Large Language Models are fundamentally reshaping digital commerce, altering how users discover relevant products and services, and in doing so, rendering traditional advertising methods increasingly obsolete. This is a profound disruption, not an incremental evolution, demanding a courageous reimagining of how businesses connect with consumers and how value is exchanged. This post will explore this transformation, assess the shortcomings of early LLM advertising strategies, and champion a more integrated, user-centric paradigm: intent-based commerce.
The Obsolete Keyword Model & The Evolving 'Answer'
Keyword vs LLM
For decades, keyword ads were imprecise proxies for user intent, leading to inefficiencies. LLMs, understanding nuanced language, now expose this flaw. The old model of guessing intent is breaking down.
Historically, successful search advertising (like early Google) aligned user intent (seeking links), the "answer" (relevant links), and monetization (ads as relevant links). This worked because the ad was the answer. Over-optimization for revenue later broke this trust, as ads often appeared "too soon" or felt misaligned, degrading the user experience. With LLMs, the user's expected "answer" to a high-intent query is no longer a list of links but a direct solution—a specific product or service. Many current LLM ad strategies falter here, creating an "answer-ad mismatch" that erodes trust if the AI's direct answer is perceived as biased or if ads are irrelevant to the clarified intent. For LLM advertising to work, the "answer" and the "ad" must seamlessly converge at the right point in the user journey, ideally after pure intent clarification.
Early LLM Advertising: "Lipstick on a Familiar Pig"?
With the old keyword model clearly insufficient and user expectations shifting towards direct, comprehensive answers, the industry is scrambling to monetize LLM interactions. However, many initial strategies are essentially attempts to transpose old interruption-based advertising mechanics onto new conversational interfaces. This approach often fails to address the new nature of "answers" or the heightened need for user trust, the underlying ad model remains largely unchanged, merely presented in an LLM-powered guise.
Common tactics include:
Ads Directly Influencing or Integrated into LLM Output: Advertisers paying to have their messages subtly (or overtly) woven into AI-generated text.
Display Ads Around Conversational Interfaces: Placing traditional banner or video ads in the UI surrounding the AI chat, sometimes explored in Perplexity.
Advertising in AI-Powered Summaries: Embedding sponsored content within or adjacent to AI-synthesized answers, as seen with Google's AI Overviews.
Sponsored Questions or Prompts: Suggesting commercially sponsored follow-up questions after an initial AI answer, a method used by platforms like Perplexity AI.
While these represent early forays, they share fundamental challenges. They often degrade the user experience and erode trust if ads feel forced, interrupt the conversational flow, or make the AI's responses seem biased. Transparency issues arise when it's unclear why certain ads are shown, and privacy concerns mount if user conversations are deeply analyzed for ad targeting without explicit consent. Crucially, these methods frequently create an "answer-ad mismatch," where the advertisement feels disconnected from the user's specific, LLM-clarified intent, rendering the ad an unwelcome interruption rather than a helpful part of the solution. These early models largely underscore the difficulty of simply transposing old advertising paradigms onto a technology that demands a more intelligent and trustworthy approach.
Answer Ad Mismatch
Navigating the New Terrain: Learnings from the Vanguard
Pioneering platforms offer key insights:
Perplexity AI: Prioritizes answer integrity by separating ads, but faces monetization challenges and risks ad irrelevance when decoupled from deep intent understanding. Their experiments with "sponsored follow-up questions," while preserving answer neutrality, reportedly encountered difficulties in delivering clear ROI for advertisers and resonating with their user base.
Google's AI Overviews: Google integrates ads within or alongside its AI-synthesized summaries. The rationale is that users receiving an AI overview are more qualified, potentially leading to higher CTRs. However, this model must maintain user trust against perceived commercial influence and navigate the "zero-click" impact on publisher value exchange.
OpenAI's ChatGPT: OpenAI has been cautious with traditional ads, emphasizing subscriptions. While leadership downplays advertising as a primary revenue path, stating product selections aren't commercially driven, any future commercial integrations must uphold the immense user trust in ChatGPT as an objective assistant, demanding exceptional transparency.
A Radically Different Approach: User-Centric Intent-Based Commerce
The Imperative for a New Model
The limitations of current models demand a user-focused approach that removes unnecessary intermediaries and ensures the user is the primary beneficiary. This isn't to argue that advertising itself is inherently negative. When executed correctly, ads serve a valuable purpose: they help users discover relevant products or services and navigate a complex marketplace to make informed choices.
Advertising becomes problematic, however, when it turns intrusive, when incentives become misaligned—prioritizing platform revenue or aggressive selling over genuine user needs—or when the "ad" itself overshadows the value of the underlying product or service it represents. This is the imbalance that many current systems exhibit.
