How Vereme works
Vereme turns appearance goals into grounded, cross-domain recommendations. This page explains the rules: what shapes a recommendation, how the business model stays aligned with your interests, and what “good” looks like inside the product.
How Vereme makes recommendations
Brand-neutral by design. No brand pays Vereme to be placed, ranked, or pushed. If a product shows up in an answer, it earned its way on fit for you — not on sponsorship.
Vereme spans 18 appearance domains — skin, hair, colour, makeup, nails, fashion, accessories, eyewear, fitness, diet, posture, fragrance, cosmetic dental, procedures, tattoos and piercings, sleep, and body hair — so a single recommendation does not pretend your life happens in one aisle. There is no separate public domains index yet; the canonical list always tracks what you see inside the product.
Recommendations are multi-factor: your skin, hair, and eye colour where relevant; body shape and fit preferences; lifestyle and time budget; spend comfort; climate; goals; allergies and intolerances; and what you already own. The point is optimisation under real constraints, not a generic routine card.
Domains are wired to each other on purpose. Posture changes how skincare sits on your neck and chest; diet and recovery change hair density and stress flares; sleep changes under-eye behaviour and barrier repair. Siloed advice reads confident and wrong; cross-domain logic is how Vereme stays honest about trade-offs.
The Pareto solver is the pragmatic layer: identify the minority of changes that move the majority of your confidence — the order-of-operations wins before the tenth product swap. It is how the product avoids vanity churn.
Signals come from published research and specialist consensus where that exists, from image models where you upload photos, from Vereme’s curated product and ingredient intelligence, and from what you tell Vereme plus what you mark as working or failing over time.
Aggregated insights
Vereme combines anonymised patterns across users — for example, what tends to work for people in similar routines, how products correlate with confidence shifts, and broad trends by cohort. These outputs are always aggregated; we apply minimum cohort sizes (typically 50+ distinct accounts in product-facing surfaces, and higher floors where licensing applies) so individual answers cannot be singled out.
Aggregated insights are never sold to brands to influence rankings. Our recommendation engine does not read partner licensing pipelines, and commerce surfaces do not steer you based on who paid for data access. Brand neutrality is structural.
Your control
You can opt out of contributing to new aggregates anytime in Settings under Privacy & data. When you opt out, your account stops feeding new cohort and pooling queries going forward. Existing aggregates that were already computed are not regenerated to remove historical contributions — that would require rebuilding every statistic on every toggle — but you will not appear in new ones. If you opt back in, you are included again from that point onward.
For data export, correction, and deletion rights, see our privacy policy.
How we make money
Subscriptions are the core business — Free, Plus, and paid Pro positioning as voice-forward and stylist-grade workflows roll out. That recurring revenue funds engineering, safety review, and editorial quality.
Affiliate relationships exist on some retailer destinations — Amazon Associates, Sephora where applicable, and similar programmes. Those links never change whether Vereme recommends an item. They only change where you choose to complete a purchase.
Vereme Verified certification is planned as a merit gate for formulas and devices — evidence-led, not pay-to-play. Brands do not buy placement inside Vereme Verified; they qualify or they do not ship under that mark.
What will never happen: taking brand money to reorder rankings, selling conversational data for ads profiling, or stuffing third-party ads into the coaching surfaces.
When an affiliate link appears, you see the retailer name beside it. Clicking simply routes checkout; it does not rewrite the recommendation logic behind the scene you already read.
See our most recent quarterly disclosure
How recommendations are evaluated
Nothing ships on a recommendation surface until it passes internal review — rubric checks, contradiction passes, and domain-lead sign-off where science or safety is involved.
Automated conflict checks flag obvious clashes — for example skincare actives against a stated allergy or routine steps that fight each other across categories.
Where the UI exposes feedback after an interaction, Vereme treats thumbs and optional notes as ground truth about usefulness. Those signals feed periodic recalibration: prompts, classifier thresholds, and merchandising rules adjust when outcomes drift.
What Vereme is not
Vereme is not a medical practice or psychiatric service. It is not a substitute for a dermatologist, physician, therapist, dietitian, or any licensed clinician when you need diagnosis, prescribing, or crisis care.
Vereme is not an industry insider desk copy — it was built independently by the founder without holding periods or sponsorship hooks from beauty conglomerates. Background on why that matters lives on who we are.
Vereme is not omniscient. When model confidence drops, the product says so. Correct Vereme bluntly when it misreads you; corrections train the personalised layer over time.