Industries
An AI beauty advisor that knows your products
AskMerra works as an AI beauty advisor for skincare and cosmetics stores. It asks about skin type and budget, recommends products you stock at today's price, and stays careful when a question touches on health.
Why a beauty store needs an AI beauty advisor
A shopper in a cosmetics store rarely knows the product name. She describes her skin: dry and a bit sensitive, oily with blackheads, red around the nose. In a physical shop, a consultant asks two questions and walks her to the right shelf. Online she lands on a category page with forty serums and a filter sidebar that expects her to know what niacinamide does.
Her question is a combination. Oily skin, a budget of €30, nothing scented, something for the evening: that is four filters, and most shoppers don't know which of your attributes match which of their worries. Some leave. Others email your team.
An AI beauty advisor fills that gap. AskMerra reads your catalog, asks one short question when a request is too vague, and recommends up to three products you actually stock, with live prices and stock on the product cards. It answers in English, Romanian, Italian, French, German or Spanish, in whichever language the shopper writes.
Questions shoppers ask, and what the assistant does
Typical questions in a skincare shop, and how AskMerra handles each one (an illustration, not results from a real store):
| Shopper asks | What the assistant does |
|---|---|
| "My skin is dry and a bit sensitive. What should I use in the evening?" | Finds products tagged for dry and sensitive skin, suggests a short routine from your catalog and mentions a patch test for anything new. |
| "Which vegan products do you have under €30?" | Treats €30 as a hard limit and recommends only products your catalog lists as vegan. |
| "Is this serum fragrance-free?" (asked on a product page) | Knows which product the shopper is looking at and checks its description, attributes and ingredient list before it answers. |
| "I want a simple routine for oily skin." | Picks products from different routine steps (cleanser, serum, moisturizer) that address the same concern. |
| "Is the retinol serum safe during pregnancy?" | Gives no verdict and suggests asking a doctor or pharmacist. |
| "Can I return a cream I've opened?" | Answers from the returns policy you added, and offers to pass the question to your team when the policy doesn't cover it. |
Here is a short, illustrative exchange on a fictional shop. The product names are invented.
Shopper: My cheeks get red easily. Is there anything calming, under €25?
Assistant: For redness, the Calm Cica Serum (€19) and the Barrier Repair Cream (€22) are both fragrance-free and in stock. Since your skin reacts easily, try the serum on a small patch first. If the redness stays or gets worse, it's worth seeing a dermatologist.
(Two product cards appear under the answer, with the live price, a stock badge and a link to each product page.)
For a longer walk-through of skincare conversations, read our guide to AI skincare recommendations.
How answers stay grounded in your catalog
An advisor that invents a product or quotes last month's price does more harm than a filter sidebar. AskMerra avoids both:
- Only your products. Recommendations come from your catalog. Product ids in the AI's answer are checked against the products it was given, and anything else is dropped.
- Live prices and stock. Prices, sale prices and stock on the cards are read from your catalog, never from the model's memory. Out-of-stock products are not recommended.
- Budgets are hard limits. "Under €30" or "sub 100 lei" becomes a price filter, and the assistant never recommends a product above a stated budget.
- Claims you can stand behind. Properties like vegan, cruelty-free or non-comedogenic come from your data, and they are checked against the product's own ingredient list: a "fragrance-free" product that lists Parfum, or a "vegan" one that lists beeswax or lanolin, loses that tag.
- Your policies, your words. Shipping, returns and payment questions are answered from the documents you write in the dashboard.
The details are in how the assistant answers and in our article on answers grounded in your catalog.
Careful with health topics
Skincare sits close to health, so the assistant holds back on purpose. It is told plainly that it is not a doctor, dermatologist or pharmacist. It never diagnoses and never promises results: no "cures", "treats", "heals" or "eliminates". For persistent, severe, painful or worsening problems, such as severe acne, rashes or an allergic reaction, it recommends seeing a professional. Questions about pregnancy or allergies get the same answer. When it suggests a new product for sensitive or reactive skin, it mentions a patch test.
