Hey, it's me, Jess. Head of Marketing at Exit Five, back with another edition of The Prompt. And today it’s all about how to rank #1 in LLMs.
It should take 30-60ish days to start showing up in ChatGPT, Perplexity, and Google AI Overviews if you know the system.
And you don't need enterprise budgets or massive content teams to make it happen.
The best part? Early-stage companies actually have an advantage right now.
While bigger competitors dominate search, they're still figuring out their Answer Engine Optimization (AEO) strategy. Which means you can move fast and claim territory in LLM results.
Something I've noticed lately while researching tools for this newsletter: the companies showing up inside LLMs aren't always the obvious ones. I'll find products I've never heard of – early-stage startups with small teams, minimal content, and barely any social presence.
And yet, they're ranking right alongside much bigger brands.
You don't need years of SEO history. You need clear expertise on specific topics, original data, and well-structured content.
Here's the system you should steal to start showing up in AI answers ASAP.
Part 1: How to Engineer Content That Ranks #1
This isn’t about praying to the algorithm gods. It’s about reverse-engineering what works and then doing it better than everyone else.
I recently listened to the Marketing Against the Grain podcast with Ethan Smith, CEO of Graphite, who’s been running this exact playbook. His team drives 30% to 120% organic traffic growth by treating AI as a research assistant, not a writer.
Here's the system they’re using right now:
Step 1: Stop Guessing, Start With Data
Before writing anything new, audit what you already have.
To pressure-test this, I tried a tool called Otterly AI. Full transparency: I’ve only used it once, so I’m not here to crown it the solution. I used it to run a quick audit on our site and do some early prompt research, basically looking at how our pages might show up (or not) in LLM-driven search.
What it confirmed: you don’t need a massive SEO machine to start understanding or improving how you show up in AI answers. There are a growing number of tools tackling this from different angles, and even your existing stack (like Google Search Console) can give you useful signals if you know what to look for.
The bigger point isn’t which tool you use. It’s that this kind of visibility is suddenly accessible way earlier than it used to be.
Step 2: Reverse-Engineer the SERP
Here’s where most people go wrong. They ask AI to “write a blog post on [keyword].” That’s like asking a chef to “make food.” You’ll get something generic and forgettable.
Instead, start with the SERP.
Export the top 10 URLs for your target query and feed them into Claude. Ask it to analyze what Google is rewarding right now:
What page types are winning? (guide, listicle, how-to)
How deep is the content?
How readable and skimmable is it?
What formats and structures keep showing up?
This isn’t guessing. It’s pattern analysis.
And it matters because LLMs don’t exist in a vacuum. They rely on many of the same signals Google does, like authority, consistency, and content that already performs well in search. If you’re winning the SERP, your odds of showing up in LLM answers go way up.
But here’s the shift: Ranking isn’t the finish line anymore. LLMs aren’t just ranking pages. They’re assembling answers. Once you understand what’s winning in search, the real opportunity is making your content easier for AI to pull from, not just rank.
Step 3: The “Human Sandwich”
Yes, AI drafts the content based on your brief. But then? A human editor spends 5 to 7 hours refining it (no, really).
They're removing fluff. They're injecting SME insights, case studies, and quotes that AI can't invent. They're weaving in internal links to build site authority. They're making sure the piece doesn't sound like it came from a content farm.
Because here's the reality: Google can smell AI slop from a mile away. But it can't penalize content that's genuinely useful, deeply researched, and written (or heavily edited) by humans.
This hybrid approach keeps you Google-safe while moving 10x faster than manual production.
Part 2: The Fast-Track Tactics for Early-Stage Companies
Here's how to start ranking in ChatGPT, Perplexity, and Google's AI Overviews quickly:
Tactic 1: Own the Ultra-Specific Questions
LLMs love detailed, specific answers to niche questions. Instead of competing for “marketing automation,” target “how to set up lead scoring in HubSpot for SaaS companies under 50 employees.”
These questions have less competition, but LLMs pull them constantly because they match real user queries perfectly.
🏆 Quick win: List 10 ultra-specific questions your customers actually ask. Write definitive answers (1,000–1,500 words). Publish. Check to see if you start showing up in AI answers within a few weeks.
Tactic 2: Create “Citation-Worthy” Data
LLMs are far more likely to reference sources that contribute something new to the ecosystem, like original data, research, or clear frameworks. You don’t need a massive survey to do this. Start small:
Analyze 50 job postings to surface salary ranges or role expectations
Interview 10 customers and extract repeatable patterns
Create a simple benchmark (conversion rates, adoption stats, time-to-value)
Build a framework with a memorable name
Original data gives AI something concrete to pull from, but it gets stronger when that data is referenced by others.
Example: our B2B Marketing Salaries Report sometimes shows up in AI answers not just because it’s original, but because it’s been cited and discussed across multiple sites.
🏆 Quick win: Pick one data point you can own and publish it with a clear methodology. Then make it easy for others to reference, clean charts, quotable stats, and a simple URL. That’s how data turns into citations.
Tactic 3: Answer the “vs.” Questions
Comparison content is LLM gold. “X vs Y” queries are exploding, and most companies ignore them.
Write honest, balanced comparisons:
“Mailchimp vs HubSpot for startups”
“SEO vs paid ads for B2B SaaS”
“In-house marketing vs agency for seed-stage companies”
Be fair. Include your competitors. LLMs reward comprehensive, unbiased answers.
