The 10 Most Common AEO Mistakes (and How to Fix Them)
Why isn't your brand showing up in ChatGPT? Most of the time it's one of these ten mistakes. Here is the diagnosis and the fix for each.
Summary — Most "we don't show up in ChatGPT" complaints trace back to one of ten root causes. Nine of them are fixable in a week. Here is the diagnosis and the fix for each.
For the framework around all of this, see What is AEO?.
The 10 mistakes, in priority order
1. AI crawlers are blocked in robots.txt
Diagnosis. Visit /robots.txt. Look for Disallow: / under GPTBot, PerplexityBot, Google-Extended, or ClaudeBot.
Fix. Allow them. See robots.txt for AI crawlers.
This is the highest-leverage 5-minute fix, and 30%+ of sites still get it wrong in 2026.
2. The homepage doesn't define what you do
Diagnosis. Read your homepage's first paragraph aloud. Could a stranger answer the question "what does this company do?" in one sentence?
Fix. Rewrite the first 100 words as a literal definition. Pattern: "[Brand] is a [category] that helps [audience] [do thing]."
LLMs lift this kind of sentence verbatim. Brand storytelling at the top costs you citations.
3. No Organization schema
Diagnosis. View the homepage source. Search for "@type": "Organization".
Fix. Add it. See schema markup for AEO for a copy-paste template. Include sameAs, logo, founder, and description at minimum.
4. Content is built for humans only — no quotable structure
Diagnosis. Open your top 10 pages. Count how many use H2/H3 headings as questions, lists, comparison tables, or numbered steps.
Fix. Restructure. The five formats LLMs love most:
- Definition at the top
- Comparison tables
- Numbered playbooks
- FAQs (with FAQPage schema)
- Listicles with explicit criteria
If your content reads like a magazine article, it gets cited less than content that reads like an encyclopedia entry.
5. Stale dates everywhere
Diagnosis. Check the visible date on your top 20 evergreen posts. Anything older than 12 months is a freshness liability.
Fix. Refresh quarterly. Add 200+ words of new info, update dateModified, and bump the visible date. Don't game it — actually improve the content.
6. No third-party citation graph
Diagnosis. Search Google for "[Brand name]" -site:[your-domain]. Count the unique authoritative third-party domains that mention you. Under 10? You have a citation-graph problem.
Fix. This is the slowest mistake to fix and the highest payoff. Pitch listicles, contribute guest posts, claim G2/Capterra/Product Hunt, work toward Wikipedia eligibility. See the authority section of the 25-step AEO checklist.
7. Optimizing only the homepage
Diagnosis. Pull your top 10 brand-mentioning AI answers (across ChatGPT, Perplexity, AI Overviews). Check which URLs of yours got cited. If 80%+ of citations are the homepage, you have a long-tail content gap.
Fix. Build a topical cluster — a pillar page plus 4–6 cluster pages, all internally linked. Long-tail prompt visibility comes from cluster pages, not the home page.
8. AI-generated thin content
Diagnosis. Did you ship 50+ pages of generated content in the last 12 months without human editing or original research?
Fix. Cut the dead weight. AI engines and Google have both retrained on this pattern. Pages that pattern-match low-effort generation are down-weighted in retrieval, regardless of factual accuracy.
Quality > quantity. One excellent article will outperform 30 mediocre ones for AEO.
9. No measurement loop
Diagnosis. Can you say what your share of voice is in ChatGPT this week, vs. last week, vs. four weeks ago? If not, you're flying blind.
Fix. Build (or buy) a prompt-tracking system across at least 100 buyer-relevant prompts and the 4 major engines. See Measuring share of voice in AI answers and the best AEO tools.
10. Treating AEO as a one-time project
Diagnosis. Look at your AEO work plan. Is it a project with an end date, or a continuous program?
Fix. Make it continuous. Models retrain. Live retrieval shifts. Competitors publish. AEO visibility is a moving target — a "we did our AEO sprint last quarter" mindset will erode within months.
A diagnostic flow
Run through this in order. Stop at the first yes:
- Are AI crawlers blocked? → fix robots.txt
- Does the homepage define what you do in <100 words? → rewrite intro
- Is Organization schema present? → add it
- Are your top 10 pages structured for quotability? → restructure
- Are dates fresh on evergreen content? → refresh
- Are 10+ authoritative third parties citing you? → start outreach
- Does only the homepage get cited? → build cluster pages
- Is there obvious thin AI content? → cut it
- Are you measuring SOV weekly? → set up tracking
- Is AEO a project or a program? → make it a program
Most teams find their gap in items 1–4 and don't need to even reach the harder ones.
What "fixed" looks like, by week
| Week | Outcome |
|---|---|
| 1 | Robots.txt + homepage intro + Organization schema |
| 2 | Top 10 pages restructured + dates refreshed |
| 3 | Citation outreach started; 5 listings claimed |
| 4 | First cluster page shipped; tracking dashboard live |
| 8 | First measurable SOV gain on tracked engines |
| 12 | Compounding visibility across multiple engines |
What's not on this list (and why)
- AI-content detection scores. Internal heuristics, not stable.
- Specific word counts. "1,500 words" isn't a magic number. Quality and density matter more.
- Backlink count alone. Citation graph > raw link count for AEO.
- Keyword density. LLMs are semantic; they don't care.
The bottom line
If you're not showing up in AI answers, the problem is almost always one of these ten things. Run the diagnostic flow, fix the top blocker, and re-measure.
Get a free AEO scan to see exactly which of these mistakes are costing you mentions.