- Parallel Universes
- Posts
- Skincare Meets AI: The New Visibility Rules (Part 2)
Skincare Meets AI: The New Visibility Rules (Part 2)
Answers beat ads
Victor G. Clatici
September 11, 2025
Bridge from Part 1
In Part 1, we stood with the patient at 3 AM and watched a decision unfold as a dialogue, not a search. We showed why recommendations replace reviews, why trust shifts to process, and why AI is the next front door to skincare decisions. Today we open that door from the brand side: why some names appear in the answer while others vanish in the scroll—and what to do about it. Start with [P1].
That 3 AM moment wasn’t a one-off; it’s a preview of how discovery decisions are being made now—at scale. While you slept, thousands of near-identical conversations played out across the Big Five, quietly redirecting millions in aggregate spend. Semrush’s forward curves point the same way: LLM value overtakes classic organic around 2026; visitors cross around 2028—faster if AI Mode becomes default [B].
Here’s the uncomfortable reality: your customer already had that conversation, received named recommendations, and a why. If the Big Five can’t find, understand, or reuse your product description, they’ll recommend something they can explain instead—often pulling from influencer comparison pages that speak in clear, quotable chunks.
The patient got their answer. Did you get their business?
Your First Customer Is an Algorithm (B2A in practice)
For the patient, the Big Five — ChatGPT, Claude, Copilot, Perplexity, and Gemini — are convenient. For a brand, they’re the first gatekeeper. The pre-sale conversation now happens before anyone reaches your site, counter, or PR. And the answer’s location is shifting: by March 2025, Google’s AI Overviews appeared on ~13% of searches (U.S. desktop) [A], with projections that LLM-mediated value overtakes classic organic around 2026, and visitors around 2028 — sooner if AI Mode becomes default [B][C]. Translation: answers beat ads; explainability beats volume.
Blunt truth: if a product can’t be explained cleanly against a patient’s constraints, it’s unlikely to appear. When you aren’t usable in the answer, someone else is — very often influencer comparison pages the models can quote because they’re already written as FAQs, side-by-side blocks, “Top 10s,” price/size callouts (and, yes, often better structured than your PDPs – Product Detail Pages).
The algorithmic gatekeeper doesn’t care about heritage or budget. It cares whether it can state one precise sentence for why your product fits this ask without distorting facts. If it can’t find clear, consistent, machine-usable information on your surfaces (the human facts you’d want repeated — and, secondarily, things like FAQs, comparison blocks, schema/JSON-LD), it doesn’t wait. It routes around you and recommends something it can explain instead.
Earn the sentence, not the slot. AI is the next front door.
The Permission Problem (why the answer skips you)
When the ask is “office-friendly sunscreen, high UVA, no white cast, under €25,” the Big Five map constraints to descriptions they can reuse. If your page speaks in adjectives and hides facts, the models reach for sources that state the usable details—filters, UVA focus, finish, layering behavior, price.
That’s why influencer pages often surface: not because they’re more authoritative, but because they’re already written in comparison-friendly blocks (FAQs, side-by-side notes, price/size callouts).
If you don’t speak clearly, the answer will speak about you without you.
Reality check: LLMs don’t “owe” you a mention. They assemble usable sentences from what’s easiest to assemble. If your product info isn’t precise, consistent, and findable, the front door opens straight into someone else’s living room.
Here’s how it plays out. Your PDP says, “advanced protection with an elegant finish.” A blogger writes, “no white cast, dries down matte, layers under makeup, €22.” The Big Five quote the blogger because that sentence actually answers the ask. The blogger becomes your accidental spokesperson—defining your product for thousands who never visit your site.
This isn’t algorithm bias. It’s information economics. Clear, structured facts beat beautiful, vague copy—every time. If you won’t be quotable, someone else will be quoted about you. AI is the next front door.
From Rankings to Answers (and what changes inside the answer)
The old SEO game said: climb a list. The new game says: earn the sentence. Not softer—stricter. A winning sentence must:
Fit the ask (clear constraints),
Hold up under scrutiny (plain reasons),
Sound human (a line a person can repeat).
That’s why answer share matters more than rank—the share of moments your name appears inside the answer when the ask fits you.
From the patient side [P1]: ask for a makeup-friendly SPF and you’ll get 2–3 candidates with a why. If yours isn’t there, it’s not because AI hates your brand; it’s because AI can’t use your page.
The shift is mechanical, not personal. When the Big Five can’t extract clean facts from your product page, they don’t wait or apologize—they route around you and quote whoever wrote the details the question needs. Your “perfect” product goes invisible while a competitor with basic, usable information gets recommended. The sentence that wins is the one that directly answers the ask in words a human can repeat without confusion.
AI is the next front door—and it opens for brands that speak in complete, usable sentences.
What AI Can Use (no playbook, just common sense)
If it can’t be quoted, it won’t be recommended
Facts that travel: What it does, for whom, how it behaves (finish/feel), where it shines, where it doesn’t.
Comparisons that matter: “high UVA focus,” “non-stripping cleanse,” “layers under makeup,” “reapply reality.”