The same rules apply when AskMerra reads your products. The enrichment step is told to use only facts present in your product data and never to invent ingredients, certifications or health effects. That protects your shoppers, and it keeps claims off your site that you never made.
Product data that makes the advice better
When you import your catalog, AskMerra reads each new or changed product and records what shoppers ask about. For a beauty store, that means:
- Skin types: oily, dry, combination, normal, sensitive, mature.
- Concerns: acne, blackheads and pores, oiliness, dryness, redness, signs of ageing, pigmentation, dullness, dark circles and sun protection, plus hair, body, lip and oral care needs.
- Key ingredients: up to eight, as written in your data.
- Routine step: cleanser, toner, serum, moisturizer, SPF, mask, exfoliant, eye care and more, which lets the assistant put together a routine and lets the dashboard show which products pair well.
- Properties: vegan, cruelty-free, fragrance-free, non-comedogenic, hypoallergenic, dermatologist tested, travel size, gift set and others.
The more your product pages say, the better the answers. Mention skin types, concerns, key ingredients and how to use the product. If your export already has columns such as skin_type or ingredients, they are suggested as product attributes during import. You can inspect and edit every product's tags in the dashboard, and your edits are kept when the product is enriched again. See the catalog docs for the full list of fields.
Setting up AskMerra for a cosmetics store
- Create your shop and choose "Beauty & cosmetics" as what you sell. The widget starts with suggested questions such as "What do you recommend for dry skin?" and "Help me build a skincare routine". You can change them, up to six per language.
- Import your catalog from a CSV, JSON or XML file, from the product feed you already send to Google Shopping, or through the Push API.
- Add your knowledge: shipping, returns, payments and FAQ entries such as "Are your products cruelty-free?". AskMerra can draft the policy texts from your website, and it can read the skincare guides on your blog and link them in answers.
- Test in the Playground with the questions your customers really ask.
- Paste one line of code before the closing body tag. Install the widget covers Shopify, WooCommerce, Magento 2 and Google Tag Manager, and platforms lists the catalog options for each. Platform guides: Shopify, WooCommerce, Magento 2 and PrestaShop.
You can import the catalog, write the knowledge base and design the widget before you choose a plan. AI features (answers, product enrichment, the Playground and the import from your website or blog) start once you choose one, and your first plan begins with a 14-day free trial.
What it costs
Plans start at €29 a month for Start, then €49 for Basic, €79 for Growth, €129 for Plus, €199 for Pro, €349 for Scale and €599 for Business, excluding VAT. The plan fee is AI usage included every month. Above it, you pay as you go at the same rates from prepaid credits, and you can set a monthly spending cap or switch pay-as-you-go off. Start includes two languages, Basic three, Growth and Plus four, and Pro and above all six. Live chat with your team is included from Growth.
Your first plan starts with a 14-day free trial: you add a card at checkout, and if you cancel before the trial ends, you pay nothing. Otherwise the plan you chose is charged when it ends. During the trial you get Pro features (live chat, all six languages, no AskMerra branding) and €10 of included AI usage, about 250 conversations, with no pay-as-you-go. There is one trial per account and per shop domain. Compare the plans on the pricing page, start your 14-day free trial, or book a demo and we will load a sample of your products so you can ask the questions your customers ask.
Frequently asked questions
Does the assistant give medical or dermatological advice?
Can it tell shoppers that a product is vegan or fragrance-free?
Does it analyse photos of the shopper's skin?
Will it recommend products that are out of stock?
Which languages does it speak?
Does it work with Shopify and WooCommerce?
Related pages
- IntegrationsAI Chatbot for Shopify: A Shopping Assistant for Your StoreAn AI chatbot for Shopify that recommends your real products with live prices and stock, in six languages. Install it today with one line of code.Read more
- IndustriesAI Chatbot for Fashion Stores: Sizing and Product AdviceAn AI chatbot for fashion and apparel stores: outfit ideas for an occasion, size answers from your own size guide, live stock and clear answers on returns.Read more
- IndustriesAI Chatbot for Electronics Stores: Specs and Product AdviceAn AI chatbot for electronics stores: it compares products on the specs you publish, checks stated compatibility, keeps to budgets and explains your warranty.Read more