
AI Citation Building Strategies for New Websites
Key Facts
- 85% of brand mentions in AI answers come from external domains, while only 13.2% originate from brand-owned websites M&C Saatchi Performance
- Branded web mentions correlate 3x more strongly with AI visibility than backlinks, with a Spearman correlation of 0.664 versus 0.218 for links sapt.ai
- YouTube mentions show the highest correlation with AI visibility at 0.737, effectively displacing traditional link-building as the core signal sapt.ai
- 82% of AI citations come from earned media rather than brand-owned content aira.net
- Gemini cites YouTube in 47.1% of answers, making video presence essential for that platform TryPrefer
- Perplexity cites Reddit in 40.8% of answers and LinkedIn in 30.4%, favoring community and professional networks TryPrefer
- Pages updated within two months earn 28% more citations, and AI platforms cite content approximately 25.7% fresher than traditional search results sapt.ai
Why Your New Site Isn't in AI Answers (Yet)
New domains assume rankings and backlinks drive AI citations, but the data shows the opposite. When a brand-new website launches, it is common to expect that established ranking factors will immediately translate into visibility within AI answers. However, the research consistently demonstrates that AI engines prioritize third-party consensus and brand authority over traditional SEO metrics, meaning new sites must earn citations through visibility elsewhere rather than relying on on-page optimization alone.
A critical finding from the research reveals that 85% of brand mentions in AI answers come from external domains, while only 13.2% originate from brand-owned websites (M&C Saatchi Performance). This statistic alone fundamentally shifts the strategy for new websites: relying solely on owned content will not generate the citations needed to appear in AI answers. Instead, new sites must focus on earning visibility across third-party platforms where AI engines actively scan for consensus. Furthermore, brand mentions across the web correlate three times more strongly with AI visibility than backlinks do, with a Spearman correlation of 0.664 for mentions compared to 0.218 for backlinks (sapt.ai). YouTube mentions show an even higher correlation at 0.737, effectively displacing traditional link-building as the core signal AI engines reward.
The zero-click reality further complicates the assumption that rankings equal citations. Approximately 68% of U.S. Google searches end without a click, and when AI Overviews appear, top organic click-through rates drop from roughly 20% to under 10% (M&C Saatchi Performance). Domain age barely factors into the equation; retrieval processes and brand knowledge decide what gets cited, not page properties. In fact, page properties explain only 2.8% of citation variation, while the engine's retrieval process explains 32% (arXiv). For new websites, this means that domain authority is secondary to being mentioned and referenced across the broader web.
To earn the first AI citations, new websites should prioritize digital PR and earned media, as 82% of AI citations come from earned media sources rather than brand-owned content (aira.net). Establishing presence on platform-specific sources also matters: Gemini cites YouTube in nearly half of answers, while Perplexity favors Reddit and LinkedIn (TryPrefer). Building brand mentions across these third-party channels, maintaining fresh content, and pursuing multi-source consensus are the most effective pathways for a new domain to gain visibility in AI answers.
New websites often assume that traditional SEO tactics, keyword stuffing, backlink acquisition, and on-page optimization, will translate directly into AI citations. The research tells a different story. AI engines do not simply index a website’s pages and serve them up; they rewrite user queries into multiple sub-queries, search the live web in real time, and cite sources based on retrieval relevance and brand consensus. According to a TryPrefer study analyzing 240 questions across major engines, 89.7% of all answers came from a live search, meaning the engine actively retrieved and ranked pages to formulate each response. This retrieval-driven process explains 32% of citation variation, while page-level properties like structured data or keyword density account for only 2.8%.
The mechanics reveal why domain authority alone rarely guarantees a citation. When a user asks a question, the engine generates new sub-queries designed to match the user's intent across the widest possible set of sources. It then ranks those retrieved pages by how well the content matches the meaning of those sub-queries, and cites mostly from the top of that engine’s own ranking. For new websites with low domain authority, this means earning a citation depends less on optimizing a single page and more on ensuring the brand appears across the specific platforms the engine prioritizes. For example, Gemini cites YouTube in 47.1% of answers, while Perplexity cites Reddit in 40.8% and LinkedIn in 30.4% of answers. ChatGPT, by contrast, rarely cites Reddit or YouTube at all, preferring Google Search Central resources.