Cautions that build trust: who should go slow, who should patch-test, what not to stack the same night.
These aren’t marketing bullets—they’re the minimum viable clarity the Big Five need to recommend your product with a straight face. When someone asks “what cleanser won’t dry out my skin,” they look for specific descriptors, not emotional copy.
Where AI finds those facts (with or without you)
If you won’t say it, someone else already did—in FAQs, comparison tables, how-to routines, and price/size callouts on retailer and influencer pages. That’s what models quote when your PDP reads like poetry.
Aggressive truth: refuse precision and you outsource your brand story to a stranger’s Top 10. That stranger becomes your default spokesperson inside the answer.
It’s simple: own the sentence or be spoken for. Every vague description is an invitation for someone else to define what you sell. AI is the next front door—and it opens for products that speak in complete, usable sentences.
The Minimal Clarity Stack (patient-safe, model-usable)
This isn’t a developer manual; it’s a clarity posture — the smallest set that makes you quotable without tricks.
Name & use-case: one-breath line a pharmacist would nod at.
Fit / misfit: who it’s for, who it isn’t — concise, human, no scare tactics.
Real-world behavior: finish/feel, layering, reapplication reality.
One honest trade-off: shine vs. comfort, film vs. reapplication, speed vs. tolerance.
Five plain-English FAQs: the questions you actually get at the counter.
Simple comparison: two near-neighbors + one alternative — why pick this vs. that.
Safety line (repeatable): the exact sentence you’d want a teen to hear at 3 AM.
Yes, JSON-LD, FAQ blocks, comparison tables, consistent naming help. No, you don’t need to think like a crawler. Think like a human who needs to be quoted accurately — because the model must produce a sentence a human can trust.
This stack works because it mirrors how people actually recommend products. When a dermatologist suggests a cleanser, they don’t recite slogans — they explain what it does, for whom, and what to expect. The Big Five need the same clarity to represent you. Skip a piece, and AI fills the gap with someone else’s description (often an influencer grid). AI is the next front door, and it opens for brands that speak in complete, trustworthy sentences.
Three Patient Moments (and why you may or may not appear)
Daily Sunscreen, Makeup-Friendly
Ask: “High UVA, no white cast, works under makeup, ≤ €25.” If your page won’t state UVA focus, finish, and layering behavior, the Big Five will quote an influencer grid that does. Being “beloved by X” won’t save you when the ask is that precise—the sentence needs the specifics.
Non-Stripping Cleanser for Breakouts
Ask: “Helps with breakouts, doesn’t strip.” If all you say is “gently cleanses,” the answer will reuse someone else’s wording that calls out surfactant type, after-wash feel, and what to pair at night. If you don’t own your sentence, someone will lend you theirs—and take the credit.
Lipstick with SPF
Ask: “Daily lip color with SPF, comfortable, low flavor.” If you never address reapply reality or comfort vs. flavor, the answer leans on bloggers who do. Not malice. Utilitarian math.
The pattern repeats across every category: the Big Five need concrete descriptors to build useful recommendations. When your product pages speak in abstractions like “transformative formula” or “luxurious feel,” they turn to sources that provide the operational details patients actually need. Those sources become the voice of your product in thousands of conversations.
This isn’t about “algorithm preferences.” It’s information utility. The cleaner and more specific your facts, the likelier you are to appear when someone asks a precise question. Vague copy doesn’t just dent rankings; it makes you invisible in the conversation where decisions happen. AI is the next front door—own the sentence or be spoken for.
ALFA — Four Principles That Survive Platform Shifts
Anchor — eight clear words you repeat everywhere. Example: “Office-friendly SPF gel — high UVA, no white cast.”
Logic — one breath: what it does, for whom, why it makes sense. Example: “Balances T-zone, cushions cheeks; good for combination skin starting evenings.”
Form — names, sizes, claims, cautions that don’t change with the weather. Example: the same tone/size/price and the same safety note on site, retailer, PR.
Aroma — a distinctive, calm voice; not a slogan machine. Example: pharmacist-plain sentences, confident without hype.
ALFA isn’t a code snippet. It’s how you become quotable on contact—to a model, to a pharmacist, to a teen at 3 AM.
The framework works because it forces consistency across every surface where your product appears. When the Big Five encounter your sunscreen on your site, a retailer page, and an influencer mention, ALFA makes sure they meet the same core facts in the same clear language. Without that, models encounter drift and default to the clearest consistent source—which may not be yours.
ALFA also future-proofs your narrative. New models will launch; existing ones will change. Brands with stable, clear sentences adapt faster than those rewriting themselves every quarter. These principles outlive formats because they’re grounded in communication clarity, not temporary tricks. AI is the next front door, and ALFA is the key that works in every lock.
The G Factor — Be Recognizable on First Touch
As you laid out in [G], the G Factor is semantic coherence: one spine, many surfaces. When that spine exists, the Big Five can recognize and reuse you without warping meaning. Without it, your story mutates as it travels from PR to retailer to reviewer—and models pick the cleanest version they can find, even if it isn’t yours.
