Marketing covers agencies, consultants and freelancers in SEO, advertising, social media, branding and public relations, along with printers and promotional product suppliers. AI answers about marketing cite software publishers and their guides, agency directories, government pages for small businesses, professional networks and review platforms, mixing advice on how to do the work with suggestions of who can do it.
Key takeaways
- Closest sector by cited sources: IT Services (5 of the 20 most cited domains in common).
Summarize this page with ChatGPTClaudePerplexity
What changed in September 2026
- Answers analyzed 7'630 (▲ 21% vs August 2026)
- First domain semrush.com (3.74 per 100 answers)
- First platform LinkedIn (1.63 per 100 answers)
Months compared: September 2026 and August 2026. The platforms that rose and fell the most are listed just below.
Biggest movers in September 2026
Most cited domains in AI answers, Marketing
5 of the 10 most cited domains in Marketing (semrush.com, ahrefs.com and hubspot.fr) do not appear among the 50 most cited domains of the whole panel.
How to read it: each bar is a domain; its length is its number of citations per 100 answers over the selected period and filters.
Domains ranked by citations per 100 answers, all periods combined.
| Rank | Rank change | Domain | Citations | per 100 answers | Share of citations | 12 months |
|---|---|---|---|---|---|---|
| 1 | Unchanged |
| 1'192 | 6.55 | 1.67% | Trend: 11.30 in Jul 2026, 3.74 in Sep 2026, high 11.30. |
| 2 | Down 6 |  ahrefs.com | 566 | 3.11 | 0.79% | Trend: 5.87 in Jul 2026, 1.53 in Sep 2026, high 5.87. |
| 3 | Down 2 |  francenum.gouv.fr | 501 | 2.75 | 0.70% | Trend: 3.33 in Jul 2026, 2.01 in Sep 2026, high 3.33. |
| 4 | Up 3 |  economie.gouv.fr | 418 | 2.30 | 0.59% | Trend: 2.81 in Jul 2026, 2.04 in Sep 2026, high 2.81. |
| 5 | Down 6 |  hubspot.fr | 397 | 2.18 | 0.56% | Trend: 4.01 in Jul 2026, 1.30 in Sep 2026, high 4.01. |
| 6 | Down 2 |  sortlist.fr | 388 | 2.13 | 0.54% | Trend: 3.75 in Jul 2026, 1.35 in Sep 2026, high 3.75. |
| 7 | Up 8 |  LinkedIn linkedin.com | 324 | 1.78 | 0.45% | Trend: 2.55 in Jul 2026, 1.64 in Sep 2026, high 2.55. |
| 8 | Up 12 |  Trustpilot trustpilot.com | 312 | 1.72 | 0.44% | Trend: 2.50 in Jul 2026, 1.61 in Sep 2026, high 2.50. |
| 9 | out of top |  HubSpot hubspot.com | 289 | 1.59 | 0.41% | Trend: 3.66 in Jul 2026, 2.12 in Aug 2026, high 3.66. |
| 10 | Up 4 |  exaprint.fr | 286 | 1.57 | 0.40% | Trend: 2.06 in Aug 2026, 2.04 in Sep 2026, high 2.06. |
| 11 | Down 1 |  scratcher.fr | 279 | 1.53 | 0.39% | Trend: 2.29 in Aug 2026, 1.76 in Sep 2026, high 2.29. |
| 12 | Up 6 |  cnil.fr | 268 | 1.47 | 0.38% | Trend: 2.24 in Jul 2026, 1.19 in Sep 2026, high 2.24. |
| 13 | Up 3 |  livestorm.co | 261 | 1.43 | 0.37% | Trend: 2.03 in Aug 2026, 1.74 in Sep 2026, high 2.03. |
| 14 | Up 6 |  eskimoz.fr | 249 | 1.37 | 0.35% | Trend: 1.67 in Jul 2026, 1.25 in Sep 2026, high 1.67. |
| 15 | Up 3 |  developer.mozilla.org | 221 | 1.21 | 0.31% | Trend: 1.65 in Aug 2026, 1.53 in Sep 2026, high 1.65. |
| 16 | out of top |  lafabriquedunet.fr | 213 | 1.17 | 0.30% | Trend: 2.43 in Jul 2026, 1.74 in Aug 2026, high 2.43. |
| 17 | new |  objetrama.fr | 193 | 1.06 | 0.27% | Trend: 2.45 in Jul 2026, 1.17 in Sep 2026, high 2.45. |
| 18 | out of top |  Wikipedia wikipedia.org | 184 | 1.01 | 0.26% | Trend: 1.84 in Jul 2026, 1.68 in Aug 2026, high 1.84. |
| 19 | Down 7 |  malt.fr | 180 | 0.99 | 0.25% | Trend: 1.53 in Aug 2026, 1.09 in Sep 2026, high 1.53. |
| 20 | Up 2 |  noiise.com | 178 | 0.98 | 0.25% | Trend: 1.38 in Aug 2026, 1.19 in Sep 2026, high 1.38. |
semrush.comRankings are rebuilt from monthly rankings: the counts at the end of a list can be minimums. Industry: the business of the analyzed website, classified by Sorank. Data since July 2026. Rank change: rank in the September 2026 monthly list against August 2026 (new: absent from the August 2026 list). A dash: fewer than 1'000 answers in one of the two months in this scope, or absent from the September 2026 list. 12 months: monthly rate per 100 answers; gaps when the domain is out of the monthly list, hollow points for months under 1'000 answers.