LLMs, with their sophisticated ability to understand user needs, introduce a new dimension to this challenge. Once an LLM has helped a user clarify their intent, it might identify a multitude of perfectly suitable options. A purely "neutral" engine could theoretically present all these options, but this would likely lead to an overwhelming and unhelpful user experience.
This reality necessitates a critical decision: how does the platform select which options to highlight? If this selection process is opaque or driven by hidden commercial incentives favoring the platform or certain sellers, user trust is inevitably compromised. The central tension, then, is to ensure that the presented choices genuinely align with the user's best interest.
The Solution: Intent-Based Bidding
This is where Intent-Based Bidding offers a structural solution by fundamentally re-architecting how choices are made and presented. Instead of advertisers bidding on keyword auctions or LLM platforms showing potentially misaligned ads, Intent-Based Bidding allows advertisers and sellers to bid directly on the verified, anonymized intent of the user. (This is the approach we are building at Intents Protocol).
Core Mechanics of Intent-Based Bidding
The selection of offers is then driven not by a central, opaque authority, but by a combination of competitive game theory among sellers and, crucially, user-defined preferences. This process is designed to fit naturally into the user journey, particularly after the initial discovery phase, rather than disrupting it. This approach seeks to structurally change not just product discovery but how commerce itself is transacted, especially as AI agents—which will prioritize value and efficiency over browsing advertisements—become increasingly prevalent.
Key principles of Intent-Based Bidding that address the selection and incentive challenge:
Verified Granular Intent (Context-Dependent): Conversational AI clarifies detailed user needs upfront. The specificity of this intent can vary. In contexts like dedicated shopping assistants, intent can become highly granular (e.g., "noise-canceling headphones, under $300, foldable, >10hr battery, for air travel"). In other environments, the expressed or inferred intent might be broader. The protocol is designed to accommodate this spectrum, focusing on the fulfillment phase once intent has been reasonably established.
Sellers Bid on Exact Intent – Driven by Game Theory: Advertisers bid to fulfill this specific, verified intent. The decision of which offers are shown is not made by an opaque platform algorithm but is influenced by the competitive dynamics of the bidding process itself. Knowing they are bidding directly on a clearly defined user need, sellers are incentivized by game theory to present their most relevant products and optimal pricing.
User Preference in Offer Ranking: The system can be designed to allow users to influence how bids are presented or prioritized. For example, a user might choose to see offers that provide the highest direct incentive to them, or those that most closely match every nuance of their stated intent. This puts the user in control.
The "Answer" IS the "Ad" – Alignment Through Transparent Choice: Because the selection of what to show is driven by the transparent mechanics of bidding on precise intent and potentially user-defined preferences, the resulting offer becomes the direct, friction-free answer the user sought at the point of commercial engagement. There's no hidden agenda.
User as Primary Beneficiary – Reinforcing Alignment: Users are rewarded (e.g., a share of the bidding fee) for their valuable, anonymized intent. This fundamentally changes the power dynamic and ensures that the system's incentives are strongly aligned with providing user value.
Privacy and User Control: Systems prioritize privacy, with users controlling how their anonymized intent is shared.
Open, Neutral Protocol: An underlying "Intents Protocol," not owned by a single company, can act as a public good. This fosters competition, empowers developers and users with choice, and crucially, prevents platform-centric bias.
Future-Proof for AI Agents: Essential for an "agentic economy" where AIs make purchases based on precise, verified intents, not by browsing ads.
Intent Bidding
This intent-based model reimagines commerce by ensuring a healthier balance between intent discovery and commercial fulfillment. It uses the clarity of LLM-defined intent to create a transparent and competitive marketplace where the "advertisements" are, in fact, the most relevant solutions, with incentives and selection mechanisms structured to prioritize user benefit and choice.
The Path Forward: Architecting an Intelligent and Equitable Future
The move from keyword-based ads to LLM-driven experiences demands a smarter, more transparent, and user-aligned advertising model. Early efforts to bolt old ad frameworks onto new tech often fall flat. What’s needed is a ground-up rethink—where verified intent drives value, and the ad is the answer. The future belongs to conversational, intent-aware systems that prioritize user trust and deliver relevance without compromise. Ads that feel like help—not noise—will define the next wave of digital commerce. This is our opportunity to build a more intelligent, equitable, and valuable commercial ecosystem for everyone.
We invite you to be at the forefront of this evolution. Explore the transformative potential of intent-based commerce, create your first intent, and experience this paradigm shift firsthand at www.inomy.shop.