82.3% of external citations in SaaS: on-page SEO alone is not enough

Summarize this article with AI

In short: 82.3% of AI Search citations in SaaS come from external sources. Aleyda Solis’s data shows it clearly: on-page SEO matters, but it’s not enough. Without social visibility, media relationships, or positioning against competitors, your brand disappears. Here’s how to reorganize your content and budgets.
82.3%of AI citations in SaaS are external
30.4%share of owned citations in e-commerce, the highest
45.7%of SaaS citations come from social networks

Just 17.7% in SaaS. The verdict is clear.

A client calls me on a Tuesday morning. He runs a SaaS platform, 320 pages of content, clean semantic silos. He’s invested €6,000 in copywriting and internal linking. Tracking works. Yet when you search his category in ChatGPT, Gemini, or Google AI Mode, his brand doesn’t show up. No mentions, no panels.

I ask him the question nobody asks. When was your last article in tech press? Your last YouTube video with a sector influencer? Your last Reddit thread with real conversation?

Silence. Nothing.

The problem wasn’t his content. It was the absence of external citations. And the figures published by Aleyda Solis in August 2026 show this isn’t isolated. It’s systemic.

She analyzed the top 10 cited domains for 15 leading brands in SaaS, e-commerce, and finance across Google AI Mode, Gemini, and ChatGPT, using Semrush Enterprise data. Each source was classified into four catégories: owned (domains controlled by the brand), social/community (YouTube, Reddit, LinkedIn, etc.), news/review (TechRadar, Forbes, Trustpilot, G2
), and competitor (competitor domains or marketplaces).

The result?

SectorOwned CitationsExternal CitationsSocialNews/ReviewCompetitorsOther Third-party
SaaS17.7%82.3%45.7%10.4%8.1%18.1%
E-commerce30.4%69.6%27.5%0.7%36.6%4.8%
Finance20.6%79.4%29.8%19.4%19.8%10.4%

Your site, even ultra-optimized, accounts for only 17 to 30% of citations. The rest comes from outside. For SaaS, more than 8 out of 10 placements are external. E-commerce, which retains the most control, barely tops 30%.

AI Search is a question of external corroboration. On-page SEO isn’t dead. It’s the foundation. But without external presence, your competitors occupy the panels.

Les chiffres de Aleyda Solis révÚlent des stratégies radicalement différentes selon le secteur. Voici comment les citations externes se distribuent : social et communautés tech dominent en SaaS, tandis que les concurrents rÚgnent en e-commerce.

Citation sources par secteur

RĂ©partition des sources d’citations externes en AI Search : SaaS vs e-commerce vs finance

SaaS social, e-commerce competitor, finance mixed: the detail that changes everything

Global averages hide a reality that shifts sector by sector. When you dive into Aleyda’s numbers, the profiles diverge completely.

SaaS: social dominates. 45.7% of total citations come from YouTube, Reddit, LinkedIn, Quora. Tech communities, video comparisons, developer threads. Traditional media (news/review) represents only 10.4%. Competitors, 8.1%. It’s clear.

E-commerce: the battle between brands. E-commerce brands cite each other. 36.6% of AI citations come from competitor domains or marketplaces (Walmart, Target, etc.). Social weighs 27.5%, press almost nothing (0.7%). For an e-commerce brand, being cited in a competitor’s product card on ChatGPT is an opportunity—or a risk.

Finance: fragile balance. Most varied sector: 29.8% social, 19.4% news/review, 19.8% competitors. Personal finance forums, Reddit, YouTube, banking comparators, business press. Your brand must exist everywhere, a bit here and there, but consistently.

Remember this: your strategy changes by sector. A SaaS betting everything on press relations is on the wrong path. An e-commerce absent from competitor marketplaces leaves 36% of the AI target on the table. Aleyda’s precise figures give you the map.

Her analysis also shows weights vary by AI platform (Google AI Mode vs Gemini vs ChatGPT). But the dominant pattern stays stable per sector.

Un phénomÚne clé émerge : bien que vos pages propres occupent seulement 10-15% des emplacements sources, leur poids dans le volume total de mentions est presque deux fois supérieur. Voici le déséquilibre révélateur.

Le paradoxe du domaine propre

Slots occupés vs. poids réel des mentions : votre site fait peu mais pÚse lourd

Trafic IA Trafic classique

Few slots, heavy weight: the owned site paradox

Another finding from the study. When you look at placements in the top 10 sources, owned domains of the brand represent only 10 to 11% of slots on average across the three sectors. In finance, 10% of source slots. In SaaS, 11%. In e-commerce, slightly better but in the same ballpark, around 15% maybe (the report doesn’t specify but the logic follows).

Yet when these owned domains appear, their weight in mentions is disproportionate. In finance, 10% of source slots generate 20.6% of total mention weight. That’s nearly double. Same phenomenon in SaaS and e-commerce: fewer appearances, but when the brand’s site is cited, it’s central.

This is the concept of « mention-dense but slot-scarce » coined by Aleyda Solis. Your site won’t often be cited first. But when it is, the AI relies heavily on it.

This is why on-page SEO truly matters. It forms the corroboration foundation: your product pages, pricing cards, technical documentation, privacy policies, guides. The AI uses these to verify facts external sources report. If your base is fuzzy, outdated, or contradictory, the AI excludes you.

I tell my B2B clients often: your semantic silo, your pillar pages—that’s your certification file. External citations are testimonies. Without the file, testimonies are worthless. Without testimonies, the file stays in a drawer.

Coordinate SEO, PR, and social: it’s no longer optional, it’s a framework

Aleyda doesn’t just give numbers. She lays the foundation for an operational framework: SEO + digital PR + social media, piloted together, with clear roles.

