Google Stock Fell 12%. 99% Of People Don't Understand What This means!

Eric says Google stock fell roughly 12% after earnings, an unusually large reaction driven by fears that AI will destroy Google Search. He argues the concern is overblown: Google ads still grew 12% despite a weak economy, cloud revenue grew about 103%, and people continue to use Google at enormous scale. Neil says the stock is cheap, though its headline P/E is affected by SpaceX; fully burdened, including stock-based compensation, he estimates roughly 26–27 times earnings. For a company growing around 24% annually on approximately $500 billion in revenue, Neil considers that affordable and says he would choose Google over the S&P 500.

Eric cites Google Search and other revenue of $63.35 billion, up 12%, Microsoft search and advertising revenue of $6.55 billion, up 10%, and Amazon AI-managed campaigns producing 67% higher long-term return on ad spend and 29% lower new-to-brand acquisition costs. He says advertising remains healthy because people and increasingly AI agents continue transacting online. He then describes Granola, valued at $1.5 billion, increasing monthly new users from AI search by 4,600% in seven months. Granola began with buyer decisions rather than keywords, converted those decisions into Google queries and AI prompts, diagnosed missing visibility, used Reddit and third-party sources, and ranked first for “best AI meeting note taker for founders.”

Eric contrasts two events: a YPO Global Marketing Summit with up to 180 attendees that sold out only at the last minute, and his newer AI summit, which sold out about two weeks before the event with less sales effort. He attributes the difference to AI’s popularity and the existing audience of roughly 1,000 people around the event. The same timing principle applies to his YouTube videos: responding immediately to a trend works, while waiting two or three days hurts performance. Neil cautions that being too early can also fail; when Eric discussed AI agents at HubSpot about a year earlier, many people neither understood nor cared, whereas current products such as Muse, Dots, and Grokbot have made the subject more compelling.

Neil describes asking Grokbot to find reputable Las Vegas water-softener companies and schedule appointments under $5,000. A manually found provider quoted about $7,600, while one AI-selected company charged roughly $4,800. In broader tests of service businesses, he says agents consider both total review count and review recency: a company with 100 reviews and 50 new reviews monthly may outrank one with 1,000 reviews but only one new review monthly. Agents also favor consistent messaging across platforms; inconsistent descriptions can prevent a business from being recommended even when it has strong reviews and digital PR.

Neil says third-party mentions matter, but the strongest content he has observed is detailed comparison material that compares a business with the major competitors customers actually recognize. An article pairing a roofing, HVAC, or cleaning company with the region’s three established providers performs better than one naming a major company alongside several obscure businesses. He believes AI agents are more likely to recommend a company repeatedly compared against the main local alternatives. Eric connects this to how AI systems weight recent and repeated information, saying marketers should keep testing how agents interpret review velocity, messaging consistency, and competitive context.

Eric demos Single Grain’s AI visibility report, which shows how often a brand appears in AI answers and then unlocks a fuller report after collecting an email. He says an AI system now performs the analysis for cents rather than the former $1–$3 cost; his account had processed 786 million tokens and 468,000 requests for $31. He also describes a talent bot sourcing qualified, AI-native candidates for $0.00459 each. Neil says detailed reports with many charts generally reduce B2B close rates. Eric therefore tells his team to highlight the customer’s clearest gaps and the top two or three fixes instead of presenting a highly technical report, a change that was already booking sales calls, including one from a company with about $927 million in revenue.