# Alfreya > Alfreya makes AI tools for fashion and jewelry stores on any platform: search that forgives typos and Turkish suffixes, outfits built from stock, try-on to scale, model photos from a single packshot, and a photo lookup for store staff. ## The products ### Discovery & Search The search box, the product page, and what goes next to the product. - [Smart Search](https://www.alfreya.com/md/en/solutions/smart-search): A search box that forgives typos, plurals and a second language. (Live with Vitruta) - [AI Shopping Assistant](https://www.alfreya.com/md/en/solutions/ai-shopping-assistant): Search by sentence, by photo, or by “same but in black”. (Live with Vitruta) - [Visual Similar Products](https://www.alfreya.com/md/en/solutions/similar-products): Its lookalikes are one scroll down. AI found them in your catalog. (Live with Vitruta) - [Outfit Recommendations](https://www.alfreya.com/md/en/solutions/complete-the-look): Two colours, six outfits. AI styled each one from your stock. (Live with Vitruta) - [Collection Merchandising](https://www.alfreya.com/md/en/solutions/smart-merchandising): New arrivals move up, broken size runs step back. Twice a day, each move explained. (Live with Vitruta) ### Try-On & Content Products on the shopper, and products on a model, without booking a shoot. - [Virtual Try-On](https://www.alfreya.com/md/en/solutions/virtual-try-on): The shopper’s own photo, wearing the jacket from this page. (Available to deploy) - [Jewelry Try-On](https://www.alfreya.com/md/en/solutions/jewelry-try-on): A necklace, ring or bracelet on the shopper, to scale. (Available to deploy) - [Ear Stack Try-On](https://www.alfreya.com/md/en/solutions/ear-stack-try-on): Shoppers stack your earrings on a reference ear, each to scale. (Live with Kısmet by Milka) - [Watch Try-On](https://www.alfreya.com/md/en/solutions/watch-try-on): One photo of the wrist, then each watch at its case size. (Available to deploy) - [Live Video Try-On](https://www.alfreya.com/md/en/solutions/live-video-try-on): A fitting-room mirror on the product page. (Available to deploy) - [AI Model Studio](https://www.alfreya.com/md/en/solutions/ai-model-studio): One packshot, worn by a model, to scale. (Live with Kısmet by Milka) ### Commerce Operations The unglamorous work: catalogs, international stores, checkout paperwork, and the shop floor. - [Multi-Store Sync & Translation](https://www.alfreya.com/md/en/solutions/store-sync-translate): New products land in your international stores translated, priced and saved as drafts. (Available to deploy) - [Checkout Consent](https://www.alfreya.com/md/en/solutions/checkout-consent): Turkish distance-selling consent at checkout, with a record saved on each order. (Available to deploy) - [In-Store Product Finder](https://www.alfreya.com/md/en/solutions/in-store-product-finder): Photograph a piece. See its price, stock and other sizes. (Live with Kısmet by Milka) - [Catalog Intelligence](https://www.alfreya.com/md/en/solutions/catalog-intelligence): AI reads each photo and title. Our products run on what it writes, and so can your shop. (Available to deploy) ## Two stores, up close - [Vitruta: Five Alfreya products on a store of hundreds of brands](https://www.alfreya.com/md/en/case-studies/vitruta): Vitruta is a multi-brand premium fashion store on Shopify. Its search box, the similar products and outfits on its product pages, and the order of its collections all run on Alfreya today. The similar products and outfits draw on Alfreya’s catalog engine, which has read each product in the store. - [Kısmet by Milka: Model photos without a shoot](https://www.alfreya.com/md/en/case-studies/kismet): Kısmet by Milka makes handcrafted fine jewelry and sells it online and in its own shops. The content team uses Alfreya’s model studio, the shop staff use the product finder, and shoppers build ear stacks from Kısmet’s earrings. A jewelry try-on is built and waiting for launch. ## Industries - [Fashion & Apparel](https://www.alfreya.com/md/en/industries/fashion-apparel): Search that finds “gömlekler”, outfits from stock, and the jacket on the shopper. - [Footwear & Accessories](https://www.alfreya.com/md/en/industries/footwear-accessories): Show where a shoe or a bag fits into a whole look. - [Fine Jewelry](https://www.alfreya.com/md/en/industries/jewelry): A ring to scale on the shopper, model shots from one packshot, and a photo lookup at the counter. - [Watches](https://www.alfreya.com/md/en/industries/watches): A 40 mm case on the shopper’s wrist, model shots, and a photo lookup at the counter. ## How a project goes 1. **We read the catalog first**: Products, attributes, how your collections are built, and what you need to sell more of this season. 2. **Then we pick one problem**: Usually the search box or the product page. Whichever one is losing you shoppers today. 3. **It looks like your store**: Recommendations follow your merchandising rules and sit in your theme. Shoppers shouldn’t be able to tell where Alfreya starts. 4. **We keep tuning**: After launch we watch what shoppers do with it and adjust. That part doesn’t end. ## Frequently asked questions ### Which platforms do you work with? Any of them. On Shopify, our products install as apps with theme blocks. On other platforms and custom builds, we connect to your catalog and storefront and set it up with your team. ### Do we need a developer? Not on Shopify: storefront pieces are theme blocks your team places, and we install them with you. On other platforms, we do the integration together with your developers. ### Does it work in Turkish? Yes. Search strips Turkish suffixes, so “çantalar” finds bags. What shoppers see is in Turkish and English. ### What does it cost? It depends on which products you use. Tell us about your store and we’ll send a quote. --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en AI index (llms.txt): https://www.alfreya.com/llms.txt # Start with the problem you want to solve. > Fifteen products in three families. The second one you add reads the same catalog, so it goes in faster. ## Discovery & Search The search box, the product page, and what goes next to the product. - [Smart Search](https://www.alfreya.com/md/en/solutions/smart-search): A search box that forgives typos, plurals and a second language. (Live with Vitruta) - [AI Shopping Assistant](https://www.alfreya.com/md/en/solutions/ai-shopping-assistant): Search by sentence, by photo, or by “same but in black”. (Live with Vitruta) - [Visual Similar Products](https://www.alfreya.com/md/en/solutions/similar-products): Its lookalikes are one scroll down. AI found them in your catalog. (Live with Vitruta) - [Outfit Recommendations](https://www.alfreya.com/md/en/solutions/complete-the-look): Two colours, six outfits. AI styled each one from your stock. (Live with Vitruta) - [Collection Merchandising](https://www.alfreya.com/md/en/solutions/smart-merchandising): New arrivals move up, broken size runs step back. Twice a day, each move explained. (Live with Vitruta) ## Try-On & Content Products on the shopper, and products on a model, without booking a shoot. - [Virtual Try-On](https://www.alfreya.com/md/en/solutions/virtual-try-on): The shopper’s own photo, wearing the jacket from this page. (Available to deploy) - [Jewelry Try-On](https://www.alfreya.com/md/en/solutions/jewelry-try-on): A necklace, ring or bracelet on the shopper, to scale. (Available to deploy) - [Ear Stack Try-On](https://www.alfreya.com/md/en/solutions/ear-stack-try-on): Shoppers stack your earrings on a reference ear, each to scale. (Live with Kısmet by Milka) - [Watch Try-On](https://www.alfreya.com/md/en/solutions/watch-try-on): One photo of the wrist, then each watch at its case size. (Available to deploy) - [Live Video Try-On](https://www.alfreya.com/md/en/solutions/live-video-try-on): A fitting-room mirror on the product page. (Available to deploy) - [AI Model Studio](https://www.alfreya.com/md/en/solutions/ai-model-studio): One packshot, worn by a model, to scale. (Live with Kısmet by Milka) ## Commerce Operations The unglamorous work: catalogs, international stores, checkout paperwork, and the shop floor. - [Multi-Store Sync & Translation](https://www.alfreya.com/md/en/solutions/store-sync-translate): New products land in your international stores translated, priced and saved as drafts. (Available to deploy) - [Checkout Consent](https://www.alfreya.com/md/en/solutions/checkout-consent): Turkish distance-selling consent at checkout, with a record saved on each order. (Available to deploy) - [In-Store Product Finder](https://www.alfreya.com/md/en/solutions/in-store-product-finder): Photograph a piece. See its price, stock and other sizes. (Live with Kısmet by Milka) - [Catalog Intelligence](https://www.alfreya.com/md/en/solutions/catalog-intelligence): AI reads each photo and title. Our products run on what it writes, and so can your shop. (Available to deploy) --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/solutions AI index (llms.txt): https://www.alfreya.com/llms.txt # Alfreya Smart Search > Smart Search fixes typos, Turkish endings and a second language while the shopper types, then opens a panel of products, categories, pages and articles from your live catalog. “Sneeker”, “çantalar” and “tişört” all find products. ## Key facts | | | |---|---| | Status | Live with Vitruta | | Product family | Discovery & Search | | Where it runs | Search box and its drop-down panel | | Platform | Shopify app, or any other platform by integration | | Languages | Turkish and English | | Setup | Install the app, flip one theme switch | | Built for | Vitruta | | Industries | Fashion & Apparel, Footwear & Accessories, Fine Jewelry, Watches | ## The problem Shopify’s built-in search matches letters, not meaning. A store full of bags answers “çantalar” with zero results because of one plural suffix. Most shoppers don’t try a second spelling. They leave. ## What we built We read your catalog: titles, product types, tags, brands. As the shopper types, the query is fixed in the browser. Typos land on a word you actually use, Turkish endings stop getting in the way, and “sneaker” and “spor ayakkabı” point to the same shelf. Results come straight from your live catalog, with no copy to sync and no extra server on the way. ## What changes Searches that used to end on an empty page now open a panel of products, categories and suggestions. There’s no extra server between the shopper and the results, and no fee per query. ## How it works 1. **We read your catalog**: Titles, types, tags and brands. Your team has nothing to set up. 2. **Fix it while they type**: Plurals, suffixes, typos and Turkish and English synonyms are sorted out in the browser, before the query goes anywhere. 