Last Tuesday I pulled up our top 5 posts by organic traffic and asked Perplexity to answer the same questions. Three of them didn't show up at all. Not buried on page one — just absent. The AI was pulling from competitor sites that weren't even ranking in the top 10 on Google.
So I picked three posts and rewrote them with GEO in mind. Here's what changed.
What I actually did
Post 1: "Best project management tools for remote teams"
Original version opened with a story about remote work trends. GEO version opened with a direct answer: "The best project management tools for remote teams in 2026 are [Tool A] for async teams, [Tool B] for hybrid, and [Tool C] for async-heavy workflows."
I moved the verdict to the top. I broke the comparison into a table with specific columns — pricing, async features, integrations, team size fit. I added a short "Why this matters" paragraph under each tool recommendation.
Post 2: "How to set up CI/CD pipeline"
Original was a 2,000-word tutorial with a narrative arc. GEO version kept the tutorial but added a TL;DR box at the top with the exact commands. I added numbered steps with exact syntax, not descriptions of what to do. I included a "Common failure" callout box after step 4 because that's where most people get stuck.
Post 3: "GEO vs SEO: what's the difference"
This one was already close but I restructured it as a comparison table first, explanation second. Added a definition block at the top:
> Generative Engine Optimization (GEO) is the practice of structuring content so AI-generated answers can accurately extract and cite it — distinct from traditional SEO which optimizes for search engine ranking signals.
The results after 3 weeks
| Post | Google position change | AI answer inclusion |
|------|----------------------|---------------------|
| PM tools | #4 → #3 | Now cited in 2/5 AI queries |
| CI/CD | #7 → #5 | Cited in 3/5 AI queries |
| GEO vs SEO | #2 → #1 | Cited in 4/5 AI queries |
The GEO vs SEO post is now pulling traffic from both Google and AI answer engines. I've been tracking this with a GEO Audit Tool that checks whether your content appears in AI-generated responses across multiple engines.
What I learned the hard way
1. AI engines don't scroll. If your answer isn't in the first 200 words, you're invisible. Front-load the conclusion.
2. Tables win. Every comparison I formatted as a table got cited. Paragraphs of the same info got ignored.
3. Specificity beats authority. Saying "Tool X costs $12/user/month" got cited more than saying "Tool X is affordable." AI engines want extractable facts.
4. Definitions matter. When I defined terms in blockquote format, the AI engines pulled those definitions verbatim.
I also ran our existing content through an AI Gravity Checker to see which pages were already AI-friendly and which needed work. About 30% of our posts had zero AI visibility despite ranking well on Google.
What I'd tell someone starting this
Don't try to optimize everything at once. Pick your top 10 posts by traffic, check if they appear in AI answers, and fix the ones that don't. The GEO vs SEO breakdown I wrote last month covers the structural differences in more detail if you're just getting started.
The pattern I'm seeing: content that ranks well on Google doesn't automatically show up in AI answers. The formatting requirements are different. Treat them as two separate distribution channels that happen to share an audience.