This platform-specific behavior creates a fragmented citation landscape. Research from M&C Saatchi Performance found that 68% of brand mentions were unique to a single AI model, meaning a brand can be prominent in ChatGPT answers while entirely absent from Perplexity or Gemini. Furthermore, third-party content earns 85% of AI citations, while only 13.2% come from brand-owned domains. For a new website, this underscores the necessity of appearing on the right third-party platforms rather than relying solely on owned content. YouTube mentions showed the highest correlation with AI visibility at 0.737, displacing traditional link-building as the core signal engines reward.
The research also highlights a critical zero-click reality. When AI Overviews appear in search results, top organic click-through rates drop from 20.02% to 9.69%, a decline of approximately 52%. Zero-click rates reach 74.3% on SERPs with AI Overviews versus 53.6% without. This shift means new websites must prioritize AI citation acquisition over traditional ranking metrics. The good news is that AI-referred visitors convert at 14.2%, compared to 2.8% for Google organic, a 5x advantage that makes earning a citation in an AI answer a high-value outcome.
For new websites with limited domain authority, the path to citations lies in digital PR and multi-source consensus. Muck Rack’s 2025 report analyzed over one million links cited by AI platforms and found that 82% of AI citations come from earned media rather than brand-owned content, with journalism alone accounting for 20-30% of citations. BuzzStream’s April 2026 analysis of 4 million citations found 14% from news publications and 53% from blog content. Brand mentions across the web correlate 3x more strongly with AI visibility than backlinks do, based on Ahrefs' analysis of 75,000 brands. To earn citations, new websites should focus on getting mentioned in industry publications, expert roundups, and comparison lists, platforms where the target AI engine already scans for consensus. Appearing in the top three positions of those lists particularly strengthens citation probability, as AirOps analysis of 21,311 brand mentions found that 80% of cited brands appeared in the top three positions of third-party lists.
Structured data and technical SEO have limited within-website effect. The arXiv study found that recommended GEO practices including FAQ blocks, summary boxes, and author boxes appear effective only across websites, not within them. However, content freshness does influence citation rates; pages updated within two months earn 28% more citations, and AI platforms cite content approximately 25.7% fresher than traditional search results. For a brand-new domain, this means maintaining active, updated profiles on YouTube, Reddit, and LinkedIn, platforms that specific engines prioritize, can be more impactful than tweaking on-page elements alone.
The New-Site Citation Framework: Entities, Platforms, and Earned Media
The research is clear: AI citation building operates on entirely different principles than traditional SEO. For new websites with low domain authority, the path to visibility runs through digital PR, platform-specific presence, and third-party consensus rather than link acquisition or on-page optimization alone. The data shows that 82% of AI citations originate from earned media, and brand mentions across the web correlate three times more strongly with AI visibility than backlinks do. This means a new site’s best investment isn’t in building more links, but in earning coverage and mentions across the platforms AI engines actively scan.
A critical differentiator is how each generative engine weights its sources. Gemini favors video content, citing YouTube in nearly half of all answers, making a brand’s YouTube presence a non-negotiable factor for visibility on that platform. Perplexity, meanwhile, leans heavily on community and professional networks, citing Reddit in over 40% of answers and LinkedIn in about 30%. ChatGPT and Claude follow their own distinct patterns, ChatGPT rarely cites Reddit or YouTube but leans on Google Search Central, while Claude avoids Reddit entirely and prefers sources like Profound. For a new website, this means there is no one-size-fits-all approach; the strategy must align with which engine the target audience uses most.
