Amazon agency for skincare and beauty brands, managed to net profit

Why does a generalist Amazon agency underperform on a beauty catalog?

Because it runs your catalog the way it runs a pet-supplies catalog: one blended ACOS target across every SKU, keywords pulled from head terms, and copy written by somebody who has never had a listing pulled for a claim. Specifically, here is what that costs you.

• One ACOS target across products with different margins. A $12 lip balm and an $80 serum do not share a break-even. Optimizing both to the same number overspends on one and starves the other, and the account average hides it from you for months.
• Shades and sizes treated as unrelated ASINs. Budget gets duplicated across siblings competing for the same shopper, and data that should pool into one signal is split into fragments too thin to act on.
• Keyword research that stops at the head term. Beauty buyers search by skin concern, ingredient and skin type: fragrance free, for oily skin, with niacinamide, non comedogenic. A generalist bids the category term and pays a premium for traffic that was never going to convert for you.
• Claims copy carried over from another category. Phrasing that is unremarkable on a cleaning product is what gets a serum flagged. Being right about keywords does not help once the listing is down.
• Returns and reviews read as noise. In beauty, shade mismatch and irritation drive returns, and one review saying it broke me out moves conversion more than any bid change will. That signal sits in the returns report and the review text, and it belongs in the ad plan and the image brief.
• Gifting handled in October. Sets and bundles are their own ASINs, with their own images and reviews starting from zero. Beauty seasonality is a catalog job months before it is a bidding job.

None of this is exotic. It is just category specific, and category specific is the whole reason to hire outside help.

What actually gets a skincare listing pulled?

The line that matters is cosmetic versus drug. A cosmetic changes how skin looks and feels. Language implying the product treats, prevents or cures a condition, or changes how the body works, reads as a drug claim - and that implication is the trigger, not any particular banned word list.

• Appearance language survives, medical language does not. Reduces the appearance of fine lines and treats wrinkles describe the same jar and are not treated the same way.
• Naming a condition is where the risk concentrates. Acne, eczema, rosacea, psoriasis, dermatitis. Positioning the product as the thing that resolves the condition moves a cosmetic into drug territory.
• Some of your SKUs genuinely are drug products in the US. Sunscreen with an SPF, acne products with an active, anti dandruff. Those carry documentation requirements a moisturizer does not, and the listing has to be built for that from the start rather than repaired after a flag.
• The claim is not only in the bullets. Images, A+ modules, the brand story and ad creative all carry claims. A compliant bullet under a non compliant infographic is still a non compliant listing.
• The restricted-term list is not published, and it moves. Copy that cleared review last year can be flagged this year, which is why claims are an ongoing check rather than a one-time launch task.

What a takedown costs is mostly downstream of the takedown. Sales velocity and organic rank decay while you appeal, review flow stops, ad spend keeps buying clicks to a dead page until somebody pauses it that day, and siblings in the variation family inherit the damage. Reinstatement then runs on Amazon timelines rather than yours.

We write claims to the policy rather than around it, and we would rather argue with you about one bullet than write an appeal. If something is already down, start with the suppressed listing fix.

What does broken shade and size parentage cost you?

Reviews, rank and conversion, in that order. A shade family split into separate parents divides its review count, its sales history and its ad budget across ASINs that then bid against each other for the same shopper.

• Reviews pool across a correct family. Twelve shades carrying 40 reviews each look far weaker than one family carrying them together, on a page where review count is the conversion lever.
• Shoppers expect a swatch row, not a second search. If the deeper tone is not selectable on the page they landed on, most of them do not go back to results for you. They go to the competitor whose shades are all on one page.
• Size variations decide where your traffic lands. Travel, full size, refill, value set. The child that shows by default is not always the one you want to pay to send traffic to, and a mispriced value size quietly cannibalizes the hero SKU.
• Not everything belongs in one family. Merging genuinely different formulas so they inherit reviews is a variation relationship problem, and it also puts reviews about a different product on the page you are buying clicks for.
• Ads follow parentage. Fixing the family changes what your campaigns are actually buying, so the catalog fix and the campaign change happen in the same week, not as two projects that surprise each other.

Which gates does a beauty catalog have to clear before any of this matters?

Beauty carries operational gates most categories do not, and each one can stall a launch for weeks. They are worth knowing before you build the plan, not after.