3. **Answer with live stock**: The panel fills with products, categories, pages and articles, at this minute’s prices and stock. Nothing to sync. 4. **Improved daily**: AI reads the day’s searches and adds the words that came back empty. Fewer dead ends each week. ## What’s included - **Turkish suffixes**: “Elbiseler” finds dresses. So does “elbiseyi”. - **Typo tolerance**: “Sneeker” is corrected to a word your catalog uses, not to a dictionary guess. - **Synonyms in two languages**: Turkish and English terms are linked. AI reviews searches daily and adds the words shoppers use that your titles don’t. - **No “no results” pages**: When a search would come back empty, the closest match is shown and the panel says so. - **Something to look at first**: Before a letter is typed, the panel shows popular searches and trending products. - **SKU and barcode lookup**: Store staff can type a SKU, a barcode or a tag and land on the product. - **Searches report**: Top queries, zero-result queries and clicks, inside the Shopify admin. ## Frequently asked questions ### How fast are the results? Instant. The widget is light, and results reach the shopper with no search server in between. ### What happens with a word it hasn’t seen before? The panel shows the closest match right away. Overnight, AI reads that search and adds the word, and the next shopper who types it lands on the right products. ## Pairs with - [AI Shopping Assistant](https://www.alfreya.com/md/en/solutions/ai-shopping-assistant): Search by sentence, by photo, or by “same but in black”. (Live with Vitruta) - [Visual Similar Products](https://www.alfreya.com/md/en/solutions/similar-products): Its lookalikes are one scroll down. AI found them in your catalog. (Live with Vitruta) - [Catalog Intelligence](https://www.alfreya.com/md/en/solutions/catalog-intelligence): AI reads each photo and title. Our products run on what it writes, and so can your shop. (Available to deploy) --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/solutions/smart-search AI index (llms.txt): https://www.alfreya.com/llms.txt # Alfreya AI Shopping Assistant > AI Shopping Assistant lets shoppers ask your store the way they would ask someone on the shop floor: a full sentence, a photo, or a follow-up like “in a lighter colour”. AI answers in their language, with products from your catalog. When nothing fits, it says so. ## Key facts | | | |---|---| | Status | Live with Vitruta | | Product family | Discovery & Search | | Where it runs | Storefront search panel and a chat window | | Platform | Shopify app, or any other platform by integration | | Languages | Answers in the shopper’s language | | Setup | Install, index the catalog, flip one theme switch | | Built for | Vitruta | | Industries | Fashion & Apparel, Footwear & Accessories, Fine Jewelry, Watches | ## The problem Nobody shops in keywords. People want “a linen blazer for a beach wedding” or “this, but in black”. A keyword box sees “linen”, “blazer”, “beach” and “wedding” as four unrelated words. The shopper rephrases twice, then gives up. ## What we built We index the catalog by what each product looks like and what it’s for. A shopper’s sentence or photo is matched against that index, follow-up questions keep the earlier context, and the reply is a set of product cards from your store. ## What changes A shopper who arrives with an occasion, a screenshot or a half-formed idea gets a short edit of your catalog. You get a report of what they asked and which conversations ended in a sale. ## How it works 1. **Index by look and use**: Each product is described by its look, material, colour and occasion. A shopper can find it without knowing what you titled it. 2. **The shopper asks**: A sentence, a photo, or a refinement such as “in a lighter colour”. 3. **Products come back**: The reply is in the shopper’s language. Each card can be opened, asked about, or used to find lookalikes. 4. **You see what sold**: The report shows top questions, the ones that went unanswered and the purchases that followed a conversation. ## What’s included - **Whole-sentence search**: Occasion, style, fit and budget are read together. “A linen blazer for a beach wedding” returns linen blazers. - **Search by photo**: Upload a picture of something you like and see the closest pieces this store carries. - **Follow-ups**: The conversation remembers what came before. Narrowing down doesn’t mean starting over. - **An honest “nothing found”**: If the catalog has no good match, the assistant says so. It won’t pad the answer with unrelated products. - **Ask and Similar on each card**: Two buttons per product: ask a question about it, or jump to pieces that look like it. - **Reports and an off switch**: Top questions, unanswered requests, attributed purchases. A pause control in the admin takes the assistant off the storefront whenever you want. ## Frequently asked questions ### Does it only show products from our store? Yes. Each card is a product from your catalog, one tap from its product page. ### Does it work alongside our keyword search? Yes. One word still gets instant keyword results, and full sentences go to the assistant. ## Pairs with - [Smart Search](https://www.alfreya.com/md/en/solutions/smart-search): A search box that forgives typos, plurals and a second language. (Live with Vitruta) - [Visual Similar Products](https://www.alfreya.com/md/en/solutions/similar-products): Its lookalikes are one scroll down. AI found them in your catalog. (Live with Vitruta) - [Outfit Recommendations](https://www.alfreya.com/md/en/solutions/complete-the-look): Two colours, six outfits. AI styled each one from your stock. (Live with Vitruta) --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/solutions/ai-shopping-assistant AI index (llms.txt): https://www.alfreya.com/llms.txt # Alfreya Visual Similar Products > Visual Similar Products puts a “more like this” row on the product page. AI reads each photo in your catalog for shape, colour, pattern and cut, then picks the closest pieces you have in stock, in the same gender and category. ## Key facts | | | |---|---| | Status | Live with Vitruta | | Product family | Discovery & Search | | Where it runs | A row on the product page | | Platform | Shopify app, or any other platform by integration | | Languages | Turkish and English catalogs | | Setup | Three steps: install, let AI read the catalog, add one block | | Built for | Vitruta | | Industries | Fashion & Apparel, Footwear & Accessories, Fine Jewelry, Watches | ## The problem Standard “related products” lean on tags, collections and vendors, which are rarely filled in the same way twice. Someone looking at a cream knit gets a random top from the same brand. Or a men’s piece on a women’s page, which is where they stop browsing. ## What we built AI reads the photos and details of each product for shape, colour, pattern and cut, labels its gender and category, and finds the closest pieces in the same store. Gender and category work as hard filters, sold-out pieces can be hidden, and a last AI pass puts the most convincing match first. From then on it keeps up with the catalog on its own. ## What changes A shopper who likes a piece but not its colour, or finds their size sold out, sees a near match without leaving the page. And a women’s page shows women’s pieces. ## How it works 1. **AI reads the catalog**: Each photo is read for shape, colour, pattern and cut. A full catalog takes minutes. 2. **It labels each piece**: Gender, category and pattern, in Turkish or English, labelled automatically. 3. **It finds the closest**: The nearest lookalikes in the same gender and category, in stock, best match first. 4. **Keeps up**: New pieces, new prices and sell-outs reach the row as they happen. ## What’s included - **Matched by appearance**: Silhouette, colour, texture and detail decide the match. Shared keywords don’t. - **Hard gender filter**: A women’s product page never suggests menswear, and the other way round. - **Same shelf first**: Suggestions come from the same category. They widen to the parent category only when the shelf is thin. - **A second AI pass**: A second AI pass reorders the candidates, putting the most convincing alternative first. - **Always up to date**: New, edited and deleted products are picked up automatically. - **Doesn’t hold up the page**: Results come from an index prepared in advance. The row loads without delaying anything else. ## Frequently asked questions ### How is this different from Shopify’s related products? Shopify’s suggestions lean on tags, collections and purchase patterns. This compares the photos, so a suggestion looks like the piece on screen. ### How quickly does the row appear? Straight away. The matches are worked out in advance, and the row loads with the page without holding anything up. ## Pairs with - [Outfit Recommendations](https://www.alfreya.com/md/en/solutions/complete-the-look): Two colours, six outfits. AI styled each one from your stock. (Live with Vitruta) - [AI Shopping Assistant](https://www.alfreya.com/md/en/solutions/ai-shopping-assistant): Search by sentence, by photo, or by “same but in black”. (Live with Vitruta) - [Smart Search](https://www.alfreya.com/md/en/solutions/smart-search): A search box that forgives typos, plurals and a second language. (Live with Vitruta) --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/solutions/similar-products AI index (llms.txt): https://www.alfreya.com/llms.txt # Alfreya Outfit Recommendations > Outfit Recommendations puts a “Shop the look” block on the product page. AI reads each colourway from its photos, plans three looks around it (everyday, evening, weekend) and fills them with pieces you have in stock, before the shopper arrives. ## Key facts | | | |---|---| | Status | Live with Vitruta | | Product family | Discovery & Search | | Where it runs | A “Shop the look” block on the product page | | Platform | Shopify app, or any other platform by integration | | Languages | Turkish and English catalogs | | Setup | Three steps: install, let AI style the catalog, add one block | | Built for | Vitruta | | Industries | Fashion & Apparel, Footwear & Accessories, Fine Jewelry, Watches | ## The problem People buy outfits. Stores sell one product page at a time. A stylist can dress the ten hero pieces by hand, not the other two thousand, and “you may also like” rows ignore colour, gender and what goes with what. ## What we built AI reads each colourway from its photos, then plans around it the way a stylist would: what goes underneath, on the feet, as a bag or an accessory, and in which colours. It picks colours that go with the piece, not colours that match it. Each slot is filled with an in-stock piece from the same store and department, with ranked back-ups behind it. Three looks per colourway are ready before anyone asks. ## What changes The product page shows three shoppable outfits for the colour on screen. A shopper who came for the dress can leave with the sandals and the bag too. ## How it works 1. **AI reads each colour**: Each colourway is read from its photos. The navy dress and the light blue one are styled apart. 2. **It plans three looks**: Everyday, evening and weekend: the pieces each look needs, in colours that go with the piece. 