The marketing industry has operated on a top-down architecture for over 30 years. The blueprint was simple: start with awareness, cast the widest possible net to capture eyeballs, and systematically funnel those prospects down toward a conversion. This logic governed the broadcast era of television and radio, and it remained the dominant framework during the first two decades of the search engine era.
In those environments, the strategy was linear. If you spent enough on top-of-funnel (TOFU) awareness, a predictable percentage of people would eventually trickle down to the bottom of the funnel (BOFU) to make a purchase. But as we transition into an era defined by artificial intelligence, assistive engines, and autonomous agents, this 128-year-old model is no longer just outdated—it is fundamentally broken.
In an AI-driven digital ecosystem, the acquisition strategy must be flipped. Search engines and AI agents do not build their recommendations from the top down; they build them from the bottom up. They must understand who you are before they can judge your credibility, and they must trust your credibility before they will ever recommend you to a user. If you continue to build from the top down, you are effectively pouring marketing budget into awareness for a brand that AI agents have no foundation to recognize or trust.
The acquisition funnel runs simultaneously in opposite directions
To understand why the funnel has flipped, we must distinguish between the user’s experience and the machine’s process. For the human user, the acquisition journey remains relatively traditional. They hear about a solution (awareness), they evaluate their options (consideration), and they make a purchase (decision). This journey still flows from wide to narrow.
This model was formalized by Elias St. Elmo Lewis in 1898. For more than a century, every marketing department in the world has followed his lead: reach first, relationship second, commitment third. In the early days of the web, this meant building a website and then using SEO or PPC to drive traffic to it. As marketing expert Philippe Lanceleur noted in 2002, building a website without a traffic strategy is like opening a shop in the middle of a wide-open field. No one finds it by accident; you have to go where the people are and lead them back to your shop.
However, the shift toward “entities” changed the prerequisites for success. When Google introduced the Knowledge Graph in 2012, it began forming its own opinions about brands, independent of specific search queries. The machine started drawing its own map of the digital world. Instead of you having to lead people across the field to your shop, the machine began building the roads itself.
In the age of AI, these roads are built from the shop outward. This means that brand understanding and reputation are no longer the “result” of a good funnel; they are the “requirement” for the funnel to exist at all. AI agents act as intermediaries. When an agent acting on behalf of a user evaluates a brand, it does so with absolute scrutiny. If the machine does not understand exactly what you offer and whom you serve, it cannot act in your favor. If it understands you but finds a competitor more credible, it will bypass you entirely. This is the ultimate zero-sum moment: a recommendation you never knew was happening, made to a prospect you never knew was looking.
The mechanics of the bottom-up build
The traditional build funnel has been reversed. While the user still experiences the funnel from top to bottom, the machine builds its recommendation engine from the bottom up. The process follows a strict hierarchy of needs:
- Understanding: Does the machine know exactly who you are and what you do? This is the foundation (BOFU).
- Credibility: Does the machine trust that you are a reliable authority? This is the middle (MOFU).
- Advocacy: Will the machine proactively recommend you to a user who hasn’t asked for you by name? This is the top (TOFU).
You can still buy awareness through paid media or direct outreach—channels you control. However, within the organic ecosystems of AI engines like ChatGPT, Claude, and Google’s Search Generative Experience (SGE), you must build from the bottom up. These algorithms operate on brand signals and entity nodes, not just keyword volume. Reach is now a byproduct of brand recognition and trust.
How the funnel becomes a guided sequence in AI
In the traditional SEO era, a search engine results page (SERP) was a collection of links that a user navigated themselves. The user was the pilot, moving from awareness to decision by clicking, browsing, and comparing. The SEO’s job was simply to secure a high-ranking slot in that composition.
Today, Large Language Models (LLMs) have fundamentally changed that dynamic. When a user asks a question, the AI reasons about the intent. It decides whether to answer directly, search for fresh data, or verify facts via a knowledge graph. It often runs “fan-out” or “cascading” queries—multiple background searches that look at a topic from different angles to provide a comprehensive answer.
This architecture allows the AI to do something revolutionary: it anticipates the user’s next move. By shaping the current answer, the AI defines the user’s acquisition journey. The user feels in control, but the AI is actually narrowing the path toward a specific conclusion. Your brand’s job is to train the machine’s expectations so that your content is the “logical next step” in that predicted sequence.
By publishing logical bridges—content that says, “if you are considering X, the next thing you must evaluate is Y”—and corroborating that information across multiple trusted platforms, you create “synapses” in the machine’s understanding. When the AI predicts the user’s next step, it will reach for your brand because you have made that connection the most logical one in its database.