Beyond platform preferences, the research highlights the power of multi-source consensus. AI engines scan for agreement across independent sources before citing a brand, and nearly 90% of third-party citations come from reviews, comparisons, and listicles. Brands that occupy the top three positions within those lists see a significant boost in citation probability. Freshness also matters, content updated within the last two months earns 28% more citations, and AI platforms preferentially cite material that is roughly 25.7% fresher than traditional search results. For a new domain, this reinforces the need to keep not only owned content current but also third-party profiles and listings active and up to date.
The zero-click reality further shifts the goalpost. With the majority of searches ending without a click and top organic CTR dropping by over 50% when AI Overviews appear, ranking in traditional results matters less than being cited within AI answers. The good news is that AI-referred traffic converts at a rate five times higher than Google organic, making citation acquisition a high-value objective. For new websites, the framework is straightforward: prioritize digital PR to earn brand mentions, establish presence on the platforms each AI engine rewards, and ensure consistent, fresh representation across review sites and industry publications. This multi-pronged approach is what the data shows actually moves the needle for low-authority domains.
New websites chasing AI citations quickly discover that traditional SEO formatting alone won't move the needle. The retrieval mechanics of generative engines favor content that answers the actual question a user is asking, positioned at the beginning of the page. Research from Superlines indicates that 44.2% of AI citations originate from the first 30% of content, meaning the opening sections must deliver the answer immediately rather than burying it in background context. This answer-first structure aligns with how ChatGPT, Gemini, and Perplexity parse queries, each engine rewrites the user question into new sub-queries and ranks pages by how well the content matches that semantic intent.
Freshness represents another critical lever. Data from SE Ranking shows that AI platforms cite content 25.7% fresher than traditional search results, and pages updated within a two-month window earn 28% more citations. For a new domain, this means maintaining a regular content update schedule isn't just about keeping information current, it's a direct citation strategy. Google's own AI Optimization Guide explicitly warns against tactics like generating inauthentic mentions or using AI-rewritten content solely to game visibility; the guidance emphasizes that genuine freshness, not mechanical churn, is what engines reward.
Structured data and schema markup occupy a nuanced space in this equation. While the arXiv research found that technical page properties like FAQ blocks, summary boxes, and author boxes explain only 2.8% of citation variation across websites, the same studies note these elements do have an effect "only across websites, not within them." For a new site with limited authority, implementing schema for organization, products, or FAQs provides a baseline of semantic clarity that helps AI engines categorize the content, but it should not be viewed as a primary citation driver. The real SEO value lies in ensuring the content itself matches user queries, pages whose titles and paragraphs directly reflect the questions people ask are cited significantly more often than those stuffed with technical tags.
Perhaps the most important formatting consideration is avoiding the "AI-rewritten content" trap. Google's guidance is clear: generating content solely for AI systems, using aggressive chunking, or creating llms.txt files does not improve visibility and may actively harm it. Instead, the focus should be on formatting that serves human readers first, with clear headings, concise answers, and naturally structured information that AI engines can reliably extract. For new websites, this means every page should answer the primary question within the first few paragraphs, update data regularly to maintain that 25.7% freshness advantage, and resist the urge to optimize exclusively for machine parsing at the expense of genuine usefulness.
Book a free AI visibility audit to see how your new site can start earning AI citations.
Your First 90 Days: Tracking Citations and Compounding Visibility
The first ninety days after launch represent a critical inflection point for any new website. In this window, the patterns you establish, both in how you format content and how you build visibility across third-party platforms, determine whether AI engines discover you at all. The research is clear: AI citation building operates on fundamentally different principles than traditional SEO. While backlinks still carry some indirect weight, branded web mentions correlate three times more strongly with AI visibility than links do (sapt.ai). For a website starting from zero domain authority, this means the priority must shift from link acquisition to earning brand consensus across the open web.