Most cited platforms in AI answers, Marketing
LinkedIn is the platform AI cites most in Marketing: 1.64 citations per 100 answers, 2.2 times its rate across all sectors over the same months (0.75).
How to read it: each bar is one of the tracked platforms; its length is its number of citations per 100 answers over the selected period.
| Rank | Rank change | Platform | per 100 answers | 12 months |
|---|---|---|---|---|
| 1 | Up 2 | LinkedIn | 1.64 | Trend: 2.12 in Jul 2026, 1.63 in Sep 2026, high 2.12. |
| 2 | Up 2 | Trustpilot | 1.62 | Trend: 2.41 in Jul 2026, 1.47 in Sep 2026, high 2.41. |
| 3 | Down 2 | HubSpot | 1.58 | Trend: 2.64 in Jul 2026, 0.92 in Sep 2026, high 2.64. |
| 4 | Down 9 | Wikipedia | 1.15 | Trend: 1.82 in Jul 2026, 0.37 in Sep 2026, high 1.82. |
| 5 | Up 3 |  Facebook | 1.04 | Trend: 1.58 in Jul 2026, 0.80 in Sep 2026, high 1.58. |
| 6 | Unchanged |  Microsoft | 0.94 | Trend: 1.27 in Jul 2026, 0.69 in Sep 2026, high 1.27. |
| 7 | Down 1 |  Salesforce | 0.71 | Trend: 1.72 in Jul 2026, 0.34 in Sep 2026, high 1.72. |
| 8 | Unchanged |  Zoom | 0.68 | Trend: 0.26 in Jul 2026, 0.64 in Sep 2026, high 1.01. |
| 9 | Up 2 |  Shopify | 0.62 | Trend: 0.85 in Jul 2026, 0.48 in Sep 2026, high 0.85. |
| 10 | Up 2 |  WordPress | 0.53 | Trend: 0.75 in Jul 2026, 0.41 in Sep 2026, high 0.75. |
Rank change: rank among the platforms cited in September 2026 against August 2026 (new: not cited in August 2026). A dash: fewer than 1'000 answers in one of the two months in this scope, or not cited in September 2026. 12 months: monthly rate per 100 answers, hollow points for months under 1'000 answers.
AI models compared
AI assistants do not agree on the first source in Marketing: ChatGPT cites economie.gouv.fr first; Gemini cites semrush.com first; Perplexity cites LinkedIn first.
How to read it: left, how the citations split between the five models; right, each model's citations per 100 answers, month by month.
First source of each AI model in Marketing
| Model | First source | per 100 answers | Second source |
|---|---|---|---|
 ChatGPT | economie.gouv.fr | 6.31 |
|
 Gemini |
| 15.64 | ahrefs.com |
 Perplexity | LinkedIn linkedin.com | 1.85 | lafabriquedunet.fr |
 Claude |  laboiteaobjets.com | 8.44 | objetrama.fr |
 Grok |  atelierbox.fr | 22.29 |  goodact.fr |
Typical marketing questions sent to AI assistants
Marketing prompts fall into three groups, and each one calls for a different kind of proof.
- Doing it yourself: "how to improve my local SEO", "how often should a small business post on LinkedIn", "how to write a cold email that complies with GDPR".
- Finding someone to do it: "best SEO agency in Lille", "freelance community manager for a restaurant", "how much does a branding agency charge".
- Buying marketing supplies: "cheap flyer printing", "custom tote bags for a trade show", "promotional pens with logo and fast delivery".
