How to sell skincare online as a small business

You've been making a body butter for your sister and her friends, three of them have asked to buy it, and someone has already said "you should sell this properly."

Before you do: make one product, keep water out of it if you're working from a kitchen, find out what your country requires before the first sale rather than after it, put a batch code on every single unit, and sell thirty by hand before you order five hundred of anything. The reason for that last one is arithmetic. A custom jar with your name moulded into it usually starts at a 500-unit minimum, and at even $1.20 a jar that's $600 sitting in your hallway before you've bought a single ingredient.

Skincare is one of the most rewarding small businesses to start and one of the easiest to start badly, and the difference is almost entirely in the boring parts.

What you say about it decides which rules apply

This is the sentence to internalise, because it catches nearly everyone.

In most regulatory systems, the line between a cosmetic and a medicine is not what's in the bottle. It's what you say the bottle does. A cream that "softens and hydrates dry skin" is a cosmetic. The same cream, with the same formula, marketed as something that "treats eczema" or "heals acne" or "repairs the skin barrier" has been claimed as a drug, and the rules for drugs are heavier by an order of magnitude: different approvals, different evidence, different consequences.

Words that reliably move you across that line: treats, cures, heals, repairs, prevents, anti-inflammatory, and any named condition. Acne, eczema, psoriasis, rosacea, dermatitis. Also anything about changing the structure of the skin rather than its appearance.

Words that generally stay on the cosmetic side: cleanses, softens, moisturises, soothes the feel of, improves the look of, adds shine.

Two traps that catch people who have been careful everywhere else:

The reposted review. A customer messages you to say your oil cleared her acne in two weeks. You screenshot it and put it on your page. That is now your claim, not hers, and it is a claim about treating a condition. Thank her privately. Don't post it.

The ingredient explainer. You write a nice post about how salicylic acid works on acne, and you link it to your product. You've connected the claim to the product even without putting it on the label.

None of this is legal advice and I'm not going to name you a statute, because the specifics differ between the US, the UK, the EU, Nigeria, India and everywhere else, and the details change. What's consistent is the shape of it: check what your national regulator requires of a cosmetics seller before you sell, and in most places that means some combination of registering your business or your facility, listing the products you make, holding a written safety assessment, and labelling to a set format. Find your regulator's own page. Not a summary on a blog, including this one.

Water is the decision that shapes everything else

If you take one technical thing from this article, take this.

Anything containing water grows things. Bacteria, mould, yeast. It will happen in a matter of weeks in a jar sitting in a warm bathroom, and often you cannot see it. Preventing it needs a proper preservative system, tested, at the right pH, in the right proportion. Vitamin E is not a preservative. Neither is grapefruit seed extract, and neither is putting it in the fridge.

So there are two paths out of a home kitchen, and only one of them is comfortable.

Anhydrous products. Balms, body butters, face oils, cleansing balms, scrubs made with oil and sugar or salt, lip products, solid perfumes. No water, so no preservative system required, though you still want an antioxidant to slow rancidity and you still need to be clean. Shelf life is usually a year or so, set by the oils.

Emulsions and anything with water. Lotions, creams, toners, gels, most cleansers, sheet masks. These need a preservative, and doing it properly means understanding pH, knowing what your preservative is and isn't effective against, and ideally getting the finished product tested. This is a real skill and it costs real money.

Almost every small brand that gets into trouble does it by making a lotion at home without a preservative because it "only has natural ingredients". Start anhydrous. It's not the beginner option, it's the sensible option, and some of the best-selling products in the category are balms.

If you want something water-based, the honest route is a contract manufacturer or a private-label supplier who has already solved the preservation and testing. That comes with the 500-unit minimum and it comes with a formula you didn't write. Trade-off, not a failure.

Batch codes, records and the recall you hope never happens

Every unit you sell gets a batch code. Every batch you make gets a record. This takes about four minutes per batch and it's the difference between a manageable problem and a catastrophic one.

A batch record holds:

• The batch code, printed or stickered on every unit from that batch
• The date you made it
• Every raw material used, with the supplier's own lot number for each
• How much you made, and how many units it filled
• Any observations. Separated slightly, smelled different, ran thin

The point of the supplier lot numbers is that if a supplier tells you a shipment of butter was contaminated, you can work out in ten minutes exactly which of your batches used it and exactly which customers bought from those batches. Without that, the only safe answer is to contact everyone who has ever bought from you.

Put a best-before date or a period-after-opening on the label too. "Use within 12 months of opening" is not decoration; it's what tells a customer their year-old jar is past it rather than defective.

Patch testing, and what to do when someone reacts

Tell every customer how to patch test, in the same words, in the product description and on the card that goes in the box:

Put a small amount on the inside of your forearm, leave it 24 hours, and don't use it on your face if you see any redness or itching.

Some people will react to something in your product no matter how gentle it is. Essential oils are a common culprit, and so are fragrance, lanolin, and a handful of botanical extracts that sound harmless. Reactions are not proof you did something wrong. They are, however, entirely your problem to handle.

