memecosmetics.fr — AI Site Grade

Analysis

MÊME Cosmetics: a cancer-care skincare brand that AI models misidentify as a general "clean makeup" brand

The site's core identity — the first dermo-cosmetic brand designed for people undergoing cancer treatment — is almost entirely invisible to AI crawlers because every page on the PrestaShop e-commerce platform renders as a ~16-word JavaScript shell with no substantive body text.

Crawler Access

All major AI bots (GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, Bytespider, Applebot-Extended) receive HTTP 200 with full byte payload from Cloudflare. No bot is blocked at the edge or in robots.txt. However, the robots.txt contains zero AI-specific directives — no User-agent: GPTBot or Disallow for any AI crawler — and the llms.txt returns a 404 (the PrestaShop 404 page). The sitemap listed in robots.txt (/1_index_sitemap.xml) redirects to the homepage rather than serving XML, effectively making it inaccessible to crawlers. The blog has its own functional sitemap (/blog/sitemap_index.xml) with 268+ URLs, but the main product catalog has no discoverable sitemap.

Content & Schema

The homepage, product pages, and content pages all return ~16 words of visible text from a plain GET — only product names and prices survive the JavaScript rendering. The blog (WordPress-based at /blog/) is the exception, serving 856+ words of rich text about cancer care, femininity, and treatment side effects. Product pages carry solid Product schema with gtin13, sku, brand, offers, and availability — but the product descriptions themselves are JS-rendered and invisible. The homepage has Organization, WebSite, and WebPage schema but no MedicalBusiness, HealthAndBeautyBusiness, or FAQPage types that would signal the oncology-care positioning to knowledge graphs. The product page for "Crème pour le visage" has an H2 titled "Questions fréquentes" with two Q&A items, but these are not marked up as FAQPage schema.

Cold-Knowledge Gap

A frontier LLM queried cold describes MÊME as a "French makeup brand founded in 2018 by Camille Masson and Marie-Christine Levet" focusing on "gender-neutral, vegan, and cruelty-free products" with a "customizable eyeshadow palette." This is almost entirely wrong. The actual brand was co-founded by Judith and Juliette (per the blog's "MÊME Story" article), not Camille Masson and Marie-Christine Levet. The brand's entire raison d'être is dermo-cosmetics for cancer patients — not general clean makeup. The model knows nothing about the oncology positioning, the "chimio" product line, the clinical studies, or the distribution through pharmacies. The gap between what AI models believe and what the site actually sells is the widest observed in this audit: the brand's core value proposition is completely absent from LLM prior knowledge.

External Signals

The site links to avis-verifies.com for customer reviews, but web searches returned no indexed external press, reviews, or Reddit threads about the brand in English or French. The blog has genuine community engagement (patient comments on the MÊME Story post), but this content lives on a subdomain (/blog/) with a separate WordPress sitemap and is not cross-linked from the main product catalog in a way that AI crawlers can easily associate. The DNS shows Google Workspace mail, Mailjet, SendGrid, and Zendesk — a mature tech stack — but the PrestaShop JS-rendering problem means AI crawlers see empty shells where they should find the brand's compelling oncology-care narrative.

Findings

1. All PrestaShop pages render as ~16-word JavaScript shells High

   Every page on the PrestaShop platform (homepage, product pages, content pages) returns only ~16 words of visible text from a plain GET. Product names and prices survive, but body content is JS-rendered and invisible to AI crawlers.

   What to change: Implement server-side rendering (SSR) or static generation for PrestaShop pages so that AI crawlers receive full HTML content. Alternatively, use a headless CMS with SSR for product descriptions and brand narrative.

2. Main sitemap redirects to homepage instead of serving XML High

   The sitemap listed in robots.txt (`/1_index_sitemap.xml`) redirects to the homepage rather than returning an XML sitemap, making the product catalog undiscoverable by crawlers.

   What to change: Fix the sitemap generation in PrestaShop to serve a valid XML sitemap at `/1_index_sitemap.xml` covering all product and category URLs.

3. llms.txt returns 404 Medium

   The site does not provide an `llms.txt` file, missing an opportunity to guide AI crawlers to key content like the brand story and product lines.

