The Day Marketing Realised Its Audience Had No Pulse

When machines started buying on our behalf, the world’s best storytellers found themselves pitching to code. Turns out, the algorithm doesn’t care about your brand voice, your mission statement, or your purpose. It just wants clean data and maybe a little confession of human weakness.

The average American supermarket carries over 30,000 distinct items, or SKUs. For a century, the primary goal of a consumer packaged goods (CPG) company like Procter & Gamble or Unilever has been to win the battle for attention on that crowded shelf. They paid for eye-level placement, designed vibrant packaging, and spent billions on advertising to build a flicker of brand recognition that would translate into a purchase decision in the fraction of a second a human shopper scans an aisle. That entire economic model is predicated on a simple fact: the consumer is human.

That fact is no longer a given.

What happens when your weekly shop is automated or one of your customers says: “Hey Google, add paper towels to my shopping list.” Or, more disruptively: “Order me more paper towels.” There is no shelf. There is no packaging. There is no moment of cognitive battle between Bounty, with its quicker-picker-upper jingle stored in your memory, and the generic store brand. There is only an intent, an algorithm, and a transaction. The consumer, in the traditional sense, has been abstracted away. In their place is the Algorithmic Consumer, and marketing to it requires a fundamentally different strategy.

This is a platform shift that threatens to upend the core tenets of brand, distribution, and advertising. The new gatekeepers are not retailers, but the AI assistants that mediate our interaction with the market. For businesses, the urgent strategic question is shifting from “How do we reach the consumer?” to “How do we become the machine’s default?”

The Great Compression: From Funnel to API Call

The classic marketing funnel: Awareness, Interest, Desire, Action (AIDA), is a model designed for the psychology of a human buyer. It’s a slow, expensive, and inefficient process.

* Awareness is built with Super Bowl ads and billboards—blunt instruments for mass attention.

* Interest is cultivated through content marketing and positive reviews.

* Desire is manufactured through aspirational branding and targeted promotions.

* Action is the final click or tap in a shopping cart.

The AI assistant acts as a powerful compression algorithm for this entire funnel. The user simply states their intent: “I need paper towels.” The stages of Awareness, Interest, and Desire are instantly outsourced to the machine. The AI evaluates options based on a set of parameters and executes the Action. The funnel is compressed into a single moment.

This has devastating implications for brands built on awareness. The billions of dollars spent by P&G on making “Bounty” synonymous with “paper towels” have created a cognitive shortcut for humans. An AI, however, has no nostalgia for commercials featuring clumsy husbands. It has an objective function to optimise. The machine’s decision might be based on:

* Price: What is the cheapest option per sheet?

* Delivery Speed: What is available for delivery in the next hour?

* User History: What did this user buy last time?

* Ratings & Reviews: What product has the highest aggregate rating for absorbency?

* User Preferences: The user may have once specified “eco-friendly products only,” a constraint the AI will remember with perfect fidelity.

The strategic imperative shifts from building a brand in the consumer’s mind to feeding the algorithm with the best possible data. Your API is your new packaging. The quality of your structured data: price, inventory, specifications, sourcing information, carbon footprint—is more important than the cleverness of your copy. This is the dawn of Business-to-Machine (B2M) marketing.

Generative Engine Optimisation (GEO)

For the past two decades, Search Engine Optimisation (SEO) has been the critical discipline for digital relevance. The goal was to understand and appeal to Google’s ranking algorithm to win placement on the digital shelf of the search results page. The coming paradigm is Generative Engine Optimisation (GEO), but it is different in several crucial ways.

SEO is still fundamentally a human-facing endeavour. The goal is to rank highly so that a human will see your link and click it. The content, ultimately, must persuade a person.

GEO is a machine-facing endeavour. The goal is to be the single best answer that the AI assistant returns to the user. Often, there is no “page two.” There is only the chosen result and the transaction. The audience is the algorithm itself.

The factors for winning at GEO are not keywords and backlinks, but logic-driven and data-centric attributes:

1. Availability & Logistics: An AI assistant integrated into a commerce platform like Amazon or Google Shopping will have real-time inventory and delivery data. A product that can be delivered in two hours will algorithmically beat one that takes two days, even if the latter has a stronger “brand.” The winner is not the best brand, but the most available and convenient option.

2. Structured Data & Interoperability: Can your product’s attributes be easily ingested and understood by a machine? A company that provides a robust API detailing its product’s every feature—from dimensions and materials to warranty information and sustainability certifications—provides the AI with the granular data it needs to make a comparative choice. A company with a beautiful PDF brochure is invisible.