When someone reports one:

1. Reply within the day. Ask them to stop using it and to see a doctor or pharmacist if it's more than mild redness.
2. Ask for the batch code and a photo.
3. Refund without arguing. Always. The refund is cheaper than the thread.
4. Write it in your batch record. Two reports from the same batch is a signal, not a coincidence.
5. Don't diagnose them. You're not qualified and it's the fastest route to making the situation worse.

And carry product liability insurance. It's the single most obviously worth-it expense in this business, and small makers routinely skip it.

What it actually costs to make thirty units

Real numbers for a 100ml cleansing balm, hand-poured, small batch:

Line	Cost per unit
Oils, butters, wax	$2.40
Antioxidant, fragrance or essential oil	$0.55
Jar, 100ml, bought in 100s not 500s	$1.35
Label, printed at home on decent stock	$0.22
Muslin cloth in the box	$0.60
Outer box and packing	$0.85
Materials total	$5.97

Then your time. Thirty units of balm is a melt, a pour, a cool, a wipe-down of every jar, labelling, batch coding, boxing. Call it four hours for the batch, so eight minutes a unit. At $25 an hour that's $3.33.

So the real cost is around $9.30, and a $28 retail price gives you $18.70 before payment fees, shipping and the cost of getting anyone to look at it. That sounds healthy, and it is, until you divide the $600 of jars you bought and the $340 of ingredients into the four units you sold in week one.

The way this business kills people is inventory, not margin. Make thirty. Sell thirty. Then make sixty. There's a longer treatment of costing your hours properly in how to price what you make, and it applies here more than most categories because the hands-on time per unit is so easy to underestimate.

Worked example: Nadia, cleansing balm, Minneapolis

Nadia makes one product. A 100ml cleansing balm, $28, in a glass jar, with a muslin cloth.

She started by making 24 units and selling them to people she knew and to two Saturday markets. Cost her about $290 in materials for the first run because she bought jars in 100s and had 76 left over. She sold the 24 in five weeks at $28, which is $672.

What she did next is the useful part. Instead of adding a face oil and a toner and a scrub, she made the same balm again, twice, and spent the time on photographs and on a customer list. By month six she was selling 60 a month, all of one product, all through Instagram and a shop link, taking payment by card and by transfer.

Her costs at 60 units a month: $5.10 a unit on materials because the jars now come in 500s, about eight hours of her time, $180 a month on shipping supplies and postage after charging most of it to customers. Revenue $1,680. She pays herself about $900 a month from it and reinvests the rest.

The toner came in month eleven, after a customer survey, and it came from a contract manufacturer because she wasn't going to preserve a water-based product in her kitchen. She was right not to.

The honest limitation

Sailo will let you type anything you like into a product description. There's no compliance checking, no ingredient list validator, and nothing that will warn you when the sentence you just wrote about your moisturiser has quietly reclassified it as a drug in your market. The same is true of every shop platform, but it's worth saying out loud in an article about a regulated category: the software is not a safety net.

What it does do is hold the ingredient list, the batch code note, the patch test instruction and the shelf life in a place buyers can read before they order, which is where they belong rather than in a story that vanishes in 24 hours. Writing those sections so they're actually read is a craft in itself and it's covered in writing product descriptions that sell.

On money: Sailo takes nothing at all on bank transfer, cash on delivery or chat orders, because it never touches the funds on those rails. Card payments cost 3% of the goods on the free plan and 1% on Business. On 60 orders of $28 that's $50 of commission on free and $17 on Business, and a subscription worth paying. On six orders it isn't. Do that division before you upgrade.

Shipping a product that melts

Two seasonal facts that catch every balm and butter maker exactly once.

A cleansing balm sitting in a delivery van in July will melt, re-solidify grainy, and arrive looking wrong. It's still fine to use, and the customer will not believe you. Either put a line in the listing between May and September saying it may soften in transit and that it's harmless, or pause shipping to hot regions, or ship with the expectation that you'll replace some.

Glass jars break, and they break at the shoulder where the box corner hits the floor. Bubble wrap the jar itself, not just the void space, and use a box that doesn't let the jar move at all. The full method for parcels that survive a bad courier is in packaging that survives the courier.

If your interest is fragrance rather than skincare, the rules and the economics are different enough to be a separate trade, and that one is covered in selling perfume and oils.

What to do next

Open your national regulator's own website today and find the page for cosmetics or personal care products. Read what a small manufacturer has to do. If you can't find it in twenty minutes, ring your local business support or trading standards office and ask them who regulates cosmetics where you are. That answer changes everything downstream, including whether you can sell the thing you've already made.

Then make one batch of one anhydrous product, thirty units, with a batch code and a written record. Sell those thirty before you buy anything in five hundreds.

And if you haven't sorted out whether you need to be registered as a business at all, do that in the same week: do you need to register your business covers the general question, and in this category the answer arrives earlier than in most.