Her message: stop treating AI visibility as a promotional layer added after on-page optimizations. Integrate it from the start of your content strategy design.

Here are the 5 axes I draw from it, tested on my own deployments:

  1. Treat external citations as a priority KPI. Not a bonus. An AI ranking indicator. Measure your appearance rate in AI panels, by source, by prompt type. If you don’t measure it, you don’t exist.
  2. Choose your Digital PR or social targets based on your sector. The numbers in the table tell you where to invest: YouTube demos, Reddit threads, and G2 comparisons for SaaS; presence on competitor marketplaces and verified reviews for e-commerce; business press and forums for finance. Don’t aim at everything. Aim precisely.
  3. Maintain your site as the canonical source. Current product cards, exact prices, error-free technical documentation. External sources pull their data from here. One inconsistency on a pricing page, and the whole chain breaks.
  4. Don’t confuse source share with mention share. A drop in your owned source appearances isn’t necessarily bad if your external mentions climb. Track both separately.
  5. Create a corroboration loop. Make sure each social post, each PR article, each review points back to your site for a specific fact. No forced links, just citation of a claim your page validates. The AI detects this coherence.

Aleyda even suggests a minimum workflow: SEO identifiés key entities and attributes for target prompts, PR builds angles for media, social generates shared expérience. All in one dashboard.

Ce client fintech illustre la transformation : en stoppant la production de contenu classique et en redéployant 70% du budget vers relations media et community, les mentions AI Search ont explosé. Voici la dynamique des gains.

Du cas client fintech : avant et aprĂšs

De 1 à 47 mentions AI en 6 mois : décomposition des levier

A finance client went from 1 to 47 AI mentions in 6 months

I observed this mechanism with a finance client, a fintech brokerage. Their site had 140 well-architected pages. The semantic silo ran across 3 pillars: credit, savings, insurance. On-page SEO was solid. Organic traffic decent. But in AI Search? One single mention, on a very specific prompt.

We analyzed the sources appearing for their market prompts. Verdict: 60% of citations came from forums (Reddit, investor communities) and comparators. Their brand was absent from these spaces.

We stopped producing new blog articles. We reallocated 70% of content budget to:

We also updated 22 site pages to ensure data consistency (interest rates, comparisons). The corroboration base was solid.

Result in 6 months: 47 unique mentions in AI panels (ChatGPT, Gemini). No link buying. No magic. Just alignment between an impeccable on-page foundation and externally relevant sector presence.

Their total organic sessions climbed 12%, but the real signal was the number of inbound leads attributable to AI search. 83 form submissions in 6 months. Profitable by month 4.

Stop writing. Start being cited.

Aleyda and field feedback confirm it: SEO investment is shifting direction.

The error we see everywhere: produce ever more on-page content thinking you’ll become a trusted source for AI. You strengthen your base, that’s useful. But without external links, you stay capped at 17.7% for SaaS or 20.6% for finance.

The priority moves. It’s no longer about publishing one more article, but getting a media to reference your card, an influencer to cite your solution, a Reddit thread to mention your name.

The working strategy in 2 steps:

For an e-commerce merchant, that means being listed on marketplace competitor cards and getting structured Trustpilot reviews. For a SaaS, producing YouTube demo videos with B2B creators. For a fintech, being cited in a Les Echos article or a personal finance forum.

Aleyda’s data spares you guessing. You know where to invest effort.

One indicator to track in 2026

The SEO person watching only Google rankings is in for a shock in 2026. AI Search isn’t theory anymore. Gemini, ChatGPT, Google AI Mode are already in use. The rules exist.

Your brand controls less than one slot in five. The rest is your ability to exist elsewhere. Good news: the map is public. Aleyda Solis’s ratios tell you how to distribute your future effort.

I’m not selling you a miracle method. I’m showing you the panels, showing you the numbers. One question alone.

When was your brand last cited without you paying?

AI Search citation audit: is your brand invisible on ChatGPT and Gemini?

I scan your external sources, your on-page corroboration, and give you the priority PR and social angles by sector. No generic recommendations: a 6-month roadmap.

Book a strategic call — 45 min

Frequently Asked Questions

How do I measure my citations in AI Search?

Use a monitoring tool to spot where your brand appears in AI panels—Brand24 or Semrush Media Monitoring do the job. Build a list of prompts matching your activity, then track how your brand ranks month over month.

Should I prioritize media relations or social networks for AI Search?

<p>It depends on your sector. In SaaS, social networks (YouTube, Reddit) lead with 45.7%. In e-commerce, competitors and marketplaces account for 36.6%. In finance, you need to combine social, press, and comparators. Aleyda Solis’s sector data tells you where to focus your budget.</p>

Does on-page SEO become secondary?

No, it becomes the verification foundation. AI uses your pages to check information provided by other sources. A reliable, current on-page base is what makes external citations credible.

How many external citations do I need to appear in AI results?

<p>There’s no magic number. What matters is having varied, relevant citations on platforms where your audience lives. Even a few quality citations (sector press, influential YouTubers, active forums) work if you’ve nailed your on-page fundamentals.</p>

Does this approach work for small sites, not just big brands?

<p>Yes. Aleyda’s data covers 15 leading brands, but the mechanism works for everyone. A small player can become the reference in a specific niche by getting citations from niche communities and perfecting on-page documentation, without facing giants head-on.</p>

Stéphane Jambu

Stéphane Jambu

SEO & AI Engineer

I build growth systems / AI / Neuroscience | 650+ clients · 80 LinkedIn testimonials · 30 years of expertise · 15 years of systems running without me.

Follow on LinkedIn