From “something for oily skin” to a full routine.
Beauty shoppers search by problem, ingredient and shade. AACSearch understands negations and attributes, answers ingredient questions from your product pages, and recommends the next step of a routine.
Queries plain keyword search gets wrong
крем для жирной кожи без спиртаskin type + negationниацинамид сывороткаingredientтональный крем светлый тёплыйshade familyspf для чувствительной кожиattribute + needчто нанести после кислотroutine questionWhere search loses customers today
“Without” is ignored
“Без спирта” returns products with alcohol because the word matched. Negation is one of the most common beauty intents.
Ingredient vocabulary
INCI names, trade names and translations differ. Shoppers rarely type what the label says.
Shade matching
Shade names are marketing; shoppers search by tone and undertone.
Repeat recommendations
“Similar items” shows five versions of the same moisturiser instead of the cleanser and SPF that complete a routine.
How AACSearch solves it, layer by layer
Each card names the capability behind it. Items marked “in development” are designed and scheduled, not live yet.
Finds the right item from the first letters — typos, synonyms, filters, meaning.
Attributes and negations
Skin type, ingredient exclusions and price phrases become filters, not competing words.
Widens the basket: similar, frequently together, personal.
Personal shelf
Ranking adapts to the brands and categories a visitor engages with; history decays so a one-off gift doesn't reshape the profile.
Diverse suggestions
MMR diversification keeps recommendation rows from filling up with near-duplicates.
Answers questions a filter can't, citing your own catalogue and docs.
Ingredient and usage answers
“Can I use retinol with vitamin C?” — answered from your product pages and articles, with citations and without medical claims beyond the sources.
Learns how items, queries and topics connect — and leads the visitor to the goal.
Routine builder
Will learn routine sequences (cleanser → toner → serum → SPF) from purchase order and suggest the next step, not another serum.
Concern clusters
Will group products, ingredients and articles into concerns like “acne-prone” or “sensitive” for guided navigation.
One visitor, from first query to goal
What the visitor does on the left, what AACSearch does on the right.
- 1
Problem
Searches “крем для жирной кожи без спирта”.
Applies skin type as a filter and excludes alcohol-containing products.
Natural-language search - 2
Learn
Asks whether niacinamide can be combined with acids.
Answers from the store's articles with a link to the source.
AI answers with citations - 3
Routine
Adds a serum to the cart.
Suggests the SPF and cleanser that usually complete this routine.
Product graph· in development
Your first two weeks
A typical rollout for this industry. Most teams are live on day one and spend the rest of the time measuring and tuning.
Day 1
Import products with skin type, ingredients and volume as fields.
Day 2
Add ingredient synonyms (INCI and trade names).
Week 1
Index articles for the assistant; check it refuses medical questions it can't source.
Week 2
Enable personal ranking for returning visitors and compare repeat purchases.
What to measure
The numbers to judge a rollout by. Compare them with an A/B split or a before/after window on the same traffic.
- Search conversion
- Purchases per search session.
- Average order value
- Routines raise basket size more than discounts do.
- Repeat purchase rate
- Personalised ranking should bring returning visitors to their products faster.
Connects to
- Shopify
- 1C-Bitrix
- WordPress / WooCommerce
- YML / CSV feed
- REST API
Capabilities used
Coming next
Questions
Search for Beauty & cosmetics, live this week
Connect a feed or a site, embed one script, and measure the difference on your own traffic.