Updated
Updated What changed on
- Added Ahrefs' March 2026 citation study.
- Added the freshness evidence, including the part usually left out: AI assistants overall cite content 25.7% fresher than organic results, but Google's own AI Overview citations are just as old as its organic results.
How to improve AI search visibility across every engine.
The 30-second answer
Eight items cover the core work across ChatGPT, Perplexity, Claude, Gemini, AI Overviews and Copilot. Roughly 70% of the work serves all engines; 30% is engine-specific. The order to do it in: bot access first (cheapest, highest impact), then schema, then named author, then content depth, then engine-specific signals. If ChatGPT is the priority engine, use the how to rank in ChatGPT guide as the deeper checklist for Bing, OAI-SearchBot and extractable answers.
What the evidence supports, and what it does not
of pages cited in Google AI Overviews also rank in the top 10 for that query.
How it was measured: 863,000 keyword SERPs and 4 million AI Overview URLs. Ahrefs measures the top 10 as blocks, counting ads, featured snippets and video packs as separate blocks.
was the same measurement in July 2025, so the overlap between ranking and being cited has roughly halved in eight months.
How it was measured: 1.9 million citations from 1 million AI Overviews. Ahrefs notes it improved its parsing between this study and the 2026 update, so part of the change may be measurement rather than behaviour.
fresher is the content AI assistants cite, compared with the content ranking in organic results.
How it was measured: 16.975 million cited URLs across ChatGPT, Perplexity, Gemini, Copilot, AI Overviews and organic Google. Average age of AI-cited pages was 1,064 days against 1,432 days for organic results. Published July 2025 and last revised April 2026, so it predates the 2026 shift Ahrefs measured in its own citation study.
is the average age of Google's top three AI Overview citations, which is the same as its organic results. The preference for fresh content does not hold for AI Overviews.
How it was measured: Same 17 million citation dataset. This is the counterweight to the headline freshness figure and is usually left out when that figure is quoted.
The eight items, ordered by impact
| Item | Matters most for | Effort | Why |
|---|---|---|---|
| AI bot access (GPTBot, ClaudeBot, PerplexityBot, Google-Extended) | All engines | 1 day | Often blocked silently by Cloudflare or security plugins. Cheapest fix, biggest impact when it applies. |
| Schema (Article, FAQPage, Person, Organization) | All engines | 1-2 weeks | Hygiene factor. Ahrefs May 2026 controlled test on 1,885 pages found schema alone did not move citation rates - but quality sites carry it and Google uses it for extraction confidence, so ship it. |
| Named author + Person schema | Perplexity, Claude, AIO | 1 week | E-E-A-T signal. Anonymous brand content loses to named-expert content. |
| Content depth (1,500+ words on priority topics) | AIO, Claude | Ongoing | Ahrefs found 44% of AIO citations from positions 11-100 - depth is the differentiator. |
| Bing index health | ChatGPT, Copilot | 1 week | ChatGPT browsing + Copilot grounding both lean on Bing's index. Google rank does not transfer. |
| Content recency + dateModified hygiene | Perplexity, AI Mode | Ongoing | Perplexity especially weights recency heavily - 38% more citation for content updated within hours. |
| Knowledge Graph + Wikipedia entity | Gemini, Claude | Months | Slow build. Pays off across Google AI surfaces and Claude's source verification. |
| Reddit and community presence | ChatGPT (context, not citation) | Months | Ahrefs found ChatGPT cites Reddit only 1.93% of the time but uses it for context constantly. Shapes how AI engines understand your category. |
The shared 70%
Schema, content depth, named author, bot access, and structural credibility signals serve every major AI engine. Get these right first; they compound across surfaces.
The engine-specific 30%
- ChatGPT and Copilot: Bing index health is the differentiator. Both ground answers in Bing. For ChatGPT specifically, follow the ChatGPT ranking sequence before worrying about second-order GEO work.
- Perplexity: Recency bias is the strongest in the category. Refresh and re-date content that still matters.
- Claude: Verifiable primary sources. Cite the studies and data you reference inline.