3. **Filled from your stock**: Each slot gets an in-stock piece from the same store and department, with back-ups ranked behind it. 4. **Keeps up**: Each piece is checked as the page loads. New colours get their looks as they arrive. ## What’s included - **Three looks per colour**: Everyday, evening and weekend. Three outfits, not one outfit three times. - **Colours that go with it**: AI picks companion colours for this variant: camel with navy, not navy with navy. - **Right department**: A men’s jacket is never styled with a women’s skirt. - **In stock today**: Each piece is checked as the page loads, with today’s price and link. - **Back-ups per slot**: When a piece sells out, the next one in line takes its place. The look holds together. - **Swap for a lookalike**: Shoppers can open any piece in a look and swap it for a lookalike. - **No wait on the page**: The product page loads a look that was finished in advance. ## Frequently asked questions ### Are the outfits made from our products? Yes, only yours, from the same department, checked for stock when the page loads. ### How fast do the looks load? Right away. The looks are built in advance, and the page loads a finished outfit. ## Pairs with - [Visual Similar Products](https://www.alfreya.com/md/en/solutions/similar-products): Its lookalikes are one scroll down. AI found them in your catalog. (Live with Vitruta) - [Virtual Try-On](https://www.alfreya.com/md/en/solutions/virtual-try-on): The shopper’s own photo, wearing the jacket from this page. (Available to deploy) - [AI Shopping Assistant](https://www.alfreya.com/md/en/solutions/ai-shopping-assistant): Search by sentence, by photo, or by “same but in black”. (Live with Vitruta) --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/solutions/complete-the-look AI index (llms.txt): https://www.alfreya.com/llms.txt # Alfreya Collection Merchandising > Collection Merchandising re-sorts your collections with a score that weighs sales, stock and size coverage, how new a piece is, and your own boosts. You preview each new order beside the live one, apply it, or let a schedule apply it. Each order is read back from Shopify to confirm it, and can be undone. ## Key facts | | | |---|---| | Status | Live with Vitruta | | Product family | Discovery & Search | | Where it runs | Collection pages and the Shopify admin | | Platform | Shopify app, or any other platform by integration | | Languages | Doesn’t depend on language | | Setup | Three steps: install, pick collections, apply or schedule | | Built for | Vitruta | | Industries | Fashion & Apparel, Footwear & Accessories, Fine Jewelry, Watches | ## The problem Most collections are sorted once, by hand, and then forgotten. Best sellers drift down, a jacket left in one size keeps the front row, and new arrivals end up on page four. Nobody with hundreds of collections has time to re-sort them weekly. ## What we built Each product gets a score from your store’s own data. Sales count most: revenue, units, conversion, add-to-cart and views over a week, a month and a quarter. Stock depth and size coverage come next, then how new it is, then your boosts. Pieces returned often lose points, and your pins hold. The proposed order is checked before it is written, and read back after. ## What changes Collections follow what people buy and what you can ship in their size. Merchandisers stop dragging products around and spend the time deciding what to push. ## How it works 1. **Pick the collections**: One by one or all at once. Brand pages and big category pages get their own preset. 2. **Each piece is scored**: Sales, stock and sizes, newness and your boosts roll into one number. Returns pull it down. 3. **Preview it**: The live order and the proposed one side by side, with the reason for each move. 4. **Apply, or let it run**: Apply it yourself or switch on the schedule. Each write is confirmed and logged. ## What’s included - **More than best-seller rank**: Revenue, units, conversion, add-to-cart and views are weighed together. - **Stock and size coverage**: A full size run rises. A piece down to one size steps back. - **Returns count against it**: A product returned more often than the rest of its collection moves down. - **Pins and new-in slots**: Lock the products that must stay put, and keep room near the top for new arrivals. - **Brand variety**: In multi-brand collections, the first row won’t be six pieces from one label. - **Why here, for each product**: Open any product and see its score split into the parts that put it there. - **Confirmed and reversible**: Each new order is read back from Shopify after it is written, and the previous one is kept for a restore. ## Frequently asked questions ### How does it decide the order? It scores each product from your sales, stock and product data, and shows you why each one sits where it does. ### Can we keep certain products at the top? Yes. Pinned products don’t move. Boosts and room for new arrivals steer the rest. ## Pairs with - [Catalog Intelligence](https://www.alfreya.com/md/en/solutions/catalog-intelligence): AI reads each photo and title. Our products run on what it writes, and so can your shop. (Available to deploy) - [Smart Search](https://www.alfreya.com/md/en/solutions/smart-search): A search box that forgives typos, plurals and a second language. (Live with Vitruta) - [Outfit Recommendations](https://www.alfreya.com/md/en/solutions/complete-the-look): Two colours, six outfits. AI styled each one from your stock. (Live with Vitruta) --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/solutions/smart-merchandising AI index (llms.txt): https://www.alfreya.com/llms.txt # Alfreya Virtual Try-On > Virtual Try-On puts a small button on the product photo. The shopper takes a picture of themselves, or picks one from their phone, and after a short wait sees that garment on their own body. Add to cart sits right under the result. We don’t keep the photo. ## Key facts | | | |---|---| | Status | Available to deploy | | Product family | Try-On & Content | | Where it runs | Product page, on the first product photo | | Platform | Shopify app, or any other platform by integration | | Languages | Turkish and English | | Setup | App install and one theme toggle | | Built for | Vitruta | | Industries | Fashion & Apparel | ## The problem The product photo shows a model. The person deciding is not that model, and has to guess how the cut, the colour and the length will look on them. Most of the hesitation on a fashion page is that guess. ## What we built Tap the button on the first product photo and the camera opens, with a self-timer if both hands are needed. One photo is enough. AI dresses the shopper in the exact garment from that product page and the result appears in the same card. ## What changes The question “how will it look on me?” gets answered on the product page, one step away from the cart. No account to create and no app to download. ## How it works 1. **A button on the photo**: It sits in the corner of the first product photo, where the shopper is already looking. 2. **One photo**: The camera opens straight away. There is a self-timer for a hands-free shot, or the shopper can upload a picture they already have. 3. **A short wait**: AI dresses their photo in the garment from the page. Tap to enlarge. 4. **Add to cart**: The button is right under the result, next to a quiet “try again”. ## What’s included - **On the image, not the buy box**: The entry point lives on the product photo and leaves the price, sizes and add to cart alone. - **Self-timer or upload**: Prop the phone on a shelf and step back. Or skip the camera and use a photo from the gallery. - **That garment, exactly**: The try-on is built from the product page’s photo. The shopper sees the jacket they are looking at, not something like it. - **Ask a friend**: Results can be downloaded, copied or shared before anyone decides. - **Photos aren’t stored**: The photo is used once, to make the try-on. The result lives in the shopper’s open tab. We keep no copy. - **Daily limits**: Each browser gets a set number of try-ons a day. We agree the number with you. ## Frequently asked questions ### Does the shopper need an account or an app? No. It all happens in a card over the product page, in whatever browser they are already using. ### What happens to the shopper’s photo? It is used once to make the try-on and never written anywhere on our side. The result goes when the tab closes, unless the shopper saves it. ### What does the shopper see? The garment on them, in their photo, with its colour, print and cut. Sizing stays with your size chart. ## Pairs with - [Live Video Try-On](https://www.alfreya.com/md/en/solutions/live-video-try-on): A fitting-room mirror on the product page. (Available to deploy) - [Outfit Recommendations](https://www.alfreya.com/md/en/solutions/complete-the-look): Two colours, six outfits. AI styled each one from your stock. (Live with Vitruta) - [AI Model Studio](https://www.alfreya.com/md/en/solutions/ai-model-studio): One packshot, worn by a model, to scale. (Live with Kısmet by Milka) --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/solutions/virtual-try-on AI index (llms.txt): https://www.alfreya.com/llms.txt # Alfreya Jewelry Try-On > Jewelry Try-On shows a necklace, ring or bracelet on the shopper’s photo to scale, from its sizes in millimetres. A camera guide for each type of piece gets them a photo that can be measured. Results stay on their phone. ## Key facts | | | |---|---| | Status | Available to deploy | | Product family | Try-On & Content | | Where it runs | Product page try-on button and camera overlay | | Platform | Shopify app, or any other platform by integration | | Languages | Turkish and English | | Setup | App install, theme toggle and product sizes | | Built for | Kısmet by Milka | | Industries | Fine Jewelry | ## The problem Jewelry is bought on scale. A pendant that looks bold in a close-up can be tiny on a neck, and a ring shot on its own says nothing about a finger. Most try-on tools paste the piece at whatever size looks nice. That answers the wrong question. ## What we built Each product carries its measurements. The camera then asks for a photo it can measure: face points for a necklace, a wrist outline for a bracelet, and for a ring, a hand outline plus two points dragged across the finger. AI places the piece on that photo at its actual size. ## What changes Shoppers find out how big a piece is on them before they pay. One good photo works for the next necklace, bracelet or ring too. ## How it works 1. **Enter the sizes**: Pendant, chain and band dimensions go into the Shopify admin once per product. 2. **The camera guides**: For a necklace the shutter fires by itself. For a bracelet or ring, an outline shows where the wrist or hand goes. 3. **Two points on a finger**: For rings, the shopper drags two points to the edges of the finger. That width sets the ring’s size. 4. **On you, to scale**: The piece comes back on the shopper’s photo to scale, saved on their device. ## What’s included - **Scale in millimetres**: Scale is worked out from the shopper’s face, wrist or hand. A small pendant looks small. A statement piece looks like one. - **Hands-free for necklaces**: Face points check framing, tilt and stillness. Then a short countdown and the photo is taken. - **Wrist and hand outlines**: Traced outlines for left and right hands show exactly where the wrist or fingers go. - **Desktop to phone**: On a laptop the button shows a QR code. Scan it and the same product opens on the phone, ready to try. - **Pose once**: Photos are kept in a small library on the device. The next ring doesn’t need a new picture. - **Stays on the device**: Photos and results are kept in the shopper’s browser for up to 30 days. Each has a delete button. ## Frequently asked questions ### Which pieces can shoppers try on? Necklaces, bracelets and rings, each with its own camera guide. You can hide the button on any product type. ### How is each piece sized? From the photo. The eye sets the scale for a necklace, the wrist outline for a bracelet and two points across the finger for a ring. The piece then sits on the shopper to scale, at the sizes on its product page. A straight, well-lit photo gives the best result. ### Where do shoppers’ photos go? A photo is sent to make the try-on. Photos and results are then kept in the shopper’s browser for up to 30 days, where they can delete them. ## Pairs with - [Ear Stack Try-On](https://www.alfreya.com/md/en/solutions/ear-stack-try-on): Shoppers stack your earrings on a reference ear, each to scale. (Live with Kısmet by Milka) - [Watch Try-On](https://www.alfreya.com/md/en/solutions/watch-try-on): One photo of the wrist, then each watch at its case size. (Available to deploy) - [AI Model Studio](https://www.alfreya.com/md/en/solutions/ai-model-studio): One packshot, worn by a model, to scale. (Live with Kısmet by Milka) --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/solutions/jewelry-try-on AI index (llms.txt): https://www.alfreya.com/llms.txt # Alfreya Ear Stack Try-On > Ear Stack Try-On lets a shopper build an ear stack from your earrings on a reference ear: studs anywhere on the ear, drops and hoops through its piercings, each at the millimetres on its product page. AI wore every piece on the ear beforehand, so it lands the moment it is picked. Your part is the list of product links. ## Key facts | | | |---|---| | Status | Live with Kısmet by Milka | | Product family | Try-On & Content | | Where it runs | A stack builder for desktop, tablet and phone | | Platform | Any platform, set up with our team | | Languages | English | | Setup | The product links of your earrings | | Built for | Kısmet by Milka | | Industries | Fine Jewelry | ## The problem Earrings are worn in stacks but sold one product page at a time. A shopper choosing a third piece has to picture it beside the two they already wear, at a size the packshot doesn’t show. ## What we built AI wears each earring on one ear ahead of time, at its real millimetres, and each hoop at every piercing it fits. The shopper taps or drags pieces onto the ear, swaps a Mini for a Maxi and sees where two pieces would overlap. ## What changes Shoppers see the stack before they buy the pieces in it. The store hands over product links and gets a stack builder with its own earrings in it. ## How it works 1. **Send the product links**: The earrings you want in the stack builder. 