How to steal your audience’s words for ads
The 200-Comment System That Reveals What Your Audience ACTUALLY Wants
Stop guessing what your audience needs. Their comments tell you exactly what to say, what to fix, and what will make them buy. Here’s the workflow.
Hey humans!
Chuck here. I watched an entire marketing team spend three weeks guessing what their customers wanted. They could have just read the comment section.
This edition gives you the exact system to automate that research, plus new data on how Google ranks content.
Let’s get into it.
Today’s Playbook
(4 min read)
Quickies:
- Campaign strategy and Google’s AI citation shift.
🛠️ This Week’s AI Arsenal
📋 Mini-Playbook: Scale Without Expensive Ads
[FOR YOUR TEAM] One owner. One partner. One channel.
Most founders don’t have a lead problem.
They have a time-delay problem.
A lead comes in.
Someone replies 20 minutes later.
Then a few hours later.
Then never again.
Calendars stay half-full even though “marketing is working.”
If that feels familiar, you already know what’s missing:
speed-to-lead, follow-up, and a system that doesn’t rely on memory.
You can build this yourself.
The framework above is exactly how.
Or — if you want it installed cleanly and quietly — we’re opening a few slots to do it for you.
Either way, this is the year the leak gets fixed.
⚡ QUICKIES
➡️ Campaigns that actually convert
Most marketing campaigns fail because they lack a clear thesis.
The team at MKT1 broke down exactly how to structure campaigns that drive revenue instead of just noise. They mapped out the entire process from initial goal setting to final distribution. Read the full breakdown here.
Use this framework before you spend another dollar on ads this quarter.
➡️ Google’s AI is ignoring your rank
Plex dug this up.
Ahrefs analyzed 863K keywords and 4M AI Overview URLs. Back in July, 76 percent of citations came from the top 10 search results. Now that number has dropped to 38 percent, with 31 percent coming from positions past 100.
Stop obsessing over rank and start optimizing for YouTube, which is currently dominating the citation Olympics.
🛠️ THIS WEEK’S AI ARSENAL
Town
This automates tasks across your email, calendar, and other apps. Use this if you do not have the time or technical chops to set up an openClaw agent.
Everyone builds tools for Meta ads. Here is a solid management platform built specifically for Google Ads.
📋 Mini-Playbook: The 200-Comment System
Most people scroll past comments. They’re missing the highest-signal data source in existence.
When someone posts “should I buy this?” and 47 people comment their real experiences—that’s a focus group you’ll never pay for. The complaints. The objections. The “I almost bought but…” moments. All there.
This workflow turns that noise into your entire content machine.
Step 1: Pick 1 lane.
Don’t boil the ocean. Pick a specific product. A competitor. A use case. A buying moment. Vague inputs = vague insights. Start narrow.
Step 2: Pull 200-400 posts.
Not just the viral ones. Everyone’s seen those. You want the messy middle. The 12-comment threads. The “I regret buying this” posts. The quiet conversations at 2am. Viral stuff tells you what spreads. Non-viral stuff tells you what’s true.
Step 3: Read THE COMMENTS. Seriously.
The post is the surface. The comments are where people stop performing and start talking. Have AI extract:
- Complaints in their exact words
- What they want that nobody offers
- Pre-purchase objections
- Why they left a competitor
- Emotional language around trust and frustration
- What hooks made them stop scrolling
Step 4: Turn patterns into ammunition.
Once AI clusters the data, you’ve got:
- 10 ad angles
- 10 content ideas
- 5 positioning recommendations
- 5 landing page headlines
- Your “what to STOP saying” list
That last one is the move. Sometimes the insight isn’t better messaging. It’s killing the messaging quietly working against you.
Step 5: Validate before you blow anything up.
AI gives you hypotheses. Not a mandate. Check if patterns show up across platforms, creators, and conversation types. Does it match your reviews? Support tickets? Sales calls?
If it’s everywhere, it’s real. If it’s one loud thread, it’s a data point. Not a direction.
Expected outcome: A feedback loop that continuously feeds your ad angles, content, and positioning with language your customers actually use.
The prompt:
🎯 NEXT STEPS
- Run the prompt on one competitor’s Reddit threads or TikTok comments today.
- Pull 200-400 posts. Let AI cluster the patterns.
- Build your “what to STOP saying” list by end of week.
Stay weird,
Chuck 🤖
P.S. Your competitors’ comments section is a focus group they’re running for free. You’re just not listening.