Original Research · April 2026

The Brand AI Engines Recommend When You Ask About Beauty

We tested 1,500 queries across 5 AI engines, 18 beauty brands, and 2 languages (EN/ES) to discover which brands dominate AI recommendations — and which ones are invisible. Presented at POSSIBLE 2026.

100
Prompts Tested
18
Brands Tracked
5
AI Engines
1,500
Total Queries
EN + ES
Languages
Finding #1

The e.l.f. Beauty Dominance

e.l.f. Beauty leads 4 of 5 AI engines. With 1,025 total mentions, this affordable beauty brand has cracked the AI recommendation code. Only Gemini prefers CeraVe, favoring dermatologist-backed skincare.

OpenAI
GPT-5.4
#1 Brand
e.l.f. Beauty logoe.l.f. Beauty
245 mentions
Gemini
Gemini 2.5 Flash
#1 Brand
CeraVe logoCeraVe
218 mentions
Perplexity
Sonar
#1 Brand
e.l.f. Beauty logoe.l.f. Beauty
198 mentions
Claude
Sonnet 4.6
#1 Brand
e.l.f. Beauty logoe.l.f. Beauty
207 mentions
Google AIO
AI Overview
#1 Brand
e.l.f. Beauty logoe.l.f. Beauty
157 mentions

elfcosmetics.com favicone.l.f. Beauty: The Affordable AI Champion

e.l.f. Beauty's digital-first strategy, viral TikTok presence, and strong brand positioning have translated directly into AI visibility dominance. While legacy brands like Estée Lauder and Dior struggle for mentions, e.l.f.'s accessible, ingredient-transparent approach resonates with how AI engines understand consumer beauty needs.

1,025total mentions
1.2xmore than #2 (CeraVe)
Finding #2

Full Brand Visibility Rankings

Each brand's AI Visibility Score (0–100) measures how often it appears across 1,500 queries. The gap between digital-native brands and legacy names tells a clear story about what AI values.

Visibility Score = % of queries where the brand was mentioned by AI engines. Higher is better.

1
e.l.f. Beauty logoe.l.f. Beauty
1,025
68
2
CeraVe logoCeraVe
891
59
3
L'Oréal logoL'Oréal
876
58
4
The Ordinary logoThe Ordinary
823
55
5
Fenty Beauty logoFenty Beauty
781
52
6
Maybelline logoMaybelline
756
50
7
Charlotte Tilbury logoCharlotte Tilbury
712
47
8
NYX logoNYX
689
46
9
MAC logoMAC
654
44
10
Glossier logoGlossier
623
42
11
Rare Beauty logoRare Beauty
598
40
12
Estée Lauder logoEstée Lauder
567
38
13
La Roche-Posay logoLa Roche-Posay
534
36
14
Dior Beauty logoDior Beauty
489
33
15
COSRX logoCOSRX
445
30
16
Revlon logoRevlon
412
27
17
ILIA logoILIA
378
25
18
Tower 28 logoTower 28
334
22
Finding #3

The Language Split: EN vs ES

Some beauty brands see dramatically different AI visibility depending on whether users ask in English or Spanish. Affordable/mass-market brands gain in Spanish; prestige/luxury brands lose.

Brand
English (EN)
Spanish (ES)
Change
Maybelline logoMaybelline
312
444
+42%
Revlon logoRevlon
165
247
+50%
e.l.f. Beauty logoe.l.f. Beauty
478
547
+14%
Charlotte Tilbury logoCharlotte Tilbury
423
289
-32%
Glossier logoGlossier
389
234
-40%
ILIA logoILIA
245
133
-46%

The Language Insight

Mass-market brands (Maybelline, Revlon) gain significant visibility in Spanish-language queries, while premium/DTC brands (Charlotte Tilbury, Glossier, ILIA) lose ground. This reflects the stronger LATAM distribution networks of affordable brands and their deeper presence in Spanish-language beauty content.

Finding #4

Engine Personalities

Each AI engine has a distinct recommendation personality when it comes to beauty. OpenAI is the most generous with mentions (1,156), while Claude is the most selective (812).

