What transfers, and what does not
The strength of retail in a treatment room is the recommendation: a person who has just examined the client's skin, hair or nails saying what to use and why. It is specific, timely and trusted, and it is the reason attachment rates in a good salon exceed anything a website achieves cold.
Online, that entire mechanism is missing. What remains is the brand, the client relationship and the knowledge. Building a shop that works means reconstructing as much of the recommendation as the format allows, rather than presenting a catalogue and hoping.
Online, the practitioner is absent. Everything the practitioner would have said has to be written down somewhere the client will actually read it.
The narrow version that works
For most independents, a defensible online retail business has four characteristics.
It sells mainly to people who already know you. Existing clients buying repeats, gift purchases, and people who found you through the service side.
It concentrates on repeats. The second purchase of something a client already uses requires no persuasion, only convenience. This is the highest-yield part of online retail for a service business and the most neglected.
It answers questions rather than listing features. Product pages built around what a practitioner would ask and say, not around the manufacturer's copy that every other stockist has also pasted in.
It does not attempt to compete for cold traffic on branded products. Searches for a specific product name are contested by the brand's own site and by national retailers with deeper pockets, better delivery economics and, frequently, lower prices.
| Asset in the room | Transfers? | What replaces it online |
|---|---|---|
| Practitioner recommendation | No | Product pages that answer what they would have said |
| Client trust | Yes | Kept by clarity, delivery and returns handling |
| Diagnosis at the chair | Partly | Guided selection, with the limits stated plainly |
| Immediate purchase | No | Repeat reminders timed to when it runs out |
| Sensory experience | No | Photography, texture description, sample options |
Source: Working model used by this paper, not a measurement.
The row that matters most is the first one. Businesses that do not replace the recommendation end up with a catalogue rather than a shop.
Product pages that do the practitioner's job
The single biggest quality difference between a good independent shop and a poor one is the product page. Manufacturer copy, duplicated across every stockist, gives a search engine no reason to prefer your page and gives a client no reason to buy from you.
What a page should contain, in roughly this order:
- Who it is for and who it is not for. The exclusion is the part clients find most useful and almost nobody writes.
- What it does, in plain terms, within the claims you are permitted to make for the product's classification.
- How to use it, including the practical details a practitioner would mention: quantity, order of application, what to avoid combining it with.
- What to expect and when, described carefully. Statements about results are claims and need to be substantiated in the way a consumer would read them.
- The questions you actually get asked, answered in the words clients use.
- Ingredients and mandatory information, accurately reproduced.
Google's guidance on helpful content is a reasonable test to apply here: would this page be worth publishing if search did not exist. A page assembled from supplier text fails that test, and a page written from the practitioner's actual advice passes it easily.
The technical floor
You do not need a sophisticated platform, but a few things are not optional.
Accurate product structured data. Where you mark up products, the markup must match what is visible on the page, including price and availability. Google publishes the requirements and the general rule that structured data must reflect visible content.
Correct handling of variants and duplicates. Multiple sizes and shades create near-identical pages. Decide the canonical structure deliberately.
Working search and filtering, because clients arrive knowing a category rather than a product name.
Speed on a phone, since that is where the traffic is. Image weight is usually the culprit.
Clear delivery and returns information before checkout, which is both a conversion issue and a consumer rights one.
| Item | Requirement | Why it matters |
|---|---|---|
| Product structured data | Must match visible price and availability | Mismatched markup can affect eligibility |
| Variant handling | One canonical page per real choice | Prevents near-duplicate pages competing |
| Mobile performance | Image weight controlled, sizes set | Most traffic, and the slowest conditions |
| Delivery and returns | Clear before checkout | Conversion and consumer rights obligation |
| Stock accuracy | Availability reflects reality | Overselling generates refunds and complaints |
Source: Requirements for structured data and content follow Google's published documentation; the rest is this paper's own framing.
Platform requirements change. Check current documentation at the time of build rather than relying on a summary.
Consumer rights when you sell at a distance
Selling online brings obligations that do not apply to a sale over the counter. Distance selling rules give consumers rights around information, cancellation and returns, and the Consumer Rights Act sets standards for goods sold to consumers. Certain categories, including sealed goods that are not suitable for return once opened for health protection or hygiene reasons, have particular treatment, and it is worth understanding exactly how that applies to what you sell rather than relying on a template returns policy copied from another site.
Publish the position clearly, apply it consistently and make sure your checkout gives the information required before the client commits. A returns policy that is unclear at the point of sale generates disputes that cost far more than the goods.
The channel that actually drives it
Paid acquisition for a small product catalogue rarely pays. The channel that does work is the one you already own: your client list, used properly. Repeat purchase reminders timed to when a product would realistically run out, restock notifications, and a small number of genuinely useful messages outperform frequent promotional sends by a wide margin.
That requires a lawful basis for marketing to those clients and a clean record of consent where consent is what you are relying on. The mechanics are covered in client data in a beauty business and the practice in email and retention.
Fulfilment, which decides whether any of it is profitable
Small parcels of liquids are an awkward category: they are heavy relative to value, they leak, and postage is a meaningful proportion of the order value. Three practical decisions.
Set a delivery threshold that reflects real cost. Free delivery on small orders converts well and loses money, which is a bad combination when it works.
Offer collection. For a business with premises this is the strongest option available: no postage, no leakage, and a client walking back into the building where the service is sold.
Test packaging in transit properly. Post a full order to yourself, and to somebody in another part of the country, before you launch.
Knowing when the shop is not worth it
If, after a full year, online sales are mostly your own clients buying repeats, that is not a failure. That is the business working as intended, and it should be judged on the margin and retention it produces rather than on its growth rate as a channel.
What is worth stopping is a shop that consumes staff time, holds stock for lines you do not sell in the room, and requires paid traffic to generate orders. That configuration transfers your service margin into someone else's advertising revenue.