Skip to main content
RoastIQBuyerLensHugoPricingBlogAbout
Book a demoSign inStart free →
Market insight · n=312 · 19 brands · published 18 May 2026 · 8 min read

Skincare ads have converged.

67% of skincare cuts in our 312-ad pool share more than six visual codes. Build Brand median 52, the lowest of 20 categories. Only four of 19 brands escape. The 12 category-default codes, the four brands that own a distinctive system, and the audit that gets you out.

Sample · n=312 Brands · 19 Median Build Brand · 52 Pattern 07 fire-rate · 43% Confidence · HIGH
Oussama Nakhil portrait
Oussama Nakhil · Founder & CEO
Multiple years buyer-side: NielsenIQ insights, then L'Oréal Groupe in global marketing insights (incl. skincare brand portfolios). Every chart pinned to model_version and benchmark_pool_version v.2026-05.
67%
Cuts sharing >6 category-default codes
52
Median Build Brand · lowest of 20 categories
4 / 19
Brands escaping the convergence

01 · The crisis, quantified.

Skincare advertising is the most converged category in our benchmark. Across 312 cuts spanning 19 brands captured between 2025-Q4 and 2026-Q2, 67% of cuts share more than six visual codes with at least three direct competitors. The category's median Build Brand score is 52, the lowest of the 20 categories we benchmark, and a full 14 points below the cross-category median of 66. Pattern 07 (the generic-category opener, one of our seven systematic failure modes) fires in 43% of cuts.

The convergence is not stylistic preference. It is the visible residue of a decade of agency-rotation, retail-channel parity briefs, and ingredient-led claims arms races that have hollowed out distinctive-asset investment. The category brief has converged on a small set of visual codes that read as "skincare" to the audience and as "interchangeable" to the algorithm. The Brand Impact penalty is structural, not aesthetic.

Convergence rate
67%
Cuts sharing >6 codes with 3+ competitors
Median Build Brand
52
14 points below cross-category median
Pattern 07 rate
43%
Generic-category opener · 7 failure modes
Escaping brands
4/19
Own ≥5 distinctive codes
The diagnosis

"Skincare did not converge by accident. It converged because every brief borrowed from every brief, and nobody invested in distinctive assets that survive a feed scroll without the wordmark."

02 · The 12 category-default codes.

The pipeline's category-attribute extractor identifies 12 visual codes that appear in >40% of skincare cuts in our pool. None of them is uniquely owned. All of them read as "skincare" to the audience because every brand uses them. Most skincare cuts in our pool feature six to ten of these. The cuts that score above the median feature five or fewer.

  • (1) Pipette / dropper. Appears in 71% of cuts. The category's most over-used opener. Once a serum tell, now a generic-skincare tell.
  • (2) Pink-or-beige palette. 64%. The default colour temperature reads as "skincare-feminine," not as any brand.
  • (3) Lab badge / clinical seal. 58%. Was a credibility cue in 2018; is a category-default in 2026.
  • (4) Lower-third claim copy. 56%. The format (small sans-serif, percentage, asterisk) is shared across the category.
  • (5) Face mid-application. 54%. The product applied to a model's face, mid-action. Universal to the category.
  • (6) Slow-mo serum drop. 51%. The 120fps drop from the pipette is now category cinema.
  • (7) Packaging beauty shot. 49%. Studio-lit bottle on a clean surface, often rotating. Replaceable with any other brand's bottle.
  • (8) Clinical superscript. 47%. Numbers and chemical names floating on screen as credibility theatre.
  • (9) Soft studio lighting. 46%. Diffused key, no shadow. Reads as "premium" and as "every other brand."
  • (10) Sans-serif lower-third. 45%. The typography choice is shared by 9 of 19 brands in our pool.
  • (11) Eyebrow-to-cheek focal vector. 43%. The camera frames the face from brow to cheekbone in a 75% of all category cuts featuring a model.
  • (12) Glow-up transition. 41%. Before-after dissolve with brightness lift. The category's signature transition; nobody owns it.

03 · Code density vs Brand Impact.

The relationship between uniquely-owned distinctive codes (not category-default codes, the opposite) and Brand Impact is monotonic and steep. Brands that own five or more uniquely-owned codes score a median Brand Impact of 78. Brands with one or two score 46. The category-default cohort, cuts featuring 8+ defaults and 0 unique codes, scores a median Brand Impact of 28.