   What to change: Create an `llms.txt` file that points AI crawlers to the blog's 'MÊME Story' article, the ethical charter, and key product pages.

4. Robots.txt has no AI-specific directives Low

   The robots.txt file contains no `User-agent` lines for GPTBot, ClaudeBot, or other AI crawlers, leaving their access unmanaged.

   What to change: Add explicit `User-agent` and `Disallow`/`Allow` rules for AI crawlers to control access and signal important pages.

5. LLMs misidentify MÊME as a general clean makeup brand High

   A frontier LLM queried cold describes MÊME as a 'French makeup brand founded in 2018 by Camille Masson and Marie-Christine Levet' focusing on gender-neutral, vegan products with a customizable eyeshadow palette. The actual brand is dermo-cosmetics for cancer patients, co-founded by Judith and Juliette. The oncology positioning is completely absent from LLM prior knowledge.

   What to change: Ensure the brand's oncology-care narrative is present in crawlable HTML on key pages (homepage, about, product pages) and reinforce with structured data (MedicalBusiness, FAQPage).

6. No MedicalBusiness or HealthAndBeautyBusiness schema on homepage Medium

   The homepage has Organization, WebSite, and WebPage schema but lacks MedicalBusiness or HealthAndBeautyBusiness types that would signal the oncology-care positioning to knowledge graphs.

   What to change: Add `MedicalBusiness` or `HealthAndBeautyBusiness` schema to the homepage and about page to explicitly signal the healthcare focus.

7. FAQ content on product pages lacks FAQPage schema Medium

   The product page for 'Crème pour le visage' has an H2 'Questions fréquentes' with two Q&A items, but these are not marked up as FAQPage schema, missing an opportunity for rich results.

   What to change: Add `FAQPage` schema markup to the FAQ section on product pages.

8. No indexed external press, reviews, or social mentions High

   Web searches returned zero results for the brand in English or French, no external press articles, no Reddit threads, and no indexed reviews on avis-verifies.com. The brand has no external signals to reinforce its authority.

   What to change: Build external signals through PR, guest posts on oncology blogs, and encourage reviews on platforms like avis-verifies.com. Ensure reviews are indexed.

9. Blog content not cross-linked from main product catalog Medium

   The blog (WordPress at `/blog/`) contains rich content about cancer care but is not cross-linked from the main PrestaShop product catalog, making it hard for crawlers to associate the two.

   What to change: Add contextual links from product pages and the homepage to relevant blog articles (e.g., 'MÊME Story' and cancer care tips).

10. Product descriptions invisible to crawlers due to JS rendering High

    Product pages carry solid Product schema with gtin13, sku, brand, offers, and availability, but the product descriptions themselves are JS-rendered and invisible. AI crawlers see only product names and prices.

    What to change: Ensure product descriptions are included in the initial HTML payload via server-side rendering or static generation.

What's working

• All major AI bots receive HTTP 200 with full payload — No AI bot is blocked at the edge or in robots.txt. GPTBot, ClaudeBot, PerplexityBot, and others all receive HTTP 200 with full byte payload from Cloudflare.
• Blog serves rich, crawlable content about cancer care — The WordPress blog at `/blog/` returns 856+ words of rich text about cancer care, femininity, and treatment side effects, providing valuable context for AI crawlers.
• Blog has a functional sitemap with 268+ URLs — The blog sitemap at `/blog/sitemap_index.xml` returns 80+ URLs (268+ total) and is properly indexed, ensuring blog content is discoverable.
• Product pages include solid Product schema with GTIN and brand — Product pages carry Product schema with gtin13, sku, brand, offers, and availability, providing structured data for search engines.
• Mature tech stack with Google Workspace, Mailjet, SendGrid, Zendesk — DNS records show Google Workspace mail, Mailjet, SendGrid, and Zendesk, indicating a professional infrastructure for email and customer support.
• Ethical charter page exists with formulation and trust commitments — The site has a dedicated 'Notre Charte Éthique' page outlining formulation and trust commitments, which can support brand authority.