How I Stopped Obsessing Over Google and Started Talking to Robots
I used to live and die by my Google Analytics dashboard. Clicks, impressions, keyword rankings—that was the whole game. For a decade, the growth formula was simple: figure out what people are Googling, and be the first result. We all did it.
Then I saw a Reddit post that made my stomach drop.
A founder wrote: "I use the web search functionality of ChatGPT. It’s been months since I don't go to Google to 'platform that do this and this.'"
I realized something: we’re all about to be doing it wrong.
The Problem with "The Future of Search"
Most articles about AI and search are full of abstract jargon. "Paradigm shifts," "knowledge graphs," "retrieval-augmented generation." It’s no wonder most marketers are just shrugging and pouring more money into Google Ads.
The real issue isn't that Google is dead—it's that it just got demoted.
Think of it this way: AI chatbots are for learning. Google is for buying.
When you have a big, messy problem, you don't type three keywords into a search bar anymore. You have a conversation:
"Explain microservices to me like I'm five."
"What's a good alternative to Salesforce for a small team that can't afford it?"
"Draft a marketing plan for my new dog-walking app."
This is the new discovery phase. It’s where your potential customers are figuring things out, forming opinions, and deciding who to trust.
Google is what you use after you've already made your decision. It’s the Yellow Pages. You search for "Salesforce pricing" or "plumber near me" when you're ready to buy. By then, the game is already over. If you're only showing up at the very end, you're not a brand; you're just a line item on a pricing comparison page.
The game has moved upstream, and your old SEO tactics can’t swim.
What Actually Works: 4 Ways to Get Noticed by AI (That Don't Suck)
So if there are no ads and no rankings, how do you get an AI to talk about you? You stop trying to "optimize" and start trying to be undeniable.
1. Be the Wikipedia for Your Niche
Stop writing 800-word blog posts about "5 Tips for X." They're invisible. When an AI needs an answer, it looks for the most comprehensive, authoritative source it can find. Your job is to create that source.
Instead of writing "Best Financial Modeling Software," create a free, university-level course on financial modeling. The kind with videos, templates, and a full glossary.
When a user asks an AI, "How does a discounted cash flow work?" it will find your definitive guide. And when it needs to recommend a tool, who do you think it's going to mention? The company that is the expert.
The key: Your product is the logical next step after your content, not the hero of it.
2. Hang Out Where Humans Actually Talk
AIs don't just learn from your website. They learn from the whole internet. They read Reddit, Quora, industry forums, and review sites. They learn from the messy, unfiltered conversations real people are having about problems in your space.
This means your team needs to be in those places, being genuinely helpful.
I’ll say it: A single, detailed Reddit comment from your founder helping someone solve a real problem is more valuable than ten generic guest posts. Why? Because you can't fake it. It’s authentic, conversational proof that you know what you're talking about. AI models and humans can spot that authenticity from a mile away.
The key: Show up to help, not to pitch. People (and AIs) will figure out what you do.
3. Answer the Real Questions
People talk to AIs like they talk to a smart colleague. The questions are long, specific, and full of context.
Old Google search: "crm software" New AI query: "I need a CRM for my 5-person startup. We mostly need email integration and can't spend more than $50 a month. What are my best options?"
Your website needs content that directly answers that kind of question. Create in-depth comparison pages ("Us vs. Them"), alternative pages ("5 Great Alternatives to BigCompetitor"), and pages for specific problems. Use simple tables—AIs love digesting structured data.
The key: Go ask an AI to compare you to your biggest competitor. The answer it gives you is your new to-do list. If it's wrong, you need to create the page that corrects it.
4. Spoon-Feed the Robots
This is the nerdy part, but it's important. You can use things like schema markup and structured data on your website to leave a "cheat sheet" for AI models. It’s a way of telling them exactly what your company is, what your product does, and what questions you can answer, all in a language they understand perfectly.
It’s the digital equivalent of neatly organizing your notes before an open-book test. It makes it dead simple for the AI to reference you correctly.
The key: Ask your developer to do a "schema audit." It’s a weekend of work that will make you exponentially easier for AIs to understand for years to come.
The Long Game
This isn't a growth hack. You won't see results overnight. You can't just reallocate your ad budget and call it a day.
The ROI isn't measured in clicks; it's measured in being the default answer. It's about building a moat of authority so deep that when someone has the problem you solve, your name is the first one that comes to mind—whether that mind is human or artificial.
This takes time. Months, not days. But it’s a defensible strategy that can't be outbid.
Getting Started (Without the Frameworks)
Pick 3 competitors. Go to ChatGPT or Perplexity and ask, "Compare my product to [Competitor A, B, and C]." Study the response. That's your content plan.
Find one big question. What's the one concept that, if everyone understood it, they would need your product? Commit to creating the single best guide on the internet about it.
Spend an hour a week in one community. Go to a relevant subreddit or forum. Don't mention your product. Just answer questions and be helpful.
Be patient.
That's it. No "synergy." No "optimization funnels." Just being the expert in the room.
The Bottom Line
The old growth game was about who could best manipulate the Google algorithm. The new game is about who is the most trustworthy, helpful, and authoritative voice in the conversation.
The best "LLM Optimization" doesn't feel like optimization at all. It feels like being generous with your knowledge.
And that's not a strategy you can hack—it's just being the undeniable expert.