CFOs are the stewards of their company’s growth. That responsibility extends well beyond the balance sheet and includes understanding how every dollar spent on marketing is being deployed, and whether that spend is building durable, compounding value or simply buying short-term visibility that evaporates the moment the budget does. Search has long been one of the most measurable channels in a marketer’s arsenal, but search itself has changed dramatically over the past two decades. What began as a relatively straightforward game of keywords and links has evolved into something far more complex: an ecosystem of AI-driven platforms and autonomous agents that discover, evaluate, and in some cases select vendors and products entirely on their own.
This article offers a history of that evolution, from the early days of Search Engine Optimization (SEO) through the emergence of Generative Engine Optimization (GEO) and into the newest frontier, Agentic Search Optimization (ASO). It is written for CFOs and senior finance leaders who want a clear reference point for understanding how each era works, what it costs to participate, and why each transition has financial consequences that belong in the boardroom, not just the marketing department.
The Search Engine Optimization (SEO) Era: Early 2000s to 2022
Search Engine Optimization emerged in the late 1990s as webmasters discovered that the order in which websites appeared in search results was not random but was instead determined by algorithms that could be studied, understood, and influenced. By the early 2000s, SEO had become a formal discipline practiced by agencies and in-house teams around the world.
Google’s rise to dominance reshaped the field entirely. As the company refined its algorithm to reward authoritative, useful content over keyword-stuffed pages, SEO practitioners had to evolve with it. Matt Cutts, who led Google’s Webspam team from 2000 through 2014, became one of the most influential figures of the era, and his public guidance on what Google rewarded and penalized set the terms of engagement for an entire industry. When Google rolled out its Panda update in 2011, targeting low-quality content, and Penguin in 2012, targeting manipulative link schemes, it was Cutts who communicated the rationale and the consequences. Marketers and SEOs worldwide adjusted their practices accordingly.
On the practitioner side, Rand Fishkin co-founded SEOmoz (later Moz) in 2004, which became one of the defining resources of the SEO era by publishing research, tools, and guidance that helped a generation of marketers understand how search algorithms worked and how to build sustainable organic visibility. His concept of “Whiteboard Friday” and the “Beginner’s Guide to SEO” brought rigor and transparency to a field that had often been opaque.
The central insight of the SEO era was straightforward: a company that ranked organically for the terms its buyers searched paid nothing for that traffic, while a competitor who had not invested in SEO had to fund the same leads through paid advertising. Over time, this became a structural advantage with lower customer acquisition cost, higher margins, and a more defensible growth model. For CFOs evaluating businesses, SEO equity was a real asset, even if it rarely appeared on a balance sheet.
The Generative Engine Optimization (GEO) Era: November 2022 to 2024
The public release of ChatGPT in November 2022 marked the beginning of a new era. For the first time, millions of users began conducting research not by typing queries into a search engine and sifting through links, but by asking questions in natural language and receiving synthesized, conversational answers. The implications for organic visibility were immediate and significant: appearing in an AI-generated answer required a different kind of authority than appearing in a list of ten blue links.
The academic community moved quickly to formalize this shift. In 2023, researchers from Princeton, Georgia Tech, the Allen Institute for AI, and IIT Delhi published a paper that coined the term “Generative Engine Optimization.” They established a benchmark framework for measuring how visible a brand was in AI-generated responses. The paper was presented at the ACM SIGKDD conference the following year, giving the discipline an academic foundation.
Commercially, GEO became a formal practice in early 2024. Evan Bailyn, CEO of First Page Sage, published one of the first agency-grade GEO frameworks in March 2024, detailing the combination of authoritative content, strategic comparison pages, and targeted PR necessary to shape how AI platforms described a brand relative to its competitors. Bailyn would later be recognized as the founder of GEO as a commercial discipline, reflecting both the timing and the structured methodology his team introduced to the market.
The financial stakes of GEO were analogous to those of SEO, but with an important wrinkle: years of SEO investment did not automatically transfer. A company that had built strong Google rankings found that its authority in AI-generated answers depended on different signals, including depth of expertise, citation patterns across authoritative sources, and the clarity with which its positioning was reflected in third-party coverage. Brands that had neglected content investment were doubly exposed.