2. **Sizes from your product pages**: Width and height come from what each page already lists. 3. **AI wears each piece on the ear**: Once per piece, and once per piercing for hoops. 4. **Checked, then live**: Each render is checked before it joins the collection. ## What’s included - **True millimetre size**: A 3.9 mm stud looks small next to an 18 mm one, as it will in the mirror. - **Through the piercing**: Drops and hoops snap to the nearest free piercing on the lobe. - **Instant on every tap**: Nothing is generated while the shopper stacks. Pieces land at once. - **Swap a piece**: The Mini becomes the Maxi in the same spot. - **Overlaps flagged**: Two pieces on the same patch of skin get a note in the stack list. - **Made for phones**: On a phone the view zooms to the ear and pieces are dragged up from the strip. ## Frequently asked questions ### Which earrings can go in? Studs, drops, hoops and ear cuffs. Studs and cuffs go anywhere on the ear; drops and hoops hang from its piercings, one piece per hole. ### Does the shopper upload a photo? No. They stack on the ear in the builder, and nothing is uploaded. ### What happens when we add new earrings? Send their product links. They are worn on the ear, checked and added to the collection. ## Pairs with - [Jewelry Try-On](https://www.alfreya.com/md/en/solutions/jewelry-try-on): A necklace, ring or bracelet on the shopper, to scale. (Available to deploy) - [AI Model Studio](https://www.alfreya.com/md/en/solutions/ai-model-studio): One packshot, worn by a model, to scale. (Live with Kısmet by Milka) - [In-Store Product Finder](https://www.alfreya.com/md/en/solutions/in-store-product-finder): Photograph a piece. See its price, stock and other sizes. (Live with Kısmet by Milka) --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/solutions/ear-stack-try-on AI index (llms.txt): https://www.alfreya.com/llms.txt # Alfreya Watch Try-On > Watch Try-On puts a watch on the shopper’s wrist from one photo, at the case diameter its product page lists. An arm outline on the camera lines up the shot. AI draws the watch onto the wrist in the photo’s light, and the same photo works for the whole catalog. ## Key facts | | | |---|---| | Status | Available to deploy | | Product family | Try-On & Content | | Where it runs | A button on the product page, or a standalone link | | Platform | Any platform, or a standalone web link | | Languages | Turkish and English | | Setup | Case diameter per watch, set up with our team | | Industries | Fine Jewelry, Watches | ## The problem Case size is the first thing a watch buyer asks about and the hardest thing to answer online. A number in millimetres means little until it sits on a wrist. Product grids frame a 30 mm and a 47 mm watch in the same square, and the studio shot on a model’s wrist tells the shopper nothing about theirs. ## What we built The back camera shows an arm outline with the case position marked. The shopper fills the outline with their wrist and takes one photo. AI places the watch at the size its case diameter calls for and wraps the strap around the wrist. The placement mark is removed before the result is shown, and the shopper can try any number of watches on the same photo. ## What changes Shoppers judge case size on their own wrist, not on a spec sheet. Watch Try-On is built and has been demonstrated. It isn’t live in a store yet. ## How it works 1. **Add the case size**: A 30 mm and a 47 mm case shouldn’t look the same on a wrist. 2. **Line up the wrist**: An arm outline on the back camera shows where the wrist goes and where the case will sit. 3. **One photo**: The shopper takes one picture. The outline tells AI where the case sits and how big it is. 4. **The whole catalog**: Tap one watch or try them all, on the same photo. ## What’s included - **Sized from the case**: The case is drawn at its listed diameter on the outline the shopper lined up. Nothing is stretched to fill the photo. - **An outline that stays put**: The arm outline is fixed on the screen. The shopper moves the wrist into it, so shots come out framed alike. - **Worn, not pasted**: AI wraps the strap around the wrist and matches the light and shadow in the photo. - **Try all**: One photo, any number of watches, three made at a time. - **The shopper’s own watch**: A photo of any watch works too, with its case diameter set on the card. ## Frequently asked questions ### How do we set it up? Send us your watches with their case diameters. We set it up with you, and it runs as a button on the product page or as a standalone link. ### How does the watch come out at the right size? The case is drawn at the diameter on its product page, against a wrist lined up in the on-screen outline. The shopper sees the watch in proportion to their wrist. ### Can shoppers compare several watches? Yes. One photo works for the whole catalog. Try all makes three at a time, and any card can be zoomed, saved or remade with ↻. ## Pairs with - [Jewelry Try-On](https://www.alfreya.com/md/en/solutions/jewelry-try-on): A necklace, ring or bracelet on the shopper, to scale. (Available to deploy) - [AI Model Studio](https://www.alfreya.com/md/en/solutions/ai-model-studio): One packshot, worn by a model, to scale. (Live with Kısmet by Milka) - [Visual Similar Products](https://www.alfreya.com/md/en/solutions/similar-products): Its lookalikes are one scroll down. AI found them in your catalog. (Live with Vitruta) --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/solutions/watch-try-on AI index (llms.txt): https://www.alfreya.com/llms.txt # Alfreya Live Video Try-On > Live Video Try-On draws a garment onto the shopper’s camera feed in real time. They turn, and the piece turns with them. It runs in the browser on the product page. Nothing to install, nothing recorded. ## Key facts | | | |---|---| | Status | Available to deploy | | Product family | Try-On & Content | | Where it runs | Product page camera overlay | | Platform | Shopify app, or any other platform by integration | | Languages | Turkish and English | | Setup | App install and one theme toggle | | Industries | Fashion & Apparel | ## The problem A still photo shows a garment from one angle. In a fitting room people turn and step back to see the whole look. That moment is missing online. ## What we built The shopper opens the camera from the product page and sees themselves wearing the garment, live. AI reads the garment once from the product photo and redraws it on each frame as the shopper moves. ## What changes A fitting-room moment on the product page, from more than one angle. Live Video Try-On is built and has been demonstrated. It isn’t live in a store yet and is ready to deploy on Shopify. ## How it works 1. **Open the camera**: A button on the product page starts the camera in a card over the page. 2. **The garment is read once**: AI writes a description of the garment from its product photo. Nobody sets anything up by hand. 3. **Wear it live**: The garment appears on the shopper and stays on as they turn and move. 4. **Keep the last frame**: When the session ends, the last frame stays on screen as a still, with a “try again” button. ## What’s included - **Real-time rendering**: The garment is redrawn on the moving video, not on a single snapshot. - **Follows movement**: Turn, step back, look over a shoulder. - **Descriptions written for you**: Each garment’s description is generated from its photo and reused. Your team writes no instructions. - **Nothing recorded**: The video streams for the session and is never recorded or stored by us. - **Timed sessions**: Each session has a visible countdown. That keeps streaming costs in check. - **Honest about slow connections**: If the connection can’t carry live video, the widget says so in the shopper’s language. ## Frequently asked questions ### How is this different from Virtual Try-On? Virtual Try-On turns one photo into one finished image. Live Video Try-On works on the moving camera feed, so the shopper sees the garment while they move. ### How does it go on our store? On Shopify, install the app and switch on its block on the product page. Other platforms connect by integration with our team. ### What is it best for? Seeing the look in motion. The shopper turns and steps back the way they would in a fitting room. Sizing stays with your size chart. ## Pairs with - [Virtual Try-On](https://www.alfreya.com/md/en/solutions/virtual-try-on): The shopper’s own photo, wearing the jacket from this page. (Available to deploy) - [Outfit Recommendations](https://www.alfreya.com/md/en/solutions/complete-the-look): Two colours, six outfits. AI styled each one from your stock. (Live with Vitruta) - [AI Model Studio](https://www.alfreya.com/md/en/solutions/ai-model-studio): One packshot, worn by a model, to scale. (Live with Kısmet by Milka) --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/solutions/live-video-try-on AI index (llms.txt): https://www.alfreya.com/llms.txt # Alfreya AI Model Studio > AI Model Studio makes your models and puts your jewelry on them. Your team generates models in batches and adds a packshot with its millimetres, and AI puts the piece on at that size. A sentence fixes what’s off. The results are files for your product pages, marketplaces and social posts. ## Key facts | | | |---|---| | Status | Live with Kısmet by Milka | | Product family | Try-On & Content | | Where it runs | Web studio for your content team | | Platform | Web app, the files work with any store | | Languages | English | | Setup | One team sign-in, nothing to install | | Built for | Kısmet by Milka | | Industries | Fine Jewelry | ## The problem Each new piece needs on-model photos, and each shoot means booking a model, a studio and a retoucher. With a catalog that turns over every season, most products end up with a packshot and nothing else. Generic AI images don’t help much with jewelry: they get the scale wrong. ## What we built AI generates models for your catalog in the format you need, then puts your packshot on them at the millimetres you type. Necklaces and chokers go on any model you made. Rings, earrings and bracelets go on hand, ear and wrist photos from your own shoots, which our team measures once. A sentence changes the light or the pose. ## What changes Kısmet by Milka’s team uses the studio to put new pieces on models from their packshots. They go on the site worn by a model, with no shoot on the calendar. ## How it works 1. **Generate models**: Up to six at a time, women or men, in eight formats. Keep the best in the gallery. 