- Gemini and AI Mode: Google ecosystem signals - YouTube, Knowledge Graph, structured data.
- AI Overviews: Top-100 position plus topical depth. The Ahrefs 38% top-10 figure means rank helps but is not deterministic.
What does NOT improve AI visibility
- Paying for ChatGPT Plus or Perplexity Pro.
- Submitting your site to AI engines directly (they crawl).
- Stuffing prompts into your content (the model sees through this).
- Hiding AI-only content behind a cloak.
- Buying AI-engine "boost" packages from agencies selling exactly that.
The highest-impact single move: publish original data
Per Digital Marketing Institute's November 2025 analysis, pages that publish original primary data see 30-40% higher AI search visibility than pages that only synthesise existing sources. The mechanism is simple: AI engines weight unique sourceable claims because they reduce hallucination risk. A page citing a statistic that exists nowhere else becomes the canonical source on that data point.
The practical version: stop rewording everyone else's research. Run one small study a quarter (a sample of 50 sites, 10 client interviews, a manual SERP audit) and publish the methodology + numbers. One sourceable original data point on a priority page outperforms three months of refreshing existing content.
Common questions on AI search visibility
Is SEO dead in 2026? No. Traditional Google search still drives the majority of organic discovery for most UK B2B verticals (StatCounter, Semrush). LLM referrals grew 72.7% YoY per Duda/SEJ April 2026 - but from a smaller base. SEO is changing under AI Overview pressure, not dying. Most teams should run both as one programme.
What are the best AI search visibility tools? Free baseline: the AI Search Visibility tool on this site. Paid tracking for mid-market: AIClicks ($59/mo), Otterly ($29/mo). Enterprise: Profound or BrandLight. Full ranked picks in the best AI rank trackers UK listicle.
How do I optimise AI search results when there is no rank? AI engines do not return a ranked list, but the citation pool is finite. Make sure you are in the pool (bot access + Bing indexing + named-author + schema) and the cited share-of-voice metric becomes measurable. You are not chasing rank 1 - you are chasing presence in the citation pool plus depth that wins repeated citation.
More on cross-engine visibility.
Is AI visibility the same across all engines?
No. Each engine weights signals differently. ChatGPT leans on Bing index and Reddit-for-context. Perplexity prioritises recency and source authority. Claude favours verifiable primary sources. Gemini draws on Google authority plus YouTube and Knowledge Graph. AI Overviews correlate with top-100 organic position plus schema. The shared foundation matters but engine-specific tuning matters too.
How long until AI visibility work shows results?
Technical fixes (schema, bot access) show in days. Content depth changes take three to six months across engines as they re-crawl and re-evaluate. Community presence work (Reddit, vertical forums) takes six to twelve months to build into citation patterns.
What is the cheapest thing I can do today?
Verify your AI bot access. Use the free AI Bot Robots.txt Tester on this site to check GPTBot, ClaudeBot, PerplexityBot, Google-Extended and others. If any are blocked, that is the cheapest fix and often the highest-impact one.
Do I need separate work for each engine?
Mostly no. About 70% of AI visibility work serves all engines (schema, content depth, named author, bot access). The remaining 30% is engine-specific - Reddit for ChatGPT, recency for Perplexity, Knowledge Graph for Gemini, FAQ schema density for AI Overviews.
How do I measure AI visibility?
Start with the free AI Search Visibility tool for a baseline. Move to a paid tracker (Profound, AIClicks, Otterly) for ongoing monitoring once you have a defined GEO programme. Sample weekly or fortnightly - daily creates noise because AI outputs vary per session.
Jason Burns
Independent UK SEO, GEO and AI consultant. 17 years in search. Portfolio includes 3M, BlackRock, Unilever and E.ON. Owner of SEO Moves Ltd since 2014.
More about Jason →Want a baseline reading before you start?
Run the free AI Search Visibility tool on your brand for a quick reading. Or send your priority queries and I run them across four engines and send the full diagnostic back.
Get the diagnostic