How AI Search Is Recommending Dermatologist-Recommended Skincare Brands

AI search is turning dermatologist-recommended skincare into a shortlist market. Buyers are no longer only searching for product pages, retailer listings, or review articles. They are asking AI systems which skincare brands to trust, which products work for acne or hyperpigmentation, which moisturizer to use, and which brands dermatologists would recommend.

The May 2026 LLM Authority Index dataset covers 614 AI observations across six AI surfaces and roughly 3.94 million modeled monthly query demand. Across that benchmark, AI recommendation power concentrates around a small group of brands: CeraVe, La Roche-Posay, Neutrogena, The Ordinary, SkinCeuticals, and Paula’s Choice. CeraVe and La Roche-Posay lead the market on valid recommendation coverage, while Paula’s Choice shows meaningful strength but is not yet consistently treated as a default category answer.

Key findings

1. CeraVe and La Roche-Posay are the default AI shortlist leaders. CeraVe earned 315 valid recommendations, a 28.83% top-three recommendation rate, and $382,255 in modeled monthly captured recommendation value. La Roche-Posay followed closely with 310 valid recommendations, a 27.36% top-three rate, and $352,599 in modeled monthly captured value.
2. Paula’s Choice is present, but not yet a default category leader. Paula’s Choice recorded 119 valid recommendations, a 7.33% top-three recommendation rate, a 3.91% rank-one rate, and $177,270 in modeled monthly captured recommendation value. That puts the brand meaningfully in the market, but behind CeraVe, La Roche-Posay, and SkinCeuticals on value-weighted visibility.
3. SkinCeuticals punches above its raw recommendation count. SkinCeuticals had fewer valid recommendations than Neutrogena and The Ordinary, but captured $219,538 in modeled monthly recommendation value, reflecting stronger performance in higher-value prompt moments such as vitamin C, aging skin, and hyperpigmentation.
4. Visibility is not the same as recommendation strength. Neutrogena earned more valid recommendations than Paula’s Choice, but Paula’s Choice had a stronger average recommendation rank and higher modeled captured recommendation value. That suggests the brand is effective when recommended, but not recommended broadly enough across the category.
5. The citation layer is heavily third-party driven. The raw citation environment includes editorial, commerce, review, retailer, and official sources, with recurring domains such as Vogue, Forbes, InStyle, Dermstore, Healthline, Ulta, Health, and Today. That pattern suggests AI systems are leaning on public evidence beyond brand-owned product pages when forming skincare recommendations.

What changed in the market

Dermatologist-recommended skincare is a trust category. Buyers care about skin type, sensitivity, active ingredients, irritation risk, budget, and whether a product feels clinically credible. That makes the category especially exposed to AI-led discovery.

A consumer asking “What is the best moisturizer for acne?” or “What are the best skincare brands?” is not simply looking for awareness. They are forming a buying shortlist. In those moments, AI systems compress the market into a few recommended brands and products. The commercial risk is that a brand can have strong awareness, strong product credibility, and strong search presence, but still lose the AI recommendation moment if competitors are framed as safer, more universal, or more frequently validated by third-party sources.

For dermatologist-recommended skincare brands, the new discovery battleground is not just ranking in Google. It is becoming the brand AI systems confidently include when buyers ask for help choosing.

What the benchmark found

The benchmark shows a clear two-brand leadership tier.

CeraVe and La Roche-Posay are the strongest broad-market AI recommendation brands. They appear to benefit from repeated association with dermatologist-style language, barrier support, sensitive skin, acne-prone skin, cleansers, moisturizers, and “safe default” recommendations.

Neutrogena and The Ordinary form the next layer of broad visibility. Neutrogena appears frequently in mainstream moisturizer, cleanser, and body-care prompts. The Ordinary is visible around ingredient-led and affordable active-ingredient recommendations, but its lower rank-one and modeled value numbers suggest it is more often included as an option than positioned as the leading answer.

SkinCeuticals is a value-weighted specialist winner. Its recommendation profile is not as broad as CeraVe or La Roche-Posay, but it performs strongly where higher-consideration product categories matter, especially anti-aging, vitamin C, and hyperpigmentation.