The data shows that 85% of brand mentions in AI answers come from external domains, with only 13.2% originating from brand-owned websites (M&C Saatchi Performance). This reality forces a new strategy: visibility must be earned off-site. Digital PR emerges as the dominant pathway. Muck Rack’s 2025 analysis of over one million links found that 82% of AI citations stem from earned media, with journalism alone accounting for 20 to 30% of those citations (aira.net). BuzzStream’s more recent review of 4 million citations reinforced this, showing 14% from news publications and 53% from blog content (aira.net). For a new site, this isn’t about publishing press releases; it is about creating newsworthy data, expert commentary, or industry insights that journalists and editors will reference. That earned media coverage then becomes the fuel AI engines scan for consensus.
Engine-specific preferences add another layer of complexity. The TryPrefer study, which audited 240 real-world queries across four major engines, revealed stark differences in what each platform cites. Gemini cites YouTube in 47.1% of answers, making video presence essential for anyone targeting that engine. Perplexity, meanwhile, cites Reddit in 40.8% of answers and LinkedIn in 30.4% (TryPrefer). ChatGPT never cites Reddit or YouTube but leans heavily on Google Search Central, while Claude avoids Reddit entirely but cites Profound most frequently (TryPrefer). This means a one-size-fits-all approach fails. If your target engine is Perplexity, maintaining active, authoritative Reddit and LinkedIn profiles becomes non-negotiable. If Gemini is the priority, a well-optimized YouTube channel with complete metadata, transcripts, descriptions, and chapter markers, carries significant weight.
Content freshness is another lever new websites can control. SE Ranking’s data shows pages updated within two months earn 28% more citations, and AI platforms preferentially cite content approximately 25.7% fresher than traditional search results (sapt.ai). This applies both to your own site and to third-party profiles. Keeping G2 listings, industry directory profiles, and review site entries current signals to AI engines that the brand is active and relevant. YouTube exemplifies this principle: its structured metadata makes it easily parseable, and the research notes YouTube overtook Reddit as the most-cited social platform in AI answers in early 2026 (sapt.ai). For a new website, publishing helpful video content and ensuring its metadata is complete offers a dual benefit: direct audience reach and AI citation potential.
Perhaps the most counterintuitive finding involves measurement. The TryPrefer study ran each question three times per engine and found run-to-run consistency varies dramatically: Perplexity showed 96% overlap between runs, Claude 68%, ChatGPT only 37%, and Gemini 35% (TryPrefer). This means a single query run will not give you a complete picture of an engine’s citation behavior. The recommendation is clear: track each engine separately, run queries multiple times, and do not blend numbers across platforms. What works for ChatGPT may not register for Perplexity, and aggregated metrics will mask these distinctions. For a new website with limited resources, this focused, engine-by-engine approach ensures every action counts toward the platforms your audience actually uses.
The zero-click reality further complicates the landscape. With 74.3% of searches ending without a click when AI Overviews appear (M&C Saatchi Performance), traditional ranking metrics matter less than AI citation acquisition. However, the research also notes a silver lining: brands cited within AI Overviews earn 35% more organic clicks and 91% more paid clicks (sapt.ai). The strategy shifts from ranking below the AI answer to being part of it. For a new website, this means the goal isn’t just to appear in AI answers, it is to earn a citation that drives qualified traffic at a conversion rate 5x higher than Google organic. AI-referred visitors convert at 14.2% versus 2.8% from Google organic (Seer Interactive), making AI visibility not just a visibility play but a revenue driver.
The 90-day framework for a new website, then, looks like this: weeks one through four focus on digital PR outreach, pitching unique data, expert commentary, or industry trends to journalists and publications. Weeks five through eight prioritize platform-specific profile optimization: a YouTube channel for Gemini, Reddit and LinkedIn participation for Perplexity, and Google Search Central resources for ChatGPT. Weeks nine through twelve shift to maintenance: updating third-party listings, refreshing owned content, and running the engine-specific query tracks three times per engine to measure citation shifts. Throughout all twelve weeks, the goal is multi-source consistency. AI engines scan for agreement across independent sources; a brand mentioned in a trade publication, reviewed on G2, and discussed on Reddit triggers citation far more reliably than a brand with a single mention anywhere on the web.