3. Cost & Economic Efficiency: Machines are ruthlessly rational economic actors. If a user’s prompt is “order more paper towels,” and no brand is specified, the primary variable for the AI will likely be optimising for cost within a certain quality band. This is a brutal force of commoditisation. Brand premiums built on psychological messaging are difficult to justify to a machine unless they are explicitly encoded as a user preference (“I only buy premium brands”).

The absurdity of this new reality can be humorous. One can imagine marketing teams of the future not brainstorming slogans, but debating the optimal JSON schema to describe a toaster’s browning consistency. The Chief Marketing Officer may spend more time with the Chief Technology Officer than with the ad agency.

The Aggregation of Preference

This shift fits perfectly within the framework of Aggregation Theory. The AI assistant platforms: Amazon’s Alexa, Google’s Assistant, Apple’s Siri and the LLMs building this out directly in their apps and websites

1. They own the user relationship. They are integrated directly into our homes and phones, capturing our intent at its source.

2. They have zero marginal costs for serving a user. Answering one query or one billion is effectively the same.

3. They commoditise and modularise supply. The paper towel manufacturers, the airlines, the pizza delivery companies; they all become interchangeable suppliers competing to fulfill the intent captured by the Aggregator.

The ultimate moat in this world is the default.

When a user says “Claude, order a taxi,” will the default be Uber or Lyft? Anthropic will have the power to make that decision. It could be based on which service offers the best API integration, which one pays Amazon the highest fee for the referral, or it could be an arbitrary choice. The supplier is in a weak position; they have been disconnected from their customer.

This creates a new, high-stakes battleground. The first time a user links their Spotify account to their Google Home, they may never switch to Apple Music. The first time a user says, “From now on, always order Tide,” that preference is locked in with a far stronger bond than brand loyalty, which is subject to erosion. It is now a line of code in their user profile. Winning that first transaction, that first declaration of preference, is everything.

We will likely see three strategic responses from suppliers:

* The Platform Play: Companies will pay exorbitant fees to be the default choice. This is the new “slotting fee” that CPG companies pay for shelf space, but on a winner-take-all, global scale.

* The Direct Play: Brands will try to build their own “assistants” or “skills” to bypass the Aggregator. For example, “Ask Domino’s to place my usual order.” This works for high-frequency, single-brand categories but is a poor strategy for most products. Nobody is going to enable a special “Bounty skill” for their smart speaker.

* The Niche/Human Play: The escape hatch from algorithmic commoditisation is to sell something a machine cannot easily quantify. Luxury goods, craft products, high-touch services, and experiences built on community and storytelling. These are categories where human desire is not about utility maximisation but about identity and emotion. The machine can book a flight, but it can’t replicate the feeling of being part of an exclusive travel club.

The Strategic Humanist’s Dilemma

This brings us to the human cost of algorithmic efficiency. A world where our consumption is mediated by machines is an intensely practical one. We might get lower prices, faster delivery, and more rational choices aligned with our stated goals (e.g., sustainability). This is the utopian promise: the consumer is freed from the cognitive load of choice and the manipulations of advertising.

However, it is also a world of profound sterility. Serendipity, discovering a new brand on a shelf, trying a product on a whim, is designed out of the system. The market becomes less of a vibrant, chaotic conversation and more of an optimised, silent database. Challenger brands that rely on a clever ad or a beautiful package to break through have no entry point. Power consolidates further into the hands of the platform owners who control the defaults.

The strategic implications are stark and urgent.

1. For CPG and Commodity Brands: The future is B2M. Investment must shift from mass-media advertising to data infrastructure, supply chain optimisation, and platform partnerships. Your head of logistics is now a key marketing figure.

2. For Digital Native Brands: Winning the first choice is paramount. The focus must be on acquisition and onboarding, with the goal of becoming the user’s explicit, locked-in preference.

3. For All Brands: Differentiate or die. The middle ground of “decent product with good branding” will be vaporised by algorithmically-selected, cost-effective generics on one side and high-emotion, human-centric brands on the other. You must either be the most efficient choice for the machine or the most meaningful choice for the human.

The age of marketing to the human subconscious is closing. The slogans, jingles, and emotional appeals that defined the 20th-century consumer economy will not work on a silicon-based consumer. The companies that will thrive in the 21st century are those that understand this shift, reorient their operations, and learn to speak the cold, ruthlessly efficient language of machines.