Brand growth

How to reduce skincare returns with personalised recommendations

Wrong-product returns and bad reviews often start with guesswork at the shelf. How personalised skincare recommendations help beauty brands keep customers and protect their name.

11 November 2026 · 7 min read · By the Claré AI team

Every beauty brand knows the sinking feeling of a return with a note that says the product did not suit my skin. The money is the small part. The bigger loss is the customer. Someone who might have bought from you for years decides the brand is not for them, and tells a friend. This post is about the quiet cost of guesswork and what you can do about it.

Why the wrong product gets bought in the first place

Most shoppers do not know their skin well. They know it feels oily by afternoon, or that something looks red, but they cannot match that to an ingredient list. So they guess. They pick by packaging, by a creator’s recommendation, by a friend’s tip or by price. Beauty e-commerce makes this easy to do and hard to undo, because the shopper cannot test the product before paying.

Guesswork looks like a sale until it does not

A guessed purchase shows up in your numbers as a conversion. Two weeks later it shows up as a complaint, a return or a quiet decision never to reorder. The sale was real. The customer was not won. If you only measure the first order, you cannot see this loss, which is why it survives for years.

Where the blame lands

When a product does not suit someone, they rarely say I chose badly. They say this brand did not work for me. Fair or not, the name that takes the hit is yours. A bad reaction becomes a review, a review becomes a doubt in the mind of the next shopper, and the cost compounds. Advice at the moment of choice is worth more than another discount.

What personalised recommendations change

  • The shopper is told what is visible on their skin today, in plain language, so the choice starts from facts.
  • The routine comes from your catalog, so every suggestion is something you stand behind.
  • Allergens the shopper declares remove products outright, so the obvious mistakes never reach them.
  • Each pick comes with a reason, which turns a recommendation into advice.

This is the design behind Claré: five questions, one selfie and a routine of four products from your own range. The point is not to sell more of everything. It is to help each shopper buy the right thing first, which is the thing most likely to be bought again.

Give people a reason to come back

Not everyone buys the moment they finish. A shopper may be on a phone on a bus, or standing in a shop with a friend waiting. That is why the report matters. It is a two-page PDF with their results and the recommended products, each with a buy link back to your store. The sale can happen tonight or next week, and the shopper remembers who helped them. You can see what it looks like in our sample report.

How to measure it honestly

Be careful with promises, including ours. We do not publish conversion figures, because we do not have your numbers. What you can do is a simple comparison on your own store.

  • Tag returns and complaints that mention skin not suiting the customer, for one month.
  • Check how many of those customers had any guidance before buying.
  • Run a personalised flow for a defined set of products and compare repeat orders after sixty to ninety days.

General guidance on measuring and improving a site experience, including how people behave on product pages, is covered well in Google’s own documentation, and the same habits of measuring before and after apply here.

A simple way to start

Three habits that make personalisation work

Tools do not fix a business by themselves. These habits turn a personalised flow into something that actually changes outcomes.

  • Put the flow where the doubt is. Add it to product pages and category pages that get many visits but few orders, not just to a hidden menu.
  • Make the reason visible. A recommendation with a one-line why is more convincing, and more honest, than a list of names.
  • Follow up on the report. A shopper who downloaded their results is warmer than a cold visitor, so treat them that way in your email and ad strategy, with permission and respect.

None of this requires a big redesign. It requires deciding that a customer who buys the wrong product is a failure of advice, not a failure of the customer, and putting effort where the advice is thinnest.

Look at your last month of returns and reviews. How many mention the product not suiting their skin? That is the group a personalised routine is built for. Then read about the cost of guessing for brands, and when you are ready, get early access for your brand.

Frequently asked questions

Does personalised skincare reduce returns?

It is designed to, because shoppers are guided to products that fit their skin and declared allergens are excluded. We do not publish return-rate figures, so measure the effect on your own store.

What does the customer take home?

A PDF report with their results and the recommended products, each with a buy link back to your store. It is built in their browser and contains no photo.

Cosmetic analysis only. Not medical advice.

Ready when you are

See Claré on your catalog.

Book a demo and we will walk through a scan, the checks and a routine built from a sample catalog, then talk about how it fits your brand.