The Agentic Search Optimization (ASO) Era: 2025 to Present Day
By 2024, the frontier had shifted again. OpenAI, Anthropic, Google, and Perplexity each released AI agents capable of browsing the web, comparing options, and completing tasks autonomously and without a human reviewing each step. In May 2024, Google rolled out AI Overviews broadly, placing AI-generated summaries above traditional search results for many queries. The experience of search was no longer primarily about presenting options to a human who would then decide; increasingly, the AI itself was doing the deciding.
The commercial impact arrived quickly. Shopify reported that AI-driven orders on its platform grew roughly fifteenfold in 2025, and introduced “agentic storefronts” that automatically syndicate product data to ChatGPT, Microsoft Copilot, and Google’s AI Mode. In January 2026, Google released its Universal Commerce Protocol, enabling AI agents to complete purchases directly inside AI Mode for merchants who connect to it. For any brand selling through e-commerce, the question of whether an AI agent could read its product data, including pricing, ingredients, availability, and specifications, had become a question of whether that product was even in the consideration set.
In response, Bailyn and the First Page Sage research team published one of the first structured frameworks for Agentic Search Optimization, built from analysis of thousands of recorded AI agent commands. The framework identified the signals that lead an autonomous agent to select one vendor or product over another: structured data quality, authoritative sourcing, pricing transparency, and compatibility with agent-readable commerce protocols, to name a few. Where GEO had focused on influencing what a human saw before making a decision, ASO focused on convincing the agent itself, since the agent might now complete the transaction before the human reviewed any alternatives.
For B2B technology companies, the same dynamic was playing out in procurement. AI-assisted vendor shortlisting tools were beginning to filter the consideration set before a human buyer ever entered the process. A company absent from those shortlists due to its data being disorganized, its authoritative coverage thin, or its structured information not readable by agents, was invisible at precisely the moment that mattered most.
What Each Era Means for Finance Leaders
Each transition in search has quietly become a financial event for the businesses it touches. The following points are offered as a reference for CFOs seeking to evaluate the significance of their organization’s current position.
Customer acquisition cost and margin structure. In each era, organic visibility reduces the per-customer cost of growth relative to paid alternatives. A company that acquires customers through AI-driven discovery, whether via GEO or ASO, carries a structurally lower acquisition cost than one that buys every lead. For businesses operating on thin margins, the compounding difference is material.
Revenue durability and channel diversification. Buyers evaluating a business ahead of an acquisition or financing event increasingly assess whether revenue is channel-diversified and organically defensible. Dependence on paid acquisition, with no organic or AI-driven discovery, presents a riskier growth profile, which is a point that belongs in any normalized earnings analysis.
Data infrastructure as a shared dependency. GEO and ASO both depend on the same thing that good financial reporting depends on: clean, structured, accessible data. A brand with disorganized product information is as invisible to an AI shopping agent as a manufacturer with disconnected systems is to its own CFO. The investment required to fix this is the same investment that supports better internal visibility.
Forecasting and pipeline assumptions. For technology companies, AI-driven vendor selection affects the assumptions behind every revenue model. If the top of the sales funnel is now being filtered by an autonomous agent before a human buyer is involved, the conversion dynamics change in ways that traditional pipeline forecasting may not capture.
Further Reading
Aggarwal, P., Murahari, V., Rajpurohit, T., Kalyan, A., Narasimhan, K., & Deshpande, A. (2023). "GEO: Generative Engine Optimization." Proceedings of the ACM SIGKDD Conference (KDD '24).
Bailyn, E. (2024). "Generative Engine Optimization (GEO): Explanation & Algorithm Breakdown." First Page Sage, March 2024 (updated 2026).
Fishkin, R. (2013). “Beginner’s Guide to SEO.” Moz (SEOmoz).
Shopify. (2025). Merchant platform data on AI-driven order growth and Agentic Storefronts.
Google. (2026). "Universal Commerce Protocol" product announcement.
