Press in BBC, Daily Mail, Tatler. Invisible in AI answers.
A sustainable haircare brand with extensive UK press coverage. Award nominations. Strong Trustpilot reviews. When buyers ask AI for the best shampoo bars in the category, the brand is never named. Competitors with thinner press footprints occupy every recommendation slot.
The brand name collides with a high-frequency English word. AI reads it as a refusal, not a haircare brand.
This is the Polyglot Penalty: brand identity fragments because the brand name itself signals something else.
Layer 1: The Polyglot Penalty
The brand surface offers no disambiguation. The title tag is the brand name and nothing else. The og:title is the brand name and nothing else. The og:description is the brand name and nothing else. The entity surface carries zero category signal, zero industry context, and zero geographic marker.
When AI tries to resolve the brand name, it encounters a high-frequency English word that means refusal. With no category context on the entity surface to override that default reading, AI never connects the brand name to the haircare category.
The retrieval pipeline does not have to guess at the competitors. It does at this brand. When it guesses, it returns the dictionary word, not the product.
Layer 3: Semantic Drift in product taxonomy.
Buyer queries are symptom-led. "Shampoo bar for oily hair." "Shampoo bar for dandruff." The brand's product taxonomy uses creative naming. The translation between buyer language and brand language exists only in small annotation text on collection pages.
A competitor solved the same category problem by naming products around the symptom. Their dandruff shampoo bar is named, marketed, and structured as a dandruff shampoo bar. It is cited repeatedly for that query. This brand is not.
When buyers ask AI to recommend a shampoo bar in this category, the brand is not named. Buyers do not discover what AI cannot describe. The press coverage earned does not contribute to algorithmic authority because the citation infrastructure does not exist on the brand surface.
The Polyglot Penalty hits any brand whose name collides with a dictionary word. Common examples include single-syllable brand names, brand names that double as verbs or interjections, and brand names that share spelling with common terms in another language.
The fix is not a rebrand. The fix is loading the entity surface with enough category context that AI cannot read the name as anything else. Every meta surface, every structured data block, and every entity property has to carry the disambiguation that the name itself does not provide.
Does your brand name signal something else to AI?
The AI Visibility Snapshot runs the diagnostic. You receive a written finding within 48 hours: how AI is reading your brand name, what category signal your entity surface carries, and what is breaking first.