2. **Add the piece**: One packshot and the sizes your product page lists. 3. **Worn at its size**: AI puts the piece on the model at those millimetres. 4. **Fix it, then download**: Change the light or the pose with a sentence, then download at 4K. ## What’s included - **Models made in the studio**: Batches of up to six, women or men, steered by up to three reference photos. - **Eight formats**: Square, portrait, story and landscape ratios, for product pages and social posts. - **Worn at its millimetres**: Pendants, chokers, rings, earrings and bracelets, at the size you type. - **Edit with a sentence**: “Turn her head a little to the right.” Typed instead of retouched. - **Flawed necklaces held back**: A necklace image with a visible flaw is stopped before it reaches you. - **Kept in History**: Try-ons stay in the studio, grouped by necklace or by model, ready to download again. ## Frequently asked questions ### How does the size stay right? You type the piece’s millimetres. The studio measures the model in the photo, and AI puts the piece on at that size. For rings, earrings and bracelets, our team measures each hand, ear or wrist photo once. ### Who are the models? AI generates them in the studio, from your reference photos if you add some. ### Where can we use the images? Download them and use them on your product pages, marketplace listings, social posts and newsletters. ## Pairs with - [Jewelry Try-On](https://www.alfreya.com/md/en/solutions/jewelry-try-on): A necklace, ring or bracelet on the shopper, to scale. (Available to deploy) - [Ear Stack Try-On](https://www.alfreya.com/md/en/solutions/ear-stack-try-on): Shoppers stack your earrings on a reference ear, each to scale. (Live with Kısmet by Milka) - [Virtual Try-On](https://www.alfreya.com/md/en/solutions/virtual-try-on): The shopper’s own photo, wearing the jacket from this page. (Available to deploy) --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/solutions/ai-model-studio AI index (llms.txt): https://www.alfreya.com/llms.txt # Alfreya Multi-Store Sync & Translation > Add a product to your main Shopify store and a translated, priced draft appears in each international store. Later edits follow it. Fields the local team changed by hand stay locked. ## Key facts | | | |---|---| | Status | Available to deploy | | Product family | Commerce Operations | | Where it runs | Shopify admin app and product-page card | | Platform | Shopify (installed on each connected store) | | Languages | Whatever language each child store sells in | | Setup | Install on both stores, approve, confirm the vocabulary | | Industries | Fashion & Apparel, Footwear & Accessories, Fine Jewelry, Watches | ## The problem Selling abroad on Shopify usually means a second store per market, and each product gets made twice. Someone copies it, translates it, prices it in pounds and tags it again. The two versions drift. The UK store launches late. Then one bulk update from head office wipes out the local team’s fixes. ## What we built Both stores install the same app and approve the link. From then on, a new product in the main store lands in the child store as a draft. The copy is translated into the child store’s language, the price comes from a per-currency international price field, and tags, product types and vendors are mapped to the terms that store already uses. Updates follow field by field. ## What changes Nobody re-enters products. The international store keeps pace with the main one, and the local team spends its time reviewing drafts and pressing publish. ## How it works 1. **Connect the stores**: Install on the main store, invite the child store and pick a language. The child store installs the same app and approves. 2. **Agree on the words**: Tags, product types and vendors are matched to the child store’s terms. AI suggests, your team confirms. 3. **Drafts arrive**: New products show up translated and priced, either as drafts or on a waiting list you pick from. 4. **Edits follow**: Edits, variants, images, prices and archiving carry over from the main store. Hand edits in the child store stay locked. ## What’s included - **Translated drafts**: Titles, descriptions, SEO fields, tags, product types, vendors and image alt text. When the source changes, only that field is translated again. - **A price per currency**: Each currency gets its own international price field. A copy without one is flagged “Needs price” in both stores. - **Hands off local edits**: Change a field by hand in the child store, variant prices included, and it is locked against overwrite. - **Vocabulary mapping**: If the main store tags it “Triko” and the UK store already says “Knitwear”, the copy gets “Knitwear”. AI proposes a match, your team has the final say. - **Bulk send and pause**: Send the existing catalog in one go, pause a whole connection, or hold edits while new products keep coming through. - **Status card and log**: A card on the product page shows sync status, price and any error, with a retry button. The activity log keeps the history. ## Frequently asked questions ### Who decides what goes live? Your team. Copies arrive as drafts with the translation and price ready to review, and you can send the whole catalog or only the products you pick. ### Can one main store feed several international stores? Yes. Each child store keeps its own language, currency, vocabulary and sync settings. ## Pairs with - [Catalog Intelligence](https://www.alfreya.com/md/en/solutions/catalog-intelligence): AI reads each photo and title. Our products run on what it writes, and so can your shop. (Available to deploy) - [Checkout Consent](https://www.alfreya.com/md/en/solutions/checkout-consent): Turkish distance-selling consent at checkout, with a record saved on each order. (Available to deploy) - [In-Store Product Finder](https://www.alfreya.com/md/en/solutions/in-store-product-finder): Photograph a piece. See its price, stock and other sizes. (Live with Kısmet by Milka) --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/solutions/store-sync-translate AI index (llms.txt): https://www.alfreya.com/llms.txt # Alfreya Checkout Consent > A checkbox above the pay button for the consent Turkish distance-selling rules require. No tick, no order. Once ticked, each document exactly as that buyer saw it is saved on the order. ## Key facts | | | |---|---| | Status | Available to deploy | | Product family | Commerce Operations | | Where it runs | Checkout, order record and confirmation email | | Platform | Shopify Plus (checkout UI extension on the payment step) | | Languages | Turkish, with custom wording supported | | Setup | Shopify Plus, a five-step guide and one checkout setting | | Industries | Fashion & Apparel, Footwear & Accessories, Fine Jewelry, Watches | ## The problem Turkish law says the Ön Bilgilendirme Formu and the Mesafeli Satış Sözleşmesi must be shown and confirmed before a distance sale. It also says the seller has to prove it, and keep the proof. A link in the footer proves nothing. Neither does a checkbox timestamp: it cannot show which text the buyer read, or that it had their name and address in it. ## What we built A checkbox with the standard sentence sits above the pay button, and each document name opens the full text in a modal. The blanks in each document (buyer name, address, the products in the cart, the totals) fill in from the live checkout as the buyer types. When the box is ticked, the documents as rendered are written onto the order, inside the store’s Shopify account. ## What changes Each order carries its own proof of pre-contractual information and consent. When a buyer, a consumer arbitration committee or a court asks for it, you export it. ## How it works 1. **Pick the documents**: Point to the store pages for the two required documents. A privacy policy or KVKK notice can go in the same box. 2. **Mark where the buyer goes**: The editor guesses where names, addresses, products and totals belong. You check each line in place. 3. **Switch it on at checkout**: A walkthrough shows where to add the block in the checkout editor and how to let it stop unticked orders. 4. **A copy for the buyer**: Paste one snippet and the accepted documents go out with the order confirmation email. ## What’s included - **Checkbox above the pay button**: The sentence names both documents. No tick, no order. - **Documents filled in for the buyer**: Buyer, billing and delivery details, order lines and totals appear inside each modal, taken from the live checkout. - **Evidence on the order**: The sentence, the full text of each document as the buyer saw it, a version mark per document and the time of acceptance. - **Kept in your store, not ours**: Records live in your Shopify account and stay there if you uninstall the app. We never store buyer data or contracts. - **Export by date or order**: Pick a date range or paste order numbers. Get a print-ready HTML file, a CSV register or JSON. - **Five-step setup**: Turkish defaults for each message and a live preview. Most stores finish without writing a word. ## Frequently asked questions ### Can we add a privacy policy or KVKK notice? Yes. Pick the page during setup. It joins the consent sentence with correct Turkish suffixes and is saved next to the two required documents. ### What about express checkout buttons? Shop Pay, Apple Pay and Google Pay take a shorter path through checkout. During setup we help you bring those orders through consent too, with a checkout validation rule or by adjusting the buttons. ## Pairs with - [Multi-Store Sync & Translation](https://www.alfreya.com/md/en/solutions/store-sync-translate): New products land in your international stores translated, priced and saved as drafts. (Available to deploy) - [Catalog Intelligence](https://www.alfreya.com/md/en/solutions/catalog-intelligence): AI reads each photo and title. Our products run on what it writes, and so can your shop. (Available to deploy) - [In-Store Product Finder](https://www.alfreya.com/md/en/solutions/in-store-product-finder): Photograph a piece. See its price, stock and other sizes. (Live with Kısmet by Milka) --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/solutions/checkout-consent AI index (llms.txt): https://www.alfreya.com/llms.txt # Alfreya In-Store Product Finder > An app for the shop floor, on a phone or tablet. An associate photographs a piece and AI finds it in the online catalog, with today’s price, stock and other sizes. The team opens it from one link. ## Key facts | | | |---|---| | Status | Live with Kısmet by Milka | | Product family | Commerce Operations | | Where it runs | Associate app on phone or tablet, plus Shopify admin | | Platform | Shopify app, or any other platform by integration | | Languages | Matches by image, any catalog language | | Setup | App install, catalog sync and an access key | | Built for | Kısmet by Milka | | Industries | Fashion & Apparel, Fine Jewelry, Watches | ## The problem A customer holds up an earring and asks if it comes in a bigger size. The tag is gone. The associate scrolls the catalog from memory or rings the back office, and by the time the answer comes, the customer has moved on to the next case. ## What we built The associate takes a photo of the piece, tag or no tag. AI matches it against the store’s online catalog, and the product comes up with today’s price, stock and variants, the other sizes of the same design, and the price in five currencies from the store’s own price list. ## What changes At Kısmet by Milka, associates answer price, stock and size questions on the shop floor in moments, from the same catalog the website sells. ## How it works 1. **Catalog sync**: Products are read from Shopify and updated as they change. A nightly check picks up anything missed. 