• Beauty and Topicals are separate approvals. Clearing one does not grant the other, and actives-led skincare is usually the harder of the two. Finding that out mid-launch costs you the launch window.
• Approval runs on documents, not persuasion. Amazon asks for things like supplier invoices, brand authorization, safety documentation and a certificate of analysis. Each rejected application burns another documentation cycle with your manufacturer, so the first application is the one worth getting right.
• Some beauty SKUs are dangerous goods. Amazon names perfumes and deodorant sprays, nail polish, alcohol-based products, aerosols, essential oils and hair dyes. They need safety documentation and a review before FBA, and storage limits can apply - which changes peak-season stock planning for that SKU specifically.
• Expiry dating is a cash question, not a labelling one. Shelf life is counted from manufacture, not from arrival, so a slow container plus a slow-turning SKU can put stock at risk while it is still perfectly saleable. That makes it a reorder-quantity and lead-time decision.
• Premium Beauty is invitational, and most scaling private-label brands do not qualify yet. It generally expects an established presence beyond Amazon. If you are Amazon-and-DTC only, the honest answer is not yet, and we will tell you what would change that instead of selling you a plan around it.

We check these in the audit rather than discovering them in month two. Where a specific requirement turns on your paperwork or your account, we verify it against your own Seller Central rather than quoting a number from a blog.

What does Amazon PPC look like on beauty margins?

Your break-even ACOS is your gross margin after cost of goods, Amazon fees and returns - per SKU, not per account. Above it you are buying rank. Below it you are buying profit. Both are legitimate spends; the failure is not knowing which one a campaign is doing. Work yours out with the break-even ACOS calculator.

• Per SKU, not blended. Beauty catalogs run a wide margin spread on purpose: a hero serum, a cheap entry product bought to win a first order, a gift set. A single target across all three fits none of them.
• Returns belong inside the margin. Shade mismatch and irritation returns are a real line item in this category and they push break-even down. Leave them out and every ACOS target you set is too generous.
• Conquesting is expensive by design. Bidding competitor brand terms in beauty means paying for traffic a defended brand is also paying to keep. It can be worth it, but only as a named acquisition budget with a ceiling and a repurchase assumption you are willing to write down and check later - never buried inside a blended ACOS number.
• Repeat purchase changes the math, and it is the easiest number to lie to yourself about. Skincare is consumable, so a first-order ACOS above break-even can still be profitable. That is only true if your own repurchase data says so. A category benchmark somebody quoted at a conference is not your data.
• Seasonality gets planned, not reacted to. Holiday gifting, Prime Day, SPF into summer, and the January reset. Budget and bid changes go in ahead of the demand curve and get logged, so in January you can separate what worked from what was simply December.

How do we actually work with you?

It starts with the free audit, and the audit is deliberately the proof. Read-only access, three business days, and a report that sizes what your catalog is leaking in dollars per month. You keep the report whether or not you hire us.

1. The audit. Wasted ad spend per SKU, break-even ACOS by ASIN from your real margins, a listing and claims check, and your shade and size variation families read line by line. Top three opportunities, sized, walked through live.
2. Scope in writing. What is included and what is explicitly out. Vagueness here is what becomes a dispute in month three.
3. One senior strategist on the account. The person who runs it is the person you spoke to. No hand-off after signing.
4. A change log. Every bid move and every listing change with a date and a reason, so when profit moves in week six we can both see which change did it.
5. Reporting in net profit. Attributed revenue labelled as attributed. Ad-attributed revenue is not incremental revenue, and we say so before you have to ask.

What we do not do: route your ad spend through our account (it stays on your own card), promise a rank or an ACOS, or ask for account-owner credentials.

One thing we will not do either: show you somebody else results. We have no published case studies, because client numbers get published only after that client has signed off in writing on those specific figures. Until then the honest proof is the audit of your own account, not a screenshot you cannot verify.

Who is this not for?

A niche you will not enforce is not a niche. We turn work down, and it is faster for both of us if you can rule us out here.

• Under roughly $50k a month. A retainer is hard to earn back at that revenue. You are better served fixing the listing yourself and running a simple campaign structure until the numbers support help.
• Well above $500k a month. At that scale you usually need an in-house owner plus specialists on call. We would be a bottleneck, and we would rather say it now.
• Looking for category-specific expertise we do not have. Beauty is where our category knowledge runs deepest. We work with strong private-label brands outside it - supplements, wellness, pet, home - and the operating discipline travels. What does not travel is the compliance and catalog detail on this page, so if your category has its own regulatory maze we will say what we know and what we would be learning.
• Marketplaces we do not run. We run US, UK and EU. Outside those, we are the wrong shop - and we will say so rather than learn a marketplace on your budget.
• Accounts with an unresolved suspension or IP complaint. Reinstatement is not a service we sell. Get the account healthy, then talk to us.
• Anyone wanting a guarantee, or rank bought in a way Amazon does not allow. The answer is no, and you will get it on the first call rather than after the invoice.

And if the audit finds your account is in decent shape, that is what the report will say. There is no pitch attached to that outcome.

Where to go next

• Amazon PPC management for beauty brands
• Listing optimization and compliant claims
• Break-even ACOS calculator
• The free Amazon account audit