// Brand Impact by distinctive-code density (n=312)

More uniquely-owned codes = higher Brand Impact. The slope is steep. The category default is the bottom band.
≥5 distinctive codes
78
3–4 distinctive codes
62
1–2 distinctive codes
46
0 distinctive codes
28

Two implications follow. First, the Brand Impact lever in skincare is the distinctive-code count, not the production-value budget. A €40k pipette-and-glow-up shoot scores below a €4k The-Ordinary-style typographic still. Second, the category's brand-equity ceiling is being capped by the brief, not by the brand. Brands that decide to invest in distinctive codes can move 30+ Brand Impact points without changing their product, their proposition, or their media budget.

04 · The four escapees.

Four brands of the 19 we benchmark have built distinctive-code systems that the pipeline flags as uniquely-owned across the category. Each owns five distinctive codes that no other brand in the pool uses, each scores 18+ Brand Impact points above the category median, and each treats their distinctive assets as part of the brief rather than as a creative variable.

// Brand 01The Ordinary, typography as brand
Build Brand · 76Brand Impact · 79Distinctive codes · 5

The Ordinary's distinctive-asset system reads as a refusal of skincare's visual grammar. No pipette pageantry, no model face, no glow-up, instead, the brand commits to a typographic language that no competitor has copied at the same craft level.

// 5 owned codes Tabular ingredient typography (monospaced) · lab-canister minimalism · no-face composition · plain-white-on-paper backgrounds · scientific tone-of-voice in copy and audio.
// Brand 02CeraVe, dermatologist narrator system
Build Brand · 72Brand Impact · 74Distinctive codes · 5

CeraVe built its escape on credibility-led storytelling, the dermatologist narrator is the brand's lead asset, and the rest of the system supports it. Credibility is the lever, not aesthetics.

// 5 owned codes Dermatologist narrator (consistent talent system) · white-blue palette · ingredient close-ups (not pipette drops) · no-pipette opening grammar · accent-based credibility (US, AU, MENA dermatologists).
// Brand 03Glossier, branded pastel ecosystem
Build Brand · 74Brand Impact · 76Distinctive codes · 5

Glossier's system reads in a single frame. The colour, the typography, the recurring talent, and the pacing form a brand-specific ecosystem that survives without the wordmark.

// 5 owned codes Branded pastel palette (millennial pink ownership) · recurring talent system (consistent faces across years) · specific kerning and tracking on lower-thirds · lowercase wordmark treatment · soft-pink-on-cream colour pairing.
// Brand 04Typology, apothecary-brown system
Build Brand · 71Brand Impact · 73Distinctive codes · 5

Typology built a French-apothecary visual language and committed to it across every cut. The brown packaging is the lead distinctive cue, and the rest of the system supports the apothecary frame.

// 5 owned codes Apothecary-brown packaging (single colour ownership) · no-model close-ups (product-only composition) · label-first composition (typography over face) · French-apothecary tone-of-voice · single-ingredient claim format.

05 · Audit your distinctive code list.

The audit is a five-step exercise. It does not require new production. It requires honesty about which of your existing visual elements are actually owned vs. shared with the category.

  • (1) Inventory. List every visual element your last 10 cuts share, colours, typography, framing, talent, lighting style, transitions, packaging treatment. Aim for 15-20 elements.
  • (2) Cross-reference. Pull three direct competitors' last 10 cuts each. Tag every element from your inventory against whether it also appears in any competitor.
  • (3) Strike the shared. Eliminate anything that appears in more than one competitor. What is left is your candidate distinctive-asset list. If less than three survive, you are in the convergent cohort.
  • (4) Test against Ehrenberg-Bass criteria. For each survivor, ask three questions: Is it uniquely yours (no competitor uses it)? Is it famous (would the category audience associate it with your brand without the wordmark)? Is it consistent (has it appeared across cuts and years)? Strike anything that fails any one.
  • (5) Brief one in every cut. Treat the survivors as non-negotiable distinctive cues. Every cut should include at least one inside the first 1.5 seconds. If the cut does not include a distinctive cue, it fails approval, not because of taste, but because the cut is structurally a category-default by definition.
The exercise

"Run the five-step audit before commissioning your next cut. If less than three elements survive Step 3, you are not briefing skincare creative, you are commissioning category wallpaper."

06 · Three before/after rebuilds.

EX-01a Convergent skincare cut · "Sharpen 58" SHARPEN · 58
Format · Generic 15sCategory codes · 9/12 firedDistinctive codes · 0Hook · Pipette + serum drop

The before. A 15-second cut for a hypothetical mid-tier skincare brand. Pipette at 0.4s, slow-mo serum drop at 1.1s, face mid-application at 2.0s, glow-up transition at 8s, clinical superscript at 11s, packaging beauty shot at 13s. Nine of the 12 category-default codes fire. Zero uniquely-owned distinctive codes. Beat the Skip 64, Build Brand 38, composite 58. Sharpen verdict. The cut is craft-competent and category-anonymous.