2. **Share one link**: The Shopify admin gives you a link with an access key. Associates open it on a phone or tablet. 3. **Photograph the piece**: AI matches the photo against your catalog. Picking a category first narrows the search. 4. **Answer the customer**: Price, stock, ring sizes and other sizes on one screen, while the customer is still standing there. ## What’s included - **Matching by photo**: Finds the piece from a photo. The tag can be missing. - **Five currencies**: Lira, dollars, euros, dirhams and pounds, from your own price list, not an exchange rate. - **Live price and stock**: Prices, stock and variants come from the online catalog and change when the store changes them. - **Size families**: Many jewelers list each size of a design as its own product. The finder groups them, which is how “do you have it bigger?” gets an answer. - **Category filter**: Narrow to necklaces, rings or any other category for faster, closer matches. - **A link, not passwords**: The team opens one link with an access key. Nobody needs a Shopify account. ## Frequently asked questions ### How does it stay up to date? Changes in Shopify come through as they happen, and a nightly check catches anything missed. A price change doesn’t trigger a full re-read. ### Does a quick snap work? Yes. The finder lists the closest matches, most likely first. Picking a category narrows them, and the one the associate taps is saved with the scan. ## Pairs with - [Visual Similar Products](https://www.alfreya.com/md/en/solutions/similar-products): Its lookalikes are one scroll down. AI found them in your catalog. (Live with Vitruta) - [Multi-Store Sync & Translation](https://www.alfreya.com/md/en/solutions/store-sync-translate): New products land in your international stores translated, priced and saved as drafts. (Available to deploy) - [Catalog Intelligence](https://www.alfreya.com/md/en/solutions/catalog-intelligence): AI reads each photo and title. Our products run on what it writes, and so can your shop. (Available to deploy) --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/solutions/in-store-product-finder AI index (llms.txt): https://www.alfreya.com/llms.txt # Alfreya Catalog Intelligence > The catalog engine Alfreya’s products run on. AI reads each photo and title, labels gender, category, pattern and season, and gives each colourway a plain colour name and a stylist’s note. Our team can point the same data at other parts of your shop. ## Key facts | | | |---|---| | Status | Available to deploy | | Product family | Commerce Operations | | Where it runs | Feeds Alfreya’s products, and other uses on request | | Platform | Shopify, or any other platform by integration | | Languages | Turkish and English catalogs | | Setup | Our team connects the catalog | | Industries | Fashion & Apparel, Footwear & Accessories, Fine Jewelry, Watches | ## The problem Catalogs are messy. A fleece jacket is filed as a sweater. Supplier colour names like “Salute” or “Bark” mean nothing to a shopper. Descriptions are pasted from the supplier, and several colourways hide behind one listing. Anything built on that data, from recommendations to filters, inherits the mess. ## What we built AI reads the whole catalog, photo by photo, and splits each listing into its colourways. Each colourway gets gender, category, pattern and season, a plain colour name and a short note in a stylist’s words, and is indexed by how it looks and by what it is. Raw category names fold into one clean taxonomy. Alfreya’s products read from it, and our team can point the same data at other jobs in your shop: filters and collections, product feeds, consistent attributes across brands, first drafts of product copy. ## What changes Visual Similar Products and Outfit Recommendations run on it. Anything else in your shop that needs clean catalog data can draw from the same picture once our team connects it. ## How it works 1. **We connect your catalog**: Our team links your store, on Shopify or any other platform. 2. **AI reads each product**: Photo and title, colourway by colourway: gender, category, pattern, season, a colour name and a stylist’s note. 3. **It feeds what you choose**: Alfreya’s products first. Filters, feeds or copy drafts if you want them, set up with our team. 4. **Changes are read again**: A new photo or new copy gets that product read again. The rest is left alone. ## What’s included - **Gender and category**: Read from the photo and the title together, for each colourway. - **Clean taxonomy**: A fleece jacket typed “Kazak” lands in jackets. Raw category names fold into one tree you can rely on. - **Stylist descriptions**: Silhouette, fabric, mood and what it goes with, in a few plain sentences. - **A colour name per colourway**: Plain colour names, read from each colourway’s own photo. Colour-aware outfits and filters depend on them. - **Indexed by look and by meaning**: One index for how a product looks, one for what it is. Both carry price and stock. - **Open to other uses**: Clean filters, product feeds, consistent data across brands, first drafts of copy. Our team sets it up with you. ## Frequently asked questions ### Can we use it outside Alfreya’s products? Yes, with our team. Categories and colour names can drive filters and collections, labels can fill product feeds, and stylist’s notes can be a first draft of copy. ### Do we buy it separately? It comes with the Alfreya products that run on it. Other uses are set up on request. ## Pairs with - [Visual Similar Products](https://www.alfreya.com/md/en/solutions/similar-products): Its lookalikes are one scroll down. AI found them in your catalog. (Live with Vitruta) - [Outfit Recommendations](https://www.alfreya.com/md/en/solutions/complete-the-look): Two colours, six outfits. AI styled each one from your stock. (Live with Vitruta) - [Smart Search](https://www.alfreya.com/md/en/solutions/smart-search): A search box that forgives typos, plurals and a second language. (Live with Vitruta) --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/solutions/catalog-intelligence AI index (llms.txt): https://www.alfreya.com/llms.txt # Watches don’t sell like sneakers. > A watch buyer asks about case size. A sneaker buyer asks what goes with it. ## [Fashion & Apparel](https://www.alfreya.com/md/en/industries/fashion-apparel) Shoppers find a piece, see what goes with it and try it on, all from the store’s own catalog. Search, similar products, outfits and collection sorting are live at Vitruta today. ## [Footwear & Accessories](https://www.alfreya.com/md/en/industries/footwear-accessories) Alfreya puts shoes, bags and accessories inside whole outfits and next to close alternatives, where a shopper can picture them. Outfit recommendations and similar products already style footwear and accessories from Vitruta’s live catalog. ## [Fine Jewelry](https://www.alfreya.com/md/en/industries/jewelry) Pieces on a model without a shoot, try-on to scale, and a way for shop staff to identify a piece from a photo. The model studio, the product finder and the ear stack builder are live at Kısmet by Milka. Jewelry try-on is built and ready to deploy. ## [Watches](https://www.alfreya.com/md/en/industries/watches) A watch on the shopper’s wrist at its real case size, on-model photos, and identification on the shop floor, all from one catalog. They run on any platform and share the same search and recommendation layer as the rest of Alfreya. --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/industries AI index (llms.txt): https://www.alfreya.com/llms.txt # Fashion & Apparel > Shoppers find a piece, see what goes with it and try it on, all from the store’s own catalog. Search, similar products, outfits and collection sorting are live at Vitruta today. ## Where these stores lose shoppers - **Searches that dead-end**: Someone types “gömlekler” on a store full of shirts and gets an empty page. Suffixes, typos and English product names do that. - **Product pages that stop at one item**: Shoppers ask what to wear with it. Thousands of colourways, new stock every week: nobody styles that by hand. - **Nobody knows how it looks on them**: The shopper sees it on a model. Every new piece needs its own shoot, and the calendar is full. ## What we’d install - [Smart Search](https://www.alfreya.com/md/en/solutions/smart-search): A search box that forgives typos, plurals and a second language. (Live with Vitruta) - [AI Shopping Assistant](https://www.alfreya.com/md/en/solutions/ai-shopping-assistant): Search by sentence, by photo, or by “same but in black”. (Live with Vitruta) - [Visual Similar Products](https://www.alfreya.com/md/en/solutions/similar-products): Its lookalikes are one scroll down. AI found them in your catalog. (Live with Vitruta) - [Outfit Recommendations](https://www.alfreya.com/md/en/solutions/complete-the-look): Two colours, six outfits. AI styled each one from your stock. (Live with Vitruta) - [Virtual Try-On](https://www.alfreya.com/md/en/solutions/virtual-try-on): The shopper’s own photo, wearing the jacket from this page. (Available to deploy) - [Live Video Try-On](https://www.alfreya.com/md/en/solutions/live-video-try-on): A fitting-room mirror on the product page. (Available to deploy) - [Collection Merchandising](https://www.alfreya.com/md/en/solutions/smart-merchandising): New arrivals move up, broken size runs step back. Twice a day, each move explained. (Live with Vitruta) - [AI Model Studio](https://www.alfreya.com/md/en/solutions/ai-model-studio): One packshot, worn by a model, to scale. (Live with Kısmet by Milka) - [Catalog Intelligence](https://www.alfreya.com/md/en/solutions/catalog-intelligence): AI reads each photo and title. Our products run on what it writes, and so can your shop. (Available to deploy) Case study: [Vitruta: Five Alfreya products on a store of hundreds of brands](https://www.alfreya.com/md/en/case-studies/vitruta) --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/industries/fashion-apparel AI index (llms.txt): https://www.alfreya.com/llms.txt # Footwear & Accessories > Alfreya puts shoes, bags and accessories inside whole outfits and next to close alternatives, where a shopper can picture them. Outfit recommendations and similar products already style footwear and accessories from Vitruta’s live catalog. ## Where these stores lose shoppers - **Hard to picture in an outfit**: A sneaker on a white background says nothing about the trousers. Shoppers hesitate. - **Accessories as an afterthought**: Belts, bags and jewelry belong in the basket. They rarely show up on the product page where the decision happens. - **Sold-out sizes**: Their size is gone. With no close alternative on the page, they leave. ## What we’d install - [Outfit Recommendations](https://www.alfreya.com/md/en/solutions/complete-the-look): Two