🏆 Quick win: Identify 5 comparison queries in your space. Include a feature table, pros/cons, and a “best for” recommendation.
Tactic 4: Update Your Existing Content with “LLM Hooks”
You probably already have content that's ranking on page one. Make it citation-worthy by adding:
A clear, quotable definition in the first paragraph
Bulleted “key takeaways” that LLMs can extract easily
Step-by-step instructions with numbered lists
Data points with sources
A “quick answer” summary box at the top
🏆 Quick win: Pick your top 3 ranking posts. Add a “Quick Answer” summary at the top (3–4 sentences). Add one unique data point or framework. Republish.
Tactic 5: Build Topic Authority, Not Just Individual Pages
LLMs don't just pull from one article. They pull from sites that demonstrate comprehensive topic authority.
Instead of writing random blog posts, cluster your content:
One pillar page (comprehensive guide)
5–7 supporting articles (deep dives on subtopics)
Internal links between all of them
Example: If you're targeting “demand generation,” create:
Pillar: “Complete Guide to Demand Generation”
Subtopics: “Account-based marketing tactics,” “Lead scoring frameworks,” “Content distribution channels,” etc.
🏆 Quick win: Pick one topic cluster. Map out 1 pillar + 5 subtopics. Write the pillar page first. Add one subtopic per week.
🔍 How To Track Whether You're Actually Showing Up
You’re probably wondering, “But how do I know if this is working?”
So let's be clear about what you can and can't track:
You cannot see what random people around the world are asking ChatGPT, access real-time global AI search trends like Google Trends, or get exact search volume for AI queries.
You can define specific queries you care about and track those, monitor your own brand visibility systematically, see patterns in what your content gets cited for, and track referral traffic from LLMs to your site in GA4.
The question isn't only “what is everyone searching?” It’s: “When people search for solutions in my space, does my brand show up?” And now you can answer that question.
🛠️ The Tools To Try
A whole new category of tools emerged to help marketers track visibility in AI search. Think of them as Google Search Console for the AI era. These tools monitor whether your brand appears when people ask questions in your space, which competitors get cited instead, and which of your pages LLMs actually pull from.
Otterly.AI
($29–$989/month based on the number of prompts monitored)
Monitors when your brand appears across ChatGPT, Perplexity, Google AI Overviews, Gemini, and Microsoft Copilot. The killer feature: Shows your “Share of AI Voice” – the % of citations you own vs. competitors for specific queries.
Semrush
($199–$549/month)
Their AI Visibility Toolkit tracks how you show up in Google AI Overviews and provides topic research tools to identify conversational queries. Best if you're already using Semrush for SEO and want to layer on basic AEO tracking without switching platforms.
Ahrefs
($129–$449/month)
Brand Radar tracks your visibility across 6 AI platforms (Google AI Overviews, ChatGPT, Perplexity, Gemini, Copilot, and AI Mode) using 190M+ real prompts. Shows which content is getting cited, competitor mentions, and includes YouTube and Reddit tracking. There’s a free tier available for basic monitoring.
Peec.ai
(Starts at ~$100/month with a 7-day free trial)
Focuses on three core metrics: Visibility (how often you show up), Position (where you rank in AI answers), and Sentiment (how you're described). The clean interface makes it easy to track prompt performance and identify which sources AI platforms cite most.
👆 These came highly recommended on our
latest AEO webinar
as the most practical options for tracking and improving AI search visibility.
Start with what you can actually execute on.
Everyone wants a shortcut. Everyone wants to prompt their way to page one. And that’s exactly why most content is mediocre.
If you want to rank #1, and stay there long enough to show up in AI answers, you have to do the work others won’t. The 5-hour edit. The SME interviews.
Because AEO isn’t about gaming the system. It’s about being the best answer. And that takes effort.
But you don't need to do everything at once. Start with what you can actually execute on.
- Jess
P.S. Even though I use AI, I am not AI. It’s really me, Jess. I wrote this sitting in my garage turned office, drinking cold brew and watching my keyword tracker like it’s the stock market.
P.P.S. Reply and tell me: Are you still doing SEO the 2019 way, or have you started thinking about AEO? I’d love to hear where you’re at.
đź«– Spilling the AI Tea (Recent News)
🚀 Perplexity Launched Comet Browser (And it’s Free for Everyone)
Perplexity released Comet, an AI-powered browser now available globally for free, featuring Background Assistants that can run multiple AI research tasks simultaneously while you browse.
Why it matters for marketers: This isn't just another browser – it's a research tool that runs AI workflows in the background while you work. Instead of toggling between ChatGPT tabs for competitor research or content ideas, Comet handles multiple research threads simultaneously.
If you're spending hours on competitive intel or market research each week, this could compress that work significantly. Worth testing if research is a regular part of your workflow.
Join us in Scottsdale, AZ (March 18–20) for an exclusive gathering of 100 B2B Marketing Executives. This is two days out of the office at a beautiful resort, specifically designed to connect top B2B marketing leaders IRL.
The lineup Includes:
Dave Kellogg (EIR, Balderton Capital) on becoming a top‑tier CMO.
Hillary Carpio (VP Growth, Snowflake) on building high‑performing teams.
Kim Storin (CMO, Zoom) on her leadership playbook.
Emily Kramer (Founder, MKT1) on Ecosystem Marketing.