EX-01b Rebuilt cut · "Scale 73", same brand, same product SCALE · 73
Format · Distinctive 15s rebuildCategory codes · 3/12 firedDistinctive codes · 4Hook · Typographic claim + sonic tag

The after. Same brand, same product, same 15-second runtime. The rebuild strips eight of the nine category-default codes and adds four uniquely-owned distinctive codes the brand develops over the audit exercise: a typographic claim plate that opens the cut (own kerning, own colour), a 2-second sonic logo tag, a packaging-silhouette opener (no pipette, no model face), and a recurring talent who appears across all the brand's cuts. Beat the Skip 76, Build Brand 71, composite 73. Scale verdict. The fix is brief discipline, not production budget, the rebuild cost is similar to the original.

EX-02 Anchored to The Ordinary's system, case study SCALE · 78
Brand · The OrdinaryFormat · 12sCategory codes · 1/12 firedDistinctive codes · 5

A canonical The Ordinary cut: tabular ingredient typography (Acid 7%, Concentration 5ml) opens the cut at 0.3s on white-on-paper. No pipette, no face, no glow-up. The lab-canister stands centred at 2s. The scientific tone-of-voice closes the cut. One of the 12 category defaults fires (sans-serif lower-third, and even that is treated distinctively). Five uniquely-owned distinctive codes. Beat the Skip 81, Build Brand 76, composite 78. The cut is the canonical version of the brand's system; it is what every cut in the category could look like if the brief committed.

07 · Implications by stakeholder.

// For the CMO

The skincare convergence is a brand-equity emergency that is invisible on the media report. Brand Impact is the lowest of 20 categories. Mandate a distinctive-code audit before approving the next campaign. The lift is in the brief discipline, not in the production budget. Look at the four escaping brands, none of them outspends the category leaders.

// For the brand team

If your brand book has more pages on "do not" rules than on owned distinctive assets, you are in the convergent cohort. Build a five-asset distinctive system on the The Ordinary / CeraVe / Glossier / Typology pattern. Strike pipette, glow-up and clinical superscript from your default visual library until you have earned them back.

// For the agency

Stop pitching "fresh takes" on the category visual language. The category is converged because every fresh take has converged. Pitch distinctive-system builds, five uniquely-owned codes committed to over multiple years, instead of cut-level executions. The four escaping brands hire creative partners who pitch systems, not films.

// For the media buyer

Tag every skincare cut by distinctive-code count in your reporting. Cuts with 0-2 uniquely-owned codes are silent Brand-Impact bleeders. Pull spend from category-default cuts before increasing budget. The 30-point Brand Impact lift from a distinctive-system rebuild outperforms any media-mix optimisation you can negotiate at the buy.

// Three things to change tomorrow
01
Run the five-step distinctive-code audit
Inventory, cross-reference, strike, test against Ehrenberg-Bass, brief into every cut. Two days of brand-team work. The category's largest single brand-health lever.
02
Strike pipette, glow-up and clinical superscript
Until you have earned distinctive variants of these category codes, treat them as banned defaults. The cuts that miss the convergence are the cuts that miss these.
03
Commit to a five-asset distinctive system for three years
The Ordinary's typography is 12 years old. Glossier's pink is 11. Distinctive assets are bought with time, not budget. Pick five and commit.

08 · Method.

Method. 312 skincare ads across 19 brands sourced from Meta Ad Library, TikTok Creative Center, TikTok Ad Library, Google Ads Transparency Center, and manual editorial curation between 2025-Q4 and 2026-Q2 capture. Each scored through the RoastIQ pipeline against benchmark pool v.2026-05 (n=2,047) via Vertex AI Gemini 2.5 Flash/Pro multimodal + Google Video Intelligence (shot/label) + Speech-to-Text (transcription) + TranSalNet-class saliency, with Zod-validated structured output. Category-default code detection (the 12-code set) is performed via the multimodal attribute extractor, validated against manual annotation on a 140-cut set at Cohen's κ 0.81. Uniquely-owned distinctive-code attribution is performed per-brand by cross-referencing each candidate code against the rest of the 19-brand pool, a code is "owned" only if no other brand in the pool uses it. Confidence label HIGH (n > 300, ρ +0.31 OOS TikTok engagement, top-vs-bottom quintile lift 6.5×). 5 KPIs: Beat the Skip (25%) · Get Noticed (20%) · Brand Impact (20%) · Sell Proposition (20%) · Build Brand (15%). We do not scrape.