Paula’s Choice sits in the most strategically interesting position. The brand is clearly not invisible. It earns valid recommendation credit, rank-one placements, and meaningful modeled value. But the benchmark suggests Paula’s Choice is often framed as an active-ingredient or product-specific specialist rather than the universal dermatologist-recommended default.

That matters because AI systems are not only deciding whether to mention a brand. They are deciding how to place it in the buyer’s mental model.

Why visibility is not enough

Raw mention presence can make a brand look healthier than it is. The stronger question is whether the brand is receiving valid recommendation credit, appearing in the top three, ranking first, and being framed in a way that supports buyer confidence.

Paula’s Choice illustrates this gap. The brand had 20.20% raw mention presence and 19.38% valid recommendation coverage, but only 7.33% top-three recommendation rate across the full benchmark. Its average recommended rank was strong when it did earn rank credit, but it was not included often enough in top positions across broad category prompts.

In practical terms, Paula’s Choice can win when AI systems understand the prompt as active-ingredient, serum, exfoliant, or skin-concern specific. But in broad “best skincare brand” and dermatologist-recommended shortlist moments, CeraVe and La Roche-Posay appear more often as the safer default answer.

The issue is not simple visibility. It is shortlist eligibility.

The citation layer

The recommendation layer appears to be shaped by a mix of editorial authority, commerce content, retailer pages, review-style lists, and brand-owned sources. In skincare, that is especially important because AI systems need evidence for product claims, skin-type suitability, ingredient positioning, dermatologist approval, and comparative trust.

The recurring public-source pattern points to a category where third-party validation matters heavily. Brand sites can support product facts, but AI systems appear to rely on wider source footprints to decide which brands are familiar, safe, recommended, and appropriate for specific skin concerns.

For CeraVe and La Roche-Posay, the public evidence layer appears to reinforce broad dermatologist-style credibility. For SkinCeuticals, the evidence layer supports premium, clinical, and active-led use cases. For Paula’s Choice, the opportunity is to expand from product-specific credibility into broader category-default credibility.

This does not mean citation frequency equals endorsement. It means public sources are part of the evidence layer AI systems may synthesize when forming answers.

What brands need to fix

Dermatologist-recommended skincare brands need to treat AI discovery as a structured evidence problem, not a prompt-level copy problem.

They need stronger coverage around the exact buyer questions that form shortlists: best skincare brands, best moisturizer for acne, best products for hyperpigmentation, best skincare for aging skin, dermatologist-recommended moisturizers, cleanser comparisons, and product-vs-product evaluations.

They also need a cleaner public evidence layer. That includes dermatologist-adjacent editorial mentions, credible retailer content, comparison pages, expert-led ingredient education, review and forum visibility, and owned content that clearly connects products to skin concerns without overclaiming.

For Paula’s Choice specifically, the strategic gap is positioning. The brand already has strong ingredient authority. The opportunity is to reinforce why that ingredient authority should also make it a default dermatologist-recommended skincare brand, not only a specialist answer for exfoliants, boosters, and actives.

How CiteWorks Studio helps

1. Map AI recommendation visibility. Track prompts, platforms, company presence, valid recommendations, top-three and rank-one performance, framing, and citation sources.
2. Identify the sources shaping AI answers. Find the editorial, review, forum, government, directory, owned, and search-visible sources that influence brand framing.
3. Build the citation architecture plan. Strengthen the public evidence layer so AI systems have more accurate, consistent, and persuasive source material to synthesize.

Commercial takeaway

AI search is not simply repeating traditional skincare brand awareness. It is building buyer shortlists from public evidence, product framing, editorial validation, retailer visibility, and prompt-specific relevance.

CeraVe and La Roche-Posay currently benefit from being treated as broad, low-risk, dermatologist-style defaults. SkinCeuticals benefits from specialist value in high-consideration active categories. Paula’s Choice has meaningful AI recommendation strength, but the benchmark suggests its next growth opportunity is to move from specialist ingredient credibility into broader default-category eligibility.

For skincare brands, the commercial question is no longer only “Are we visible?” It is “Are we recommended when the buyer is ready to choose?”

CTA

Want to know how AI systems are recommending your skincare brand?

CiteWorks Studio can help map where your brand appears, where competitors are recommended instead, which sources are shaping AI answers, and what needs to change to improve recommendation-stage visibility.

Request an AI Visibility Audit or Citation Architecture Review.