The compounding effect of this approach becomes visible over time. Each earned media mention, each platform-specific profile, each fresh content update adds another data point into the AI engines’ retrieval processes. After 90 days, you can benchmark where you stand with a free AI visibility audit, seeing which engines cite your brand and where the gaps remain. The new website that treats its first 90 days as a citation-building sprint, prioritizing earned media over links, optimizing for engine-specific platforms, and measuring engine-by-engine, positions itself not just to be found, but to convert. AI-referred traffic converts at 14.2%, and with the right foundation in place, your site can start earning those citations from day one.
Sequencr.ai reports that generative AI search drives 10%+ of traffic, confirming that AI-mediated discovery is no longer experimental, it is a meaningful traffic channel alongside traditional search.
First Page Sage notes ChatGPT holds 59.70% of the AI search market share as of May 2025, though this dominance has slowly eroded from 76.4% in January 2024 to 74.2% in April 2025, while Perplexity more than doubled from 2.7% to 6.3% over the same period.
M&C Saatchi Performance found that 85% of brand mentions in AI answers come from external domains, with only 13.2% from brand-owned websites, and that zero-click rates reach 74.3% on SERPs with AI Overviews versus 53.6% without.
sapt.ai reports branded web mentions correlate 0.664 with AI visibility versus 0.218 for backlinks, and that pages updated within two months earn 28% more citations.
TryPrefer found Gemini cites YouTube in 47.1% of answers and Perplexity cites Reddit in 40.8% of answers, with run-to-run overlap as low as 37% for ChatGPT, underscoring the need for engine-specific tracking.
aira.net states 82% of AI citations come from earned media rather than brand-owned content, with journalism alone accounting for 20-30% of citations.
Sequencr.ai further notes AI-referred visitors convert at 14.2% versus 2.8% for Google organic, a 5x advantage for brands earning AI citations.
First Page Sage also reports AI-driven search traffic grew 7x in one year, and 38% of B2B decision-makers trust AI platforms for vendor shortlisting.
M&C Saatchi Performance quotes Google's AI Optimization Guide: tactics such as creating llms.txt files or aggressive content chunking do not improve visibility in AI-powered search experiences.
arXiv finds 32% of citation variation is explained by the engine's retrieval process versus only 2.8% by page properties, emphasizing that every page an engine cites is one its own search retrieved.
arXiv adds that retrieval mechanisms of generative search engines remain opaque, creating opportunities for AI-generated sources with misleading information to compete with accurate ones.
TryPrefer emphasizes that for ChatGPT, two runs of the same question shared, on average, only 37% of the sites cited across both, recommending a minimum of three runs per question per engine for reliable data.
First Page Sage notes Google AI Overviews appeared for 75 of 80 questions (93.8%) in study, and Google AI Mode returned answers for 76 of 80 questions (95%).
Your Site Is New, Your Citations Don't Have to Wait
The playbook for AI citation building has fundamentally changed: 85% of brand mentions in AI answers come from third-party domains, brand mentions correlate three times more strongly with AI visibility than backlinks, and 82% of citations stem from earned media. For a new website, that means the fastest path to being cited by ChatGPT, Gemini, and Perplexity runs through digital PR, engine-specific platforms like YouTube and Reddit, and multi-source consensus, not on-page tweaks. And the payoff is real: AI-referred visitors convert at 14.2% versus 2.8% for Google organic, a five-fold advantage that makes citations a revenue channel, not a vanity metric. Your next steps are straightforward: start pitching newsworthy data to journalists, build active profiles on the platforms your target engine cites most, keep content fresh within a two-month window, and track each engine separately with at least three runs per question. At SentFirst, we help new brands get discovered faster, and if you want a clear picture of where you stand today, book a free AI visibility audit to see how your new site can start earning AI citations.