colours, six outfits. AI styled each one from your stock. (Live with Vitruta) - [Visual Similar Products](https://www.alfreya.com/md/en/solutions/similar-products): Its lookalikes are one scroll down. AI found them in your catalog. (Live with Vitruta) - [Smart Search](https://www.alfreya.com/md/en/solutions/smart-search): A search box that forgives typos, plurals and a second language. (Live with Vitruta) - [Collection Merchandising](https://www.alfreya.com/md/en/solutions/smart-merchandising): New arrivals move up, broken size runs step back. Twice a day, each move explained. (Live with Vitruta) - [AI Shopping Assistant](https://www.alfreya.com/md/en/solutions/ai-shopping-assistant): Search by sentence, by photo, or by “same but in black”. (Live with Vitruta) Case study: [Vitruta: Five Alfreya products on a store of hundreds of brands](https://www.alfreya.com/md/en/case-studies/vitruta) --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/industries/footwear-accessories AI index (llms.txt): https://www.alfreya.com/llms.txt # Fine Jewelry > Pieces on a model without a shoot, try-on to scale, and a way for shop staff to identify a piece from a photo. The model studio, the product finder and the ear stack builder are live at Kısmet by Milka. Jewelry try-on is built and ready to deploy. ## Where these stores lose shoppers - **Scale is hard to judge online**: Photographed alone, a 0.19 carat ring could be any size. - **On-model imagery is costly**: New pieces need model shots for the site and campaigns. A shoot for each one rarely fits the launch date. - **Similar pieces on the floor**: Two earrings, same shape, different stone. The customer is waiting for a price. ## What we’d install - [Jewelry Try-On](https://www.alfreya.com/md/en/solutions/jewelry-try-on): A necklace, ring or bracelet on the shopper, to scale. (Available to deploy) - [Ear Stack Try-On](https://www.alfreya.com/md/en/solutions/ear-stack-try-on): Shoppers stack your earrings on a reference ear, each to scale. (Live with Kısmet by Milka) - [AI Model Studio](https://www.alfreya.com/md/en/solutions/ai-model-studio): One packshot, worn by a model, to scale. (Live with Kısmet by Milka) - [In-Store Product Finder](https://www.alfreya.com/md/en/solutions/in-store-product-finder): Photograph a piece. See its price, stock and other sizes. (Live with Kısmet by Milka) - [Smart Search](https://www.alfreya.com/md/en/solutions/smart-search): A search box that forgives typos, plurals and a second language. (Live with Vitruta) - [Visual Similar Products](https://www.alfreya.com/md/en/solutions/similar-products): Its lookalikes are one scroll down. AI found them in your catalog. (Live with Vitruta) - [Catalog Intelligence](https://www.alfreya.com/md/en/solutions/catalog-intelligence): AI reads each photo and title. Our products run on what it writes, and so can your shop. (Available to deploy) Case study: [Kısmet by Milka: Model photos without a shoot](https://www.alfreya.com/md/en/case-studies/kismet) --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/industries/jewelry AI index (llms.txt): https://www.alfreya.com/llms.txt # Watches > A watch on the shopper’s wrist at its real case size, on-model photos, and identification on the shop floor, all from one catalog. They run on any platform and share the same search and recommendation layer as the rest of Alfreya. ## Where these stores lose shoppers - **Case size is hard to judge**: 42 mm means little until it’s on a wrist. An expensive purchase stalls right there. - **Many references, small differences**: Dial colours, straps and sizes turn one model into a dozen variants that look the same in photos and in the case. - **On-wrist shots for every reference**: They sell watches. Shooting a whole range is slow and expensive. ## What we’d install - [Watch Try-On](https://www.alfreya.com/md/en/solutions/watch-try-on): One photo of the wrist, then each watch at its case size. (Available to deploy) - [AI Model Studio](https://www.alfreya.com/md/en/solutions/ai-model-studio): One packshot, worn by a model, to scale. (Live with Kısmet by Milka) - [In-Store Product Finder](https://www.alfreya.com/md/en/solutions/in-store-product-finder): Photograph a piece. See its price, stock and other sizes. (Live with Kısmet by Milka) - [Smart Search](https://www.alfreya.com/md/en/solutions/smart-search): A search box that forgives typos, plurals and a second language. (Live with Vitruta) - [Visual Similar Products](https://www.alfreya.com/md/en/solutions/similar-products): Its lookalikes are one scroll down. AI found them in your catalog. (Live with Vitruta) --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/industries/watches AI index (llms.txt): https://www.alfreya.com/llms.txt # Two clients, up close > What runs at Vitruta and Kısmet by Milka, and what’s still on the way. ## [Alfreya × Vitruta](https://www.alfreya.com/md/en/case-studies/vitruta) **Five Alfreya products on a store of hundreds of brands.** Vitruta is a multi-brand premium fashion store on Shopify. Its search box, the similar products and outfits on its product pages, and the order of its collections all run on Alfreya today. The similar products and outfits draw on Alfreya’s catalog engine, which has read each product in the store. ## [Alfreya × Kısmet by Milka](https://www.alfreya.com/md/en/case-studies/kismet) **Model photos without a shoot.** Kısmet by Milka makes handcrafted fine jewelry and sells it online and in its own shops. The content team uses Alfreya’s model studio, the shop staff use the product finder, and shoppers build ear stacks from Kısmet’s earrings. A jewelry try-on is built and waiting for launch. --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/case-studies AI index (llms.txt): https://www.alfreya.com/llms.txt # Alfreya × Vitruta: Five Alfreya products on a store of hundreds of brands > Vitruta is a multi-brand premium fashion store on Shopify. Its search box, the similar products and outfits on its product pages, and the order of its collections all run on Alfreya today. The similar products and outfits draw on Alfreya’s catalog engine, which has read each product in the store. | | | |---|---| | Client | [Vitruta](https://vitruta.com) | | Industry | Premium fashion retail | | Platform | Shopify | | Live | Smart Search, Visual Similar Products, Outfit Recommendations, Collection Merchandising, AI Shopping Assistant | | Built | Virtual Try-On | ## The problem A big multi-brand catalog has all the discovery problems at once. One brand calls it a blazer, another a ceket. Gender is sometimes in the title and sometimes nowhere. Shoppers search in Turkish and English, and a product page shows one item when they want to know what to wear it with. Sorting dozens of collections by hand couldn’t keep up with new arrivals. ## What we built The visible parts went in one by one: Smart Search in the search box, an AI shopping assistant beside it, similar products and outfits on product pages, automatic sorting on the collections. Similar products and outfits run on Alfreya’s catalog engine, which labels each colourway by gender and category, folds the brands’ category trees into one and describes each piece the way a stylist would. Virtual try-on is built on the same catalog and waiting for launch. ## What we built - [Smart Search](https://www.alfreya.com/md/en/solutions/smart-search): A search box that forgives typos, plurals and a second language. (Live with Vitruta) - [Visual Similar Products](https://www.alfreya.com/md/en/solutions/similar-products): Its lookalikes are one scroll down. AI found them in your catalog. (Live with Vitruta) - [Outfit Recommendations](https://www.alfreya.com/md/en/solutions/complete-the-look): Two colours, six outfits. AI styled each one from your stock. (Live with Vitruta) - [Collection Merchandising](https://www.alfreya.com/md/en/solutions/smart-merchandising): New arrivals move up, broken size runs step back. Twice a day, each move explained. (Live with Vitruta) - [AI Shopping Assistant](https://www.alfreya.com/md/en/solutions/ai-shopping-assistant): Search by sentence, by photo, or by “same but in black”. (Live with Vitruta) - [Virtual Try-On](https://www.alfreya.com/md/en/solutions/virtual-try-on): The shopper’s own photo, wearing the jacket from this page. (Available to deploy) ## Highlights - **“çantalar” finds bags**: Suffixes, typos and synonyms in two languages are sorted out while the shopper types. - **Ask in a sentence, or with a photo**: Open to every shopper on Vitruta. AI answers in their language, with Vitruta’s products and prices. - **Similar products and full outfits**: A lookalike when the colour or size is wrong, and three outfits per colour, all from stock. - **Collections in a better order, twice a day**: New arrivals and full size runs move up. Each move comes with its reason. - **One catalog engine underneath**: Similar products and outfits read gender, category and colour from Alfreya’s catalog engine. - **Built, not live yet**: Virtual Try-On works on Vitruta’s product pages. Shoppers can’t use it yet. --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/case-studies/vitruta AI index (llms.txt): https://www.alfreya.com/llms.txt # Alfreya × Kısmet by Milka: Model photos without a shoot > Kısmet by Milka makes handcrafted fine jewelry and sells it online and in its own shops. The content team uses Alfreya’s model studio, the shop staff use the product finder, and shoppers build ear stacks from Kısmet’s earrings. A jewelry try-on is built and waiting for launch. | | | |---|---| | Client | [Kısmet by Milka](https://kismetbymilka.com.tr/) | | Industry | Luxury jewelry | | Platform | Shopify | | Live | AI Model Studio, In-Store Product Finder, Ear Stack Try-On | | Built | Jewelry Try-On | ## The problem A ring makes sense once you see it on a hand, but booking a model shoot for each new piece is slow and expensive. In the shops, many designs differ only by a stone, a setting or the colour of the gold, and staff need the right price, spec and stock while the customer is standing there. ## What we built We worked on both sides of the brand. The content team uploads one product photo to the model studio and gets the piece on a model, to scale, with no shoot. In the shops, an associate photographs a piece and the finder shows what it is, what it costs and whether it’s in stock. A try-on for necklaces, bracelets and rings is built on the same catalog, not live yet. ## What we built - [AI Model Studio](https://www.alfreya.com/md/en/solutions/ai-model-studio): One packshot, worn by a model, to scale. (Live with Kısmet by Milka) - [In-Store Product Finder](https://www.alfreya.com/md/en/solutions/in-store-product-finder): Photograph a piece. See its price, stock and other sizes. (Live with Kısmet by Milka) - [Jewelry Try-On](https://www.alfreya.com/md/en/solutions/jewelry-try-on): A necklace, ring or bracelet on the shopper, to scale. (Available to deploy) - [Ear Stack Try-On](https://www.alfreya.com/md/en/solutions/ear-stack-try-on): Shoppers stack your earrings on a reference ear, each to scale. (Live with Kısmet by Milka) ## Highlights - **Campaign imagery without a shoot**: One packshot becomes the on-model shot the team uses on the site and in campaigns. - **A photo instead of a price tag**: Staff photograph a piece and see its price, stock and other sizes. - **Try-on, built and waiting**: Necklaces, bracelets and rings at their real millimetres, on the shopper’s photo. Not live yet. - **Sizes from the product pages**: The studio and the stack builder use the millimetres Kısmet already lists, so a pendant on a model is the pendant in the box. --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/case-studies/kismet AI index (llms.txt): https://www.alfreya.com/llms.txt # What sits under all fifteen > Why the second product goes in faster than the first. ## What the platform does - **Reads the catalog**: Our catalog engine labels and describes each product once. Similar products and outfits read from it. - **Works with your platform**: Ready-made Shopify apps. Any other platform, connected with your team. - **Turkish and English**: From search grammar to the hints on the try-on camera. - **Online and on the shop floor**: The website and the staff app quote the same price. ## How products reach your store - **On Shopify**: Apps and theme blocks. Install from a link, place them in the theme editor. - **On any other platform**: Custom builds too. We connect your catalog and storefront with your developers. - **Standalone apps**: Model Studio and the in-store finder are web apps your team signs in to. ## On data - **Only the access it needs**: Each integration asks only for the access it uses. - **We don’t keep shopper photos**: A try-on photo makes one result, then it’s gone from our side. - **Stores are kept apart**: One store’s catalog and analytics never feed another’s. ## How a project goes 1. **We read the catalog first**: Products, attributes, how your collections are built, and what you need to sell more of this season. 