{"slug": "kantar-identifies-brands-large-language-model-visibility-gap", "title": "Kantar Identifies Brands' Large-Language-Model Visibility Gap", "summary": "Kantar executive director Karin Du Chenne told a Mumbrella360 audience that some of the world's leading brands are effectively invisible in large-language-model search results because they have not produced content optimized for machine readability. Du Chenne said LLMs assemble answers from a \"patchwork\" of citations and signals such as reviews and ratings, and argued that brands consistently known for a single, distinctive attribute are more likely to surface in LLM responses. The finding highlights a growing gap between traditional brand-building channels and the signal types that generative search systems consume.", "body_md": "# Kantar Identifies Brands' Large-Language-Model Visibility Gap\n\nAccording to Mumbrella reporting, Kantar executive director Karin Du Chenne said some of the world's leading brands are effectively invisible in large-language-model (LLM) search results because they have not produced content optimized for machine readability. Du Chenne told a Mumbrella360 audience, \"We're no longer just marketing to humans, we're marketing to machines as well,\" and said LLMs assemble answers from a \"patchwork\" of citations and signals such as reviews and ratings, per Mumbrella. She argued that brands that are consistently known for a single, distinctive attribute are more likely to surface in LLM responses, while brands that rely primarily on emotive mass-media advertising can be underrepresented in model outputs, according to Mumbrella.\n\n### What happened\n\nAccording to Mumbrella reporting, Kantar executive director **Karin Du Chenne** told a Mumbrella360 audience that some leading brands are effectively invisible in LLM search results because they have not built content for machine readability. \"We're no longer just marketing to humans, we're marketing to machines as well,\" Du Chenne said, per Mumbrella. The article reports Du Chenne described LLM answers as drawn from a \"patchwork\" of citations and signposts, including reviews and ratings, and said brands that are consistently associated with a single, distinctive attribute are more likely to \"climb to the top,\" per Mumbrella.\n\n### Editorial analysis - technical context\n\nKantar's description highlights how LLM-based retrieval and answer-generation rely on distributed signals rather than a single canonical source. Industry-pattern observations show modern generative search systems weight frequency, consistency, and corroborating citations when constructing responses. For practitioners, that shifts some SEO concerns from purely human-facing copy and backlinking to machine-friendly structure and repeatable, attributable claims across web properties.\n\n### Industry context\n\nReporting frames this as a broader gap between traditional brand-building channels, such as emotive television advertising, and the signal types LLM systems consume. Industry observers note that brands built primarily through offline or emotionally resonant channels can underperform in algorithmic discoverability unless their web presence encodes consistent, machine-readable signals. This is consistent with recent conversations in marketing technology about structured data, schema, and reputation signals becoming inputs to model-driven interfaces.\n\n### What to watch\n\n- •Volume and consistency of third-party citations and reviews for brand queries\n- •Use of structured metadata and schema across brand-owned properties\n- •Emergence of vendor tools that translate marketing content into machine-readable signals\n\n### Industry context\n\nThese indicators will help observers determine whether brands are adapting their content and measurement practices to the realities of LLM retrieval. The article does not include documentation of specific industry-wide adoption metrics or vendor rollouts, and Kantar has not provided technical specifications in the Mumbrella coverage.\n\n## Scoring Rationale\n\nThe report flags a practical operational gap that affects marketing, search, and data teams implementing `LLM`-driven discovery, but it does not introduce new models or technical methods. It is notable for practitioners responsible for retrieval, metadata, and measurement.\n\nPractice interview problems based on real data\n\n1,500+ SQL & Python problems across 15 industry datasets — the exact type of data you work with.\n\n[Try 250 free problems](/problems)", "url": "https://wpnews.pro/news/kantar-identifies-brands-large-language-model-visibility-gap", "canonical_source": "https://letsdatascience.com/news/kantar-identifies-brands-large-language-model-visibility-gap-5903f68d", "published_at": "2026-05-28 01:31:28.768699+00:00", "updated_at": "2026-05-28 01:31:31.723131+00:00", "lang": "en", "topics": ["large-language-models", "artificial-intelligence", "natural-language-processing"], "entities": ["Kantar", "Karin Du Chenne", "Mumbrella"], "alternates": {"html": "https://wpnews.pro/news/kantar-identifies-brands-large-language-model-visibility-gap", "markdown": "https://wpnews.pro/news/kantar-identifies-brands-large-language-model-visibility-gap.md", "text": "https://wpnews.pro/news/kantar-identifies-brands-large-language-model-visibility-gap.txt", "jsonld": "https://wpnews.pro/news/kantar-identifies-brands-large-language-model-visibility-gap.jsonld"}}