button or the cleverness of an ad headline, but AI agents don’t care about your graphic design. They care about how your data is structured.
If your website is a beautiful but unreadable mess of heavy JavaScript and buried PDFs, the AI will simply bypass you for a competitor whose site is logically mapped and clearly contextualised. Marketers need to stop acting solely like magazine editors and start acting a lot more like digital librarians.
Embrace the “How Did You Hear About Us?” Metric. We have been spoiled by the false comfort blanket of software attribution. We wanted a dashboard to tell us exactly which ad drove which dollar. In a zero-click, dark social world, you have to become comfortable with the messy reality of self-reported attribution.
You need to start aggressively asking your customers, “How did you hear about us?” on high-intent forms. You will be shocked at how many of them type in “ChatGPT,” “a private Slack group,” or “my friend sent me a text.” Your software will never track those channels, which means if you don’t ask the customer directly, you will misattribute the sale and optimise your budget against a lie.
Stop Hoarding Content Behind Forms. For a decade, content was just bait. It was a shiny object we dangled in front of a user to trick them into giving us their email address. In an AI-driven world, your content is training data. If you lock your best insights, your deepest research, and your most valuable perspectives behind an MQL form, the LLMs cannot crawl it, read it, or learn from it.
And if the LLMs can’t learn from you, they will learn from your competitors who are giving their knowledge away for free. We have to transition from using content to capture demand, to using content to feed the machines that will ultimately recommend us.
Community is the Only Algorithm You Can Control. Because people know that traditional search results are gamed and that AI can occasionally hallucinate, they are retreating into private communities to verify information with actual human beings.
Specialised group chats and niche forums are where the actual, final buying decisions are being made. Marketers need to stop treating these spaces like distribution channels to dump their links into, and start treating them as places to genuinely participate, build trust, and foster word-of-mouth.
We are leaving the era of the tracked click and entering the era of the conversational prompt. Stop trying to force buyers into rigid, trackable cohorts just to satisfy an outdated analytics dashboard.
Start focusing on being the brand that both human beings and AI agents implicitly trust, even if you can’t perfectly measure the exact path they took to get to your checkout page.
ALSO READ: Connected Packaging Goes Big with Gamified Brand Storytelling