2. **Then we pick one problem**: Usually the search box or the product page. Whichever one is losing you shoppers today. 3. **It looks like your store**: Recommendations follow your merchandising rules and sit in your theme. Shoppers shouldn’t be able to tell where Alfreya starts. 4. **We keep tuning**: After launch we watch what shoppers do with it and adjust. That part doesn’t end. --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/platform AI index (llms.txt): https://www.alfreya.com/llms.txt # Built into Shopify > Apps and theme blocks that sit inside your store. Not on Shopify? We work with other platforms too. ## Why we build inside Shopify - **Your theme, untouched**: Drag blocks into place. The theme code stays as it was. - **Prices are never stale**: Change a price at noon. Shoppers see it at noon. - **No new login**: The apps live in the Shopify admin you already use. Model Studio is the one with its own sign-in. ## Live in four steps 1. **Install the app**: From a link. It asks only for the permissions it uses. 2. **Connect your catalog**: AI reads and labels your products once. The rest builds on that. 3. **Place the blocks**: Drag them where you want them in the theme editor. 4. **Watch and adjust**: Reports sit in the admin. We go through them with you and tune. ## All fifteen, and where they live | Product | Where it runs | Setup | |---|---|---| | [Smart Search](https://www.alfreya.com/md/en/solutions/smart-search) | Search box and its drop-down panel | Install the app, flip one theme switch | | [AI Shopping Assistant](https://www.alfreya.com/md/en/solutions/ai-shopping-assistant) | Storefront search panel and a chat window | Install, index the catalog, flip one theme switch | | [Visual Similar Products](https://www.alfreya.com/md/en/solutions/similar-products) | A row on the product page | Three steps: install, let AI read the catalog, add one block | | [Outfit Recommendations](https://www.alfreya.com/md/en/solutions/complete-the-look) | A “Shop the look” block on the product page | Three steps: install, let AI style the catalog, add one block | | [Collection Merchandising](https://www.alfreya.com/md/en/solutions/smart-merchandising) | Collection pages and the Shopify admin | Three steps: install, pick collections, apply or schedule | | [Virtual Try-On](https://www.alfreya.com/md/en/solutions/virtual-try-on) | Product page, on the first product photo | App install and one theme toggle | | [Jewelry Try-On](https://www.alfreya.com/md/en/solutions/jewelry-try-on) | Product page try-on button and camera overlay | App install, theme toggle and product sizes | | [Ear Stack Try-On](https://www.alfreya.com/md/en/solutions/ear-stack-try-on) | A stack builder for desktop, tablet and phone | The product links of your earrings | | [Watch Try-On](https://www.alfreya.com/md/en/solutions/watch-try-on) | A button on the product page, or a standalone link | Case diameter per watch, set up with our team | | [Live Video Try-On](https://www.alfreya.com/md/en/solutions/live-video-try-on) | Product page camera overlay | App install and one theme toggle | | [AI Model Studio](https://www.alfreya.com/md/en/solutions/ai-model-studio) | Web studio for your content team | One team sign-in, nothing to install | | [Multi-Store Sync & Translation](https://www.alfreya.com/md/en/solutions/store-sync-translate) | Shopify admin app and product-page card | Install on both stores, approve, confirm the vocabulary | | [Checkout Consent](https://www.alfreya.com/md/en/solutions/checkout-consent) | Checkout, order record and confirmation email | Shopify Plus, a five-step guide and one checkout setting | | [In-Store Product Finder](https://www.alfreya.com/md/en/solutions/in-store-product-finder) | Associate app on phone or tablet, plus Shopify admin | App install, catalog sync and an access key | | [Catalog Intelligence](https://www.alfreya.com/md/en/solutions/catalog-intelligence) | Feeds Alfreya’s products, and other uses on request | Our team connects the catalog | --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/shopify AI index (llms.txt): https://www.alfreya.com/llms.txt # What we keep, and what we don’t. > Short answers for your legal team. Ask us for the long ones. ## How we handle data - **Each product keeps only what it needs**: Try-on photos are handled differently from catalog data. The table below goes product by product. - **Stores are kept apart**: One store’s index, recommendations and analytics never feed another’s. - **Only the permissions a product needs**: Each app asks for the access it uses. Smart Search asks for none. - **A “no” to tracking holds**: Our storefront analytics follow Shopify’s customer-privacy settings. No consent, no tracking. - **Where it runs**: Google Cloud and Vercel, in EU and US regions depending on the product. The AI search index sits in Warsaw, inside the EU. Ask us for the region of each product you use. - **Who we are, legally**: Alfreya is run by Konsiyer Teknoloji A.Ş. in Istanbul, under Turkish data-protection law (KVKK). ## Product by product | Product | What it handles | What happens to it | |---|---|---| | Smart Search | What shoppers type | Sent from the browser to your store’s own search. Searches are counted for your report. | | AI Shopping Assistant | Questions and photos shoppers send | Answered from your catalog. Questions are kept for your report. | | Visual Similar Products | Your catalog | Read and indexed for your store only. No shopper data. | | Virtual Try-On | The shopper’s photo | Sent once to our image provider to make the result. We keep no copy. The result stays in the shopper’s tab. | | Jewelry Try-On | The shopper’s photo | Sent to make the try-on. Photos and results stay in the shopper’s browser for up to 30 days. | | Ear Stack Try-On | Nothing from the shopper | Pieces are stacked on a reference ear. Nothing is uploaded. | | Live Video Try-On | The camera stream | Streams for a short, timed session. Never recorded. | | Checkout Consent | Buyer details inside the documents | Saved on the order in your Shopify store. We keep none of it. | | In-Store Product Finder | Photos staff take of a piece | Stored for your store, with the match the associate confirmed. | Questions about your data? We’ll answer for your store and the products you use, in as much detail as your legal team wants. info@alfreya.com --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/security AI index (llms.txt): https://www.alfreya.com/llms.txt # Alfreya Stylist: Where Alfreya started. > A free stylist at chat.alfreya.com. Ask in words, a photo or your voice. Open Alfreya Stylist: https://chat.alfreya.com/en | | | |---|---| | Products to discover | 200,000+ | | Ask with | Words, a photo or your voice | | Languages | English, Turkish | | For shoppers | Free to try | ## How shoppers use it - **Describe the occasion**: “What to wear to a rooftop party” works. - **Show a photo, or just speak**: It finds pieces that look like it. - **Refine in conversation**: Similar, cheaper or pricier. No starting over. - **Buy from the brand**: Checkout happens on the brand’s own site. ## Your store’s search was tested here first - **Photo and chat search**: Shoppers used them here before any store had them. - **Same search underneath**: Pointed at your catalog instead of hundreds of brands. - **Questions as people ask them**: Full sentences, two languages, a budget, a wedding. ## Put it on your store - [AI Shopping Assistant](https://www.alfreya.com/md/en/solutions/ai-shopping-assistant): Search by sentence, by photo, or by “same but in black”. (Live with Vitruta) - [Smart Search](https://www.alfreya.com/md/en/solutions/smart-search): A search box that forgives typos, plurals and a second language. (Live with Vitruta) - [Visual Similar Products](https://www.alfreya.com/md/en/solutions/similar-products): Its lookalikes are one scroll down. AI found them in your catalog. (Live with Vitruta) - [Outfit Recommendations](https://www.alfreya.com/md/en/solutions/complete-the-look): Two colours, six outfits. AI styled each one from your stock. (Live with Vitruta) Are you a fashion brand? We can list your products in Alfreya Stylist. Ask us how it works. https://chat.alfreya.com/en/brand-partner ## Questions about Alfreya Stylist ### Is Alfreya Stylist free? Yes. You can try a couple of chats without an account. A free account keeps favourites, collections and a style profile. ### Which languages does it speak? English and Turkish. Prices are shown in Turkish lira. ### How do I buy a piece? Tap it and you go straight to the brand’s store, where checkout happens. ### Can I offer the same experience on my store? Yes. The AI Shopping Assistant and Smart Search put the same kind of search on your store, using your catalog only. --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/stylist AI index (llms.txt): https://www.alfreya.com/llms.txt # Fashion software, made in Istanbul. > For fashion and jewelry stores, on the website and at the counter. ## At a glance | | | |---|---| | Founded | 2025 | | Headquarters | Istanbul, Türkiye | | Products | 15 | | Retailers in production | Vitruta, Kısmet by Milka | | Platforms | Any, with Shopify apps ready | | Languages | Turkish, English | ## Our story We started with an AI stylist for shoppers. It taught us how badly most software reads a fashion catalog. In 2025 we turned to stores. ## Why we exist Shoppers decide with their eyes. Most store software sees a dress as a row in a spreadsheet. Ours starts from the photo. ## How we build Inside live stores, with the retailer, starting from one sales problem. --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/about AI index (llms.txt): https://www.alfreya.com/llms.txt # Contact sales > Send the address and what bothers you about it. We’ll say where we’d start. - Email: info@alfreya.com - Office: Konsiyer Teknoloji A.S., Ataşehir, İstanbul, Türkiye - A person replies, usually within a business day. --- Alfreya (Konsiyer Teknoloji A.S.) · Founded 2025 · Istanbul · info@alfreya.com HTML: https://www.alfreya.com/en/contact AI index (llms.txt): https://www.alfreya.com/llms.txt