The mutation problem is real (and expensive). Your press release says “advanced UV protection,” a retailer lists “broad-spectrum sunscreen,” an influencer calls it “reef-safe SPF.”
Models treat these as different concepts, not the same product. Result: fragmented visibility and confused recommendations. The G Factor fixes this by establishing one authoritative description that travels intact.
How to hold the spine (human, not technical):
Canonical sentence (use everywhere): “Office-friendly SPF gel — high UVA focus, no white cast.”
Alias map (allowed ≠ anything goes): “broad-spectrum” = OK; “advanced protection” = avoid; “reef-safe” = use only if substantiated.
Copy lock: same one-breath line on site, retailer, PR, influencer brief.
If every surface says the same clear sentence, the Big Five meet the same product each time—and reuse your wording. AI is the next front door. The G Factor makes you recognizable on first touch.
People, Not Playbooks
Nobody Has Five Years’ Experience (and that’s fine)
Your point in [H] stands: hire for curiosity, precision, taste. Give one mandate: make our story explainable on contact.
Three simple working rules:
One-breath test: if a human can’t say it in one breath, rewrite.
Single owner per sentence: one accountable person for the canonical line.
Weekly drift check (15 min): site ↔ retailer ↔ PR ↔ briefs; fix the first crack.
As you argue in [I], an accountable expert gives the answer a human shape. People still buy from people; the algorithm just introduces them. The AIO Author:
Signs the stance: safety line, who-for / who-not-for, one honest trade-off.
Speaks consistently: same tone across channels; pharmacist-plain over hype.
Faces feedback: when answers drift, the Author corrects the sentence, not the slogan.
The human element matters more, not less, when AI mediates first contact. Patients want to feel there’s real expertise behind what touches their skin daily. The AIO Author provides that credibility bridge—making algorithmic introductions feel trustworthy. AI is the next front door, but people still want to know who is answering when they knock.
Metrics That Matter (answers, not vanity)
Share of Answer — how often you’re named inside the answer when the ask fits you. Low = you’re absent where choices happen.
Time-to-Mention — days from publish/update to first appearance in an answer. Slow = your information is hard to find or hard to reuse.
Consistency Index — how many wordings you use for the same thing across surfaces (fewer is better). High = you’re teaching models (and people) to treat one product as many.
Patient-Fit Feedback — do the reasons given in answers match why people actually buy? Mismatch = your value prop is being retold by strangers.
These aren’t “AI metrics.” They’re clarity metrics. Improve them and the rest tends to follow.
Traditional KPIs—page views, rank—mean less when most interactions happen inside conversations that never click through. Share of Answer tells you if you exist where decisions happen. Time-to-Mention shows how quickly the Big Five can find and process your facts—lag often points to scattered or opaque information.
Consistency Index exposes drift: the same product described three different ways across site, retailer, and PR fragments your presence and confuses answers. Patient-Fit Feedback closes the loop: if people buy for comfort-under-makeup but answers praise you for “luxury feel,” you’re being recommended for the wrong reasons.
Think of these as your health bar:
Green: you’re quotable on contact; appearances rise naturally.
Yellow: answers mention you, but for fuzzy or inconsistent reasons.
Red: you’re missing from precise asks; someone else is speaking for you.
Optimize for communication clarity, not algorithmic tricks. AI is the next front door—these metrics tell you whether you’re visible when someone knocks.
Risk-First Hygiene (no heroics)
If you’re not sure, say so. Honest limits beat overreach.
Keep the safety line identical everywhere (one-breath, pharmacist-plain).
Don’t overfit to one platform; the Big Five shift often.
When a mistake appears in the wild, fix the sentence at its source, not the slogan.
Why This Is Urgent
AI is the next front door. If you’re not usable at that door, the conversation reaches for sources that are—often influencer comparison pages that speak in the sentences patients need. This isn’t a moral judgment. It’s physics: answers form from what’s easiest to assemble. Make your story the easiest truthful sentence to assemble, and you’ll keep showing up where it matters.
The window for shaping your narrative is narrowing. Every week you wait, competitors and influencers gain more opportunities to define you inside AI conversations. Once these patterns harden—in user behavior and model memory—reversing them gets exponentially harder. The brands that act now claim semantic territory that’s difficult to dislodge later.
This isn’t about perfecting everything before launch. It’s about establishing clear, consistent communication that the Big Five can reliably use. Start with your hero SKU, get the clarity stack right, then expand. The cost of waiting now exceeds the risk of imperfect early action.
This week — rewrite one product page so a human could quote it in one breath.
This article is informational and not medical advice.
Dr. Victor Gabriel Clatici, MD Originator of LLM Nutritionist • 30 Years in Dermatology • 20+ Years in Anti-Aging #ġ Bucharest, Romania | September 09, 2025
#LLMNutritionist #AIO #MedAIMark #B2A #SkincareAI #BeautyTech #AIOptimization
Sources & Context (at a glance)
Reply