The Invisible Funnel: How AI Is Rewriting the Buyer Journey

Think about the last product or service you bought. 

It’s highly unlikely that you just woke up one morning and bought the first thing that came into your head. There was a process – a journey of discovery, research, comparison, and ultimately, commitment. 

That journey is as old as commerce itself. 

However, with the introduction of AI, this process is now happening somewhere different.

Back to basics: Marketing 101

To understand where AI is disrupting consumer behaviour, we need to revisit something foundational.

More than a century ago, philosopher John Dewey introduced what became known as the Five Stage Model of Consumer Behaviour, which is often referred to as the Buyer Decision Process.

Despite everything that’s changed since then, the psychology that drives our decision-making is still the same. 

People are still people. We still experience problems, look for solutions and compare options. Ultimately, we commit, and afterwards, we still decide whether it was worth it.

There’s nothing revolutionary about this framework. What is revolutionary is how clearly it shows us where AI is reshaping the game. 

The stages haven’t changed, but the infrastructure has.

A personal example

Recently, my pair of cheap earbuds went through the washing machine, rendering them useless. I had arrived at Stage One. I had a problem, and I required new wireless earbuds. 

Stages Two and Three are where things get interesting, and where marketers are feeling this disruption most acutely.

In a traditional search-centric world, my information research would have taken me to Google. I might have typed in “best wireless earbuds” or “are wireless earbuds worth it” and landed on a series of articles, comparison guides, and brand pages crafted precisely to capture that search intent. Marketers spent years (and significant budget) creating content to show up at exactly this moment.

But this was shifting long before ChatGPT burst onto the scene in late 2022. The concept of “search everywhere optimisation” had already entered the marketing lexicon. Consumers weren’t just searching Google. They were going to YouTube for expert reviews, Reddit for real-world user experiences, Instagram for aesthetic inspiration, TikTok for short-form comparisons. The search journey had fragmented across platforms.

The introduction of AI has accelerated this fragmentation into something else entirely.

The shifting search landscape

Estimates suggest that nearly 84% of knowledge workers in New Zealand now use generative AI regularly. A study by Gartner anticipates that traditional search engine volume will drop by 25% in 2026. These statistics represent a fundamental shift in how people access information.

In my earbuds journey, I might now open an AI assistant and ask: “What are the best wireless earbuds for someone who commutes daily, exercises occasionally, and wants good noise cancellation, for under $300?”

In one conversation, the AI synthesises information that previously required a dozen separate searches across multiple platforms. It presents a shortlist tailored to my specific context, drawing on product reviews, forum discussions, expert comparisons, and manufacturer specs it has already consumed and indexed.

My journey through Stages Two and Three now happens inside an AI interface. 

Here’s the problem for brands: you can’t see it.

The invisible funnel

Traditionally, a brand selling wireless earbuds would invest heavily in content designed to capture search traffic across each stage of the funnel.

At Stage Two (Information Research), this might look like:

• “How do wireless earbuds work?”
• “Wired vs wireless earbuds”
• “What is active noise cancellation?”
• “Best earbuds for iPhone”

At Stage Three (Evaluation of Alternatives):

• “Sony WF-1000XM5 vs Bose QuietComfort Ultra Earbuds”
• “Best noise-cancelling earbuds 2026”
• “AirPods Pro review”

Each of these queries represented a traceable touchpoint. A user clicked a link, landed on a page, and a marketer got data: impressions, clicks, time on page, conversions. The funnel was measurable.

Now, a consumer might ask all of those questions in a single AI conversation. The AI has already consumed the content – your articles, your competitor’s guides, Reddit threads, YouTube transcripts, review aggregators – and is synthesising an answer on your behalf.

You only find out you existed in that journey when the consumer arrives at Stage Four, types your brand or product name into a browser, and makes a purchase.

The top-of-funnel traffic isn’t disappearing – the decisions are still being made – they’re just being made somewhere you can’t see.

What this means for your marketing

The instinct many marketers have right now is to panic. If AI is eating our search traffic, what’s the point of content?

The answer is that the point of content has never been more important, but the audience has changed.

You’re no longer writing only for humans on a search results page. You’re writing for AI models that are ingesting, evaluating, and redistributing information on behalf of those humans. The principles of good content – experience, expertise, authoritativeness, and trustworthiness – haven’t changed. But the distribution channels, and the way your content gets found, have.

Here’s how to think about it:

Content quality still wins – but for a new judge

AI models are sophisticated evaluators of content quality and authority. Thin, keyword-stuffed articles designed to game an algorithm will serve you even less well in an AI-mediated world. Deep, genuinely helpful content that answers real questions comprehensively is what gets referenced and cited.

Brand presence across platforms matters more than ever

AI doesn’t build its understanding of your brand from your website alone. It draws on everything: your Google Business profile, social media presence, media coverage, product listings, customer reviews, forum mentions and third-party comparisons. A brand that exists richly and consistently across the internet is a brand that AI can confidently represent to a consumer.

Earned media and third-party validation are critical

When a consumer asks an AI assistant for the best wireless earbuds, the AI is more likely to recommend brands that appear authoritatively across multiple independent sources. Reviews, editorial mentions, expert comparisons, and genuine community discussions all contribute to the picture AI builds of your brand.

Owned channels remain your conversion anchor

While the top of the funnel is increasingly invisible, Stage Four (Purchase Decision) still happens on your terms. Your website, your checkout experience, your pricing clarity, your trust signals: these matter enormously. When a consumer arrives at your site primed by an AI recommendation, they are ready to convert.

Hedge your bets across platforms

Instagram, YouTube, news and media publications, Reddit, TikTok, podcasts – all of these feed AI’s understanding of the world. A marketing strategy that treats SEO for Google as the only game in town is a strategy built for 2016. Think about where authentic conversations about your category are happening, and show up there.

The human at the centre

Here’s the most important thing to hold onto amid all of this: people are still people.

They still put their earbuds through the wash and need a new pair. They still feel uncertain about spending $300 on a product they haven’t held in their hands. And most of all, they still want to feel confident that they’re making a good decision, that they aren’t being ripped off, and that someone out there has had the same experience and found it worthwhile.

The role of marketing has always been to help people on that journey – to reduce uncertainty, build trust, and make the path to a good decision easier.

AI hasn’t changed that mandate. But it has changed the infrastructure through which that help is delivered.

The brands that will thrive in this new landscape are not the ones who figure out how to game the AI. They’re the ones who become genuinely, comprehensively, and consistently helpful – so helpful that every AI model that encounters them comes away with the same clear picture: this is a brand worth recommending.

That’s not a new idea. It’s just more important than it’s ever been.

– Robert Hoek

• Tags:
• AI
• Digital marketing
• SEO