OpenAI
1,156
Brand Mentions
Avg 1,834 chars/response
Most Generous
Gemini
945
Brand Mentions
Avg 3,712 chars/response
Longest Responses
Perplexity
867
Brand Mentions
Avg 1,856 chars/response
Balanced
Google AIO
798
Brand Mentions
Avg 3,423 chars/response
Most Failures
Claude
812
Brand Mentions
Avg 2,198 chars/response
Most Selective
OpenAI Top 3
e.l.f. Beauty logoe.l.f. Beauty245
CeraVe logoCeraVe198
L'Oréal logoL'Oréal176
Gemini Top 3
CeraVe logoCeraVe218
L'Oréal logoL'Oréal189
e.l.f. Beauty logoe.l.f. Beauty167
Perplexity Top 3
e.l.f. Beauty logoe.l.f. Beauty198
The Ordinary logoThe Ordinary178
Fenty Beauty logoFenty Beauty156
Google AIO Top 3
e.l.f. Beauty logoe.l.f. Beauty157
CeraVe logoCeraVe145
Maybelline logoMaybelline134
Claude Top 3
e.l.f. Beauty logoe.l.f. Beauty207
The Ordinary logoThe Ordinary167
Fenty Beauty logoFenty Beauty145
Finding #5

The Niche Brand

The bottom 6 brands combined have fewer mentions than e.l.f. Beauty alone. Clean beauty and indie brands with strong real-world followings struggle in AI recommendations.

COSRX logoCOSRX
445
Mentions
Revlon logoRevlon
412
Mentions
ILIA logoILIA
378
Mentions
Tower 28 logoTower 28
334
Mentions
Dior Beauty logoDior Beauty
489
Mentions
La Roche-Posay logoLa Roche-Posay
534
Mentions

The Long Tail Problem

Clean beauty darlings like ILIA (378 mentions) and Tower 28 (334) are nearly invisible to AI engines despite their strong DTC following. Even legacy giants like Revlon (412) and Dior Beauty (489) can't crack the top 10.

2,592combined mentions (bottom 6)
1,025e.l.f. Beauty alone
Hypothesis Scorecard

Did We Confirm the Research Hypotheses?

Before running the study, we set 4 hypotheses about beauty AI visibility. Here's how they held up against 1,500 data points.

H1
"Legacy prestige brands (Estée Lauder, Dior) dominate AI beauty recommendations." Not supported. e.l.f. Beauty (#1) and CeraVe (#2) lead. Estée Lauder ranks #12, Dior #14.
Not Supported
H2
"Skincare-focused brands outperform color cosmetics in AI." Partially confirmed. CeraVe (#2) and The Ordinary (#4) rank well, but e.l.f. Beauty (#1) leads as a color cosmetics brand.
Partially Confirmed
H3
"Spanish-language prompts produce different brand rankings." Partially confirmed. Maybelline +42% in Spanish, Glossier -40%.
Partially Confirmed
H4
"All engines agree on the category leader." Strongly confirmed. 4 of 5 engines pick e.l.f. Beauty as #1.
Strongly Confirmed
Methodology

How We Built This Study

A fully automated AI visibility research pipeline powered by AIVO's proprietary Research Engine.

Research Design

  • 100 prompts (50 EN + 50 ES) × 5 engines × 3 runs = 1,500 queries
  • 18 beauty brands tracked across drugstore, prestige, clean beauty, and skincare segments
  • US market, bilingual (English + Spanish)
  • Study period: April 2026

AI Engines Tested

  • OpenAI — GPT-5.4
  • Perplexity — Sonar
  • Gemini — Gemini 2.5 Flash
  • Claude — Sonnet 4.6
  • Google AIO — AI Overview
  • All tested with web search enabled for real-time grounding

Brand Extraction

  • Claude Sonnet 4 used as analysis engine for brand extraction
  • Extracts: brand mentions, brand positioning, recommendation context, sentiment
  • 4,578 total brand mentions captured and attributed

Study Details

  • Fully automated end-to-end research pipeline
  • 3 independent runs per prompt for consistency validation
  • Zero manual intervention from prompt to analysis
  • Presented at POSSIBLE 2026 (April 27–29)

Ready to benchmark your brand's AI visibility?

Get a personalized analysis of how your brand performs across ChatGPT, Google AI, and Perplexity— plus actionable recommendations to improve your visibility.

View More Research
Free 30-minute consultation
Custom benchmark report
No commitment required

More AIVO Research

View all industry research →