What we do not claim. The Brand Impact and Build Brand penalties are model predictions calibrated against public engagement and click-intent signals. They do not measure in-market sales, ROAS, attributed conversion or brand recall. The four "escaping" brands are escaping on Brand Impact, we make no claim about their financial outcomes vs the rest of the category. The before/after rebuild example is illustrative; brand-identifying detail is held back where licence required.

FAQ.

What is the skincare ad convergence crisis?
67% of skincare ad cuts in our 312-ad pool share more than six visual codes with at least three direct competitors. The category's median Build Brand score is 52, the lowest of the 20 categories we benchmark. Pattern 07 (generic-category opener, one of our seven systematic failure modes) fires in 43% of cuts. The category is structurally converged.
What are the 12 category-default skincare codes?
Pipette, pink-or-beige palette, lab badge, lower-third claim, face mid-application, slow-mo serum drop, packaging beauty shot, clinical superscript, soft studio lighting, sans-serif lower-third, eyebrow-to-cheek focal vector, glow-up transition. Most skincare cuts in our pool feature six to ten of these. Brand Impact falls in monotonic relation to the count.
Which brands are escaping the convergence?
Four brands of the 19 we benchmark: The Ordinary (tabular ingredient typography, lab-canister minimalism, no-face composition, plain-white-on-paper, scientific tone-of-voice). CeraVe (dermatologist narrator, white-blue palette, ingredient close-ups, no-pipette opening, accent-based credibility). Glossier (branded pastel palette, recurring talent system, specific kerning, lowercase wordmark, soft-pink-on-cream). Typology (apothecary-brown packaging, no-model close-ups, label-first composition, French-apothecary tone, single-ingredient claim). Each owns five distinctive codes that no other brand in the category uses.
What is the relationship between code density and Brand Impact?
Brands with five or more uniquely-owned distinctive codes score a median Brand Impact of 78. Brands with three or four score 62. Brands with one or two score 46. Brands with zero unique codes, the category-default cohort, score 28. The relationship is monotonic, steep, and stable across pool versions.
How do you audit your distinctive code list?
Five steps: (1) inventory every visual element your last 10 cuts share; (2) cross-reference against three category competitors' cuts; (3) strike anything that appears in more than one competitor; (4) test what remains against Ehrenberg-Bass criteria of uniqueness, fame, consistency; (5) brief one of the survivors into every cut as a non-negotiable distinctive cue inside the first 1.5 seconds.
Why is skincare specifically more converged than other beauty categories?
Three structural reasons: (1) the ingredient-led claims arms race standardised "credibility theatre" elements (lab badges, percentages, clinical superscript), (2) agency-rotation and retail-channel parity briefs hollowed out long-term distinctive-asset investment, (3) the 2018-2022 indie-skincare wave borrowed from the same Pinterest visual vocabulary and converged. Cosmetics and fragrance have not converged to the same degree.
What is the confidence label on this report?
HIGH. n=312 cuts across 19 brands, held-out OOS Spearman ρ +0.31 on engagement, top-vs-bottom quintile lift 6.5×. The 12-code default set detection is validated at Cohen's κ 0.81 against a 140-cut annotation set. Uniquely-owned distinctive-code attribution is performed per-brand by cross-reference against the rest of the 19-brand pool.

Sources. All ads in the underlying pool are sourced from official public transparency tools (Meta Ad Library API, TikTok Creative Center, TikTok Ad Library, Google Ads Transparency Center) or manual editorial curation with documented licence terms. We do not scrape. Pipeline. Vertex AI Gemini 2.5 Flash/Pro (multimodal) + Google Video Intelligence (shot/label) + Speech-to-Text (transcription) + TranSalNet-class saliency, all scored against benchmark pool v.2026-05 (n=2,047) via Zod-validated structured output. Validation. Held-out OOS Spearman ρ +0.31 TikTok engagement (n=700), +0.30 TikTok CTR (n=691), +0.32 YouTube view counts (n=403). Attribute-detection accuracy ~85% (formal IRR paper in pipeline). Construct. We predict engagement and click intent against public behavioural outcomes, not sales, ROAS, attributed conversion or brand recall. The four "escaping" brand identifications are based on Brand Impact and distinctive-code system completeness; we make no claim about their financial outcomes vs the rest of the category. Reproducibility. Every chart is pinned to model_version and benchmark_pool_version.