Understandability, Credibility, and Deliverability: The UCD Model
To succeed in a bottom-up strategy, brands must focus on the three dimensions of visibility at the point of “Display.” This is the moment the machine presents your brand to the user. Success here depends on three layers:
Understandability (U): The Decision Layer
Understandability is the “Bottom of the Funnel” for the machine. If the AI doesn’t have a clear entity record for your brand, it will hedge its language. You may have seen AI responses that say a brand “claims to be” a leader or “appears to offer” a service. This is a failure of Understandability. When a user is ready to buy, they need a confirmation, not a hedge. If the machine is confused, the user becomes hesitant. We call this the Doubt Tax.
Credibility (C): The Consideration Layer
Credibility is the “Middle of the Funnel.” This is where the AI compares you to your competitors. The machine looks for N-E-E-A-T-T signals (Niche, Experience, Expertise, Authoritativeness, Trustworthiness, and Transparency). If the AI has higher confidence in your competitor’s credibility than yours, you will be excluded from shortlists and “best of” recommendations. This is the Ghost Tax—being invisible during the critical comparison phase.
Deliverability (D): The Awareness Layer
Deliverability is the “Top of the Funnel.” This is advocacy. It happens when the AI surfaces your brand to someone who wasn’t even looking for you. The machine only does this when it is fully convinced of your Understandability and Credibility. It treats your brand as the “reference option” for a category. Failure here results in the Invisibility Tax, where you are never mentioned to prospects researching the broader market.
The business case for UCD: Retiring the three taxes
Think of the major AI platforms—Google, ChatGPT, Perplexity, Claude, Copilot, Siri, and Alexa—as seven virtual employees working for you 24/7. They are either selling your brand or they are selling for your competitors. Your goal in the AI era is to train these “employees” to represent you accurately.
Optimizing for these machines is not about gaming an algorithm; it is about reinforcing the world’s opinion of your brand. A Brand SERP is essentially Google’s summary of what the world thinks of you. By improving Understandability, Credibility, and Deliverability, you are retiring the taxes that drain your conversion rates:
- Retire the Doubt Tax by cleaning up your structured data, maintaining a consistent “Entity Home,” and ensuring your schema is accurate.
- Retire the Ghost Tax by securing third-party mentions, earned media, and co-citations from sources the machine already trusts.
- Retire the Invisibility Tax by creating content that serves as “proof” of your expertise rather than just a marketing claim.
Acquisition is one act in a 15-stage play
The acquisition funnel is often treated as the end-all-be-all of marketing because that is where the transaction happens. However, in the AI world, the transaction is just one gate in a much longer pipeline. There are nine gates before the user ever sees your brand at “Display,” and there are five gates after they have been “Won.”
The stages that follow a purchase—Onboarding, Performance, Integration, Devotion, and Codification—are where the real power lies. Every satisfied customer creates signals that feed back into the machine’s foundation. This creates a flywheel effect. When a client has a positive outcome and that outcome is documented or mentioned online, it strengthens the AI’s confidence in your brand’s Credibility (C) and Understandability (U).
This means that your post-purchase customer success strategy is actually part of your pre-purchase AI acquisition strategy. The machine is a constant participant in the entire lifecycle of your business. If you fail at the later stages, the machine will notice the lack of positive signals or the presence of negative ones, and it will slowly stop recommending you at the “Display” gate.
Strategy: Where to begin the flip
If you are ready to move toward a bottom-up acquisition strategy, your first diagnostic instruments are your Brand SERP and your AI Résumé (the conversational summary an AI provides about your business). Read them through the lens of the machine’s confidence.
If the AI gets basic facts wrong or hedges its descriptions, you have an Understandability problem. You must focus on your “Entity Home”—the single authoritative source of truth for your brand (usually your About page or a dedicated entity page) and ensure your structured data is flawless.
If the AI’s summary is accurate but uninspiring, or if you are missing from competitive comparisons, you have a Credibility problem. You need to focus on off-site signals: reviews, industry awards, and mentions in high-authority publications that the AI uses to ground its answers.
If you are simply not showing up for broad, category-level queries, you have a Deliverability problem. You need to build “content bridges” that show the machine why your brand is the natural solution to the problems users are asking about.
The 128-year-old top-down funnel is a relic of an era where humans did all the filtering. In the AI era, the machines do the filtering for us. To get through that filter, you cannot start with a loud shout at the top of the funnel. You must start with a solid foundation at the bottom. Build for understanding first, earn credibility second, and only then will the machines grant you the reach and awareness that fuels modern growth.