In the first group, agencies compete for citations with software publishers that maintain large libraries of free guides. In the second, they compete with directories. The third behaves like e-commerce.
Who gets cited: tool publishers, directories and official guidance
Marketing software publishers are the most visible educators. Semrush, Ahrefs and HubSpot run extensive blogs and academies, so AI models cite them for definitions, tutorials and benchmarks, even when the user was looking for an agency.
Agency directories and freelance marketplaces such as Sortlist and Malt structure the "who can help me" answers, while LinkedIn and Trustpilot provide profiles and reviews that confirm an agency exists and is appreciated.
Public sources appear whenever a rule is involved: the CNIL for cookies, consent and email prospecting, government portals for the digital transition of small businesses. In English answers, technical web references such as MDN Web Docs and Cloudflare come up on questions about site speed, structured data or security, a sign of how technical many SEO prompts have become.
What this means for an agency or a freelancer
Agencies sell expertise, and an AI can only judge expertise from what is published. A firm that writes generic articles on "why SEO matters" competes with software publishers that have far more content and authority, and it usually loses. A firm that publishes original material (a case study with real figures, a documented method, a local market analysis) gives the AI something it cannot find elsewhere.
The same logic applies to provider searches. Answers to "SEO agency in Nantes" tend to cite directories and the agencies whose own site has a clear page for that city and service. Specialisation helps: an agency positioned on "SEO for e-commerce" or "paid social for B2B" matches precise prompts better than a generalist full-service agency.
GEO (Generative Engine Optimization, optimising content so that AI assistants cite it) is also becoming a service that agencies sell. An agency's own visibility in AI answers is then the most convincing demonstration it can offer.
Marketing agencies: actions to appear in AI answers
- Publish original data or methods on a regular basis (a client benchmark, a test, a local study) instead of generic advice already covered by software publishers.
- Create service and city pages only for combinations you truly serve, each with named references, price ranges and the team members involved.
- Complete your Sortlist, Malt, LinkedIn and Trustpilot profiles with the same positioning as your website, and request a review after each project.
- State your specialist positioning in the page title and first paragraph ("SEO agency for e-commerce brands") so that precise prompts match it.
- For printing and promotional products, describe formats, minimum quantities, delivery times and prices as text on each product page, so AI models can answer purchase questions directly.
- Measure your own presence in AI answers for your core services with an AI mention tracker such as Sorank, and turn the results into a case study for clients.
Related industries
Sectors sharing the most domains with Marketing among their 20 most cited, then sectors with a similar number of answers.
Frequently asked questions
Why does ChatGPT cite Semrush or HubSpot when I ask for a marketing agency?
Because those publishers own a large share of the educational content on marketing. When a prompt mixes advice and provider search, the AI often explains the method with software guides, then lists agencies taken from directories or from their own sites. Agencies with distinctive, well-documented expertise are more likely to be cited in the provider part of the answer.
How can an SEO agency check whether AI assistants recommend it?
Ask the prompts your prospects would use, in each AI assistant, at regular intervals: "best SEO agency in [city]", "SEO agency for [sector]". Note whether your agency is named, whether your site is cited and which competitors appear. Because answers vary from one run to the next, weekly tracking over several weeks gives a more reliable picture than a single test.
Do agency directories like Sortlist help with AI visibility?
Yes, for provider-search prompts. AI models that search the web retrieve directories because they group agencies by service and location, with reviews. A complete profile can bring your name into answers, and a consistent website lets the model confirm the details and sometimes cite your own domain directly.
Which websites do AI assistants cite for Marketing businesses?
semrush.com (6.55 citations per 100 answers), ahrefs.com (3.11) and francenum.gouv.fr (2.75), across 18'190 answers from ChatGPT, Claude, Gemini, Grok and Perplexity about Marketing businesses since July 2026.
Which platforms do AI assistants cite in Marketing?
LinkedIn (1.64 citations per 100 answers), Trustpilot (1.62) and HubSpot (1.58).
Method and sources
387'471 answers from ChatGPT, Claude, Gemini, Grok and Perplexity, from August 2025 to September 2026, to the questions Sorank asks every week to follow its clients' visibility (88% of them in French). A citation is a source domain given in an answer; the rate is the number of citations per 100 answers. Data snapshot of 2 October 2026.
Full methodology · Raw data (JSON)
How to cite this page: Sorank Data, “Websites AI cites most in Marketing (Sep 2026)”, data.sorank.com, data snapshot of 2 October 2026.
Track the citations of your own site with Sorank's AI mention tracking.
