Everything PR News
Research Report/Beauty

Ingredient Transparency vs Heritage Branding: Which Wins AI Citation Share

Two skincare communication models. One built on ingredient specificity, the other on brand heritage. The AI answer engine picks a winner — and it isn't the one with the longer history.

EPR Editorial TeamEPR Editorial Team 5 min read

Two skincare brands launch the same hero product on the same week. Same active. Same concentration. Same price point.

Brand A is fifty years old. Its product page leads with the maison's history, the founder's vision, the laboratory's heritage. The ingredient appears halfway down, in marketing-grade prose.

Brand B is six years old. Its product page leads with the ingredient name, the concentration, the pH, the suggested use, and a list of what the formula does not contain. The brand story is two sentences at the bottom.

Ask ChatGPT which one to use. It picks Brand B. So does Claude. So does Perplexity. So does Gemini. The pattern repeats across the entire category.

This is the structural shift that has redrawn the skincare competitive map. The two communication models — heritage and ingredient transparency — are no longer equal-and-different. They produce measurably different outcomes inside the AI answer surface.

For the ranked map of which model is winning across beauty, see The Beauty Citation Share Index 2026: Who Owns the AI Answer.

What heritage branding gives an answer engine.

Almost nothing.

Heritage branding is built on emotional association — feelings about a brand, scenes from a campaign, the gravity of a name. None of that is extractable into an answer to the question "what should I use for melasma."

An answer engine reading a heritage-led product page finds atmospheric copy, founder lore, and references to laboratories and Parisian apothecaries. It does not find: ingredient name, concentration, pH, claim substantiation, suggested use, contraindications, comparison data. What it cannot extract, it cannot cite.

The historical assumption — that brand equity built across decades would carry into any new discovery surface — has failed at the AI layer. Equity that lives in feeling does not translate into equity that lives in retrieval. La Mer still sells. It does not get cited.

What ingredient transparency gives an answer engine.

The answer.

Ingredient-transparent brands publish what an engine needs to lift: the active, the percentage, the formulation logic, the use case, the exclusions, the comparisons. The product page reads like the answer to a question a buyer would ask.

This is why brands like The Ordinary, Paula's Choice, Drunk Elephant, CeraVe, and La Roche-Posay dominate skincare retrieval. The brands' product pages do half the engine's work for it.

Paula's Choice took the model furthest: it published an Ingredient Dictionary that is now referenced by retrieval engines as a structured source, not just a brand property. The brand became infrastructure. Heritage brands cannot replicate this by adding an ingredient page; they have to rebuild how the entire brand communicates.

Where heritage still wins.

Fragrance.

The fragrance category is the structural exception. Smell does not extract into ingredient transparency the way a serum's actives do. The buyer is not asking "what percentage of bergamot is in this." They are asking "what does a Tom Ford fragrance smell like compared to a Chanel." Both questions resolve through editorial association, not chemical specification.

Dior, Chanel, Tom Ford, Maison Francis Kurkdjian, Le Labo — all dominate fragrance AI retrieval through heritage and editorial co-mention. The category rewards story because the product itself is irreducibly subjective.

This is the defensive line legacy luxury houses should hold. Fragrance is a heritage stronghold. Skincare is not.

The middle path doesn't work.

Several heritage brands have attempted hybrid positioning — keeping the maison narrative on the homepage, layering ingredient transparency on the product page. The result tends to be neither.

Answer engines see inconsistent entity signals: clinical claim on one page, atmospheric brand language on another, conflicting tone across the retailer site, the brand site, and editorial coverage. Inconsistency makes resolution harder, and unresolved entities get dropped from answers rather than risked.

The brands that have moved successfully — SkinCeuticals being the cleanest example — committed fully. The brand voice is clinical. The product pages are technical. The earned coverage at Allure and Byrdie emphasizes the research, not the prestige. Consistent signal across every surface.

Half-measures produce half-citations.

Where this leaves legacy prestige skincare.

Exposed.

The brands most vulnerable are the ones whose entire equity sits in heritage and lifestyle association — and whose product pages still read like ad copy. Without a structural communication shift, these brands will continue losing skincare citation share to brands that publish what the engines need.

The strategic options are narrow. Defend through fragrance, where heritage still wins. Commit to ingredient transparency as a brand voice rebuild, which most legacy houses will not do because it threatens the maison narrative they have spent decades constructing. Or accept that heritage equity will continue compounding only in the dimensions answer engines do not weight — and lose the skincare buyer entirely to brands that speak the new language.

The full discipline framework lives in Beauty AI Communications: The Complete 2026 Guide. The GEO mechanics behind ingredient extractability are unpacked in How Beauty Brands Win in the GEO Era. The entity-level case study is Why EltaMD Is the First Beauty Brand the Answer Engines Recommend.

Other research

See all

Most brands are invisible inside AI search. Is yours?

EPR publishes the data every week.

Free. Weekly. Unsubscribe anytime.