Software di etichettatura automatica dei prodotti per l'e-commerce

Automated product tagging uses AI to assign descriptive attributes like material, color, style and use case to every product in a catalog, from product images and existing product data. In Hypotenuse AI, tags follow your list of values, and the AI suggests new values for your team to approve. For the basics, read our guide to product tagging in ecommerce.

AI product tagging uses a model that reads product images and text to decide which tags apply. It picks up details no one wrote down, such as a spaghetti-strap top that should be tagged sleeveless.

A product tag is a single attribute value on a product. A linen shirt might carry Color: White, Material: Linen, Sleeve length: Short sleeve and Occasion: Beach. Shoppers use tags through filters and search, and search engines read them to understand the product.

You connect or upload your catalog and choose the attributes to tag. The AI reads each product's images and data, researches missing details on the web when needed, and maps every value to your list of values. Your team then reviews tags by confidence score and approves them in bulk. Tagging runs on the same engine as our product data enrichment.

Yes. Hypotenuse AI tags products in bulk, for catalogs from a few thousand SKUs to several million. Confidence scores keep your team's review focused on the tags that need a person.

Yes. The AI tags against your attributes, allowed values and formatting rules. When it finds a value that isn't on your list, it suggests the value for your team to approve, so your list stays under your control. Manage attributes and values with AI taxonomy automation.

Yes. Hypotenuse AI tags and localizes product data in over 40 languages, with vocabulary rules so key terms always translate the same way. See AI bulk translation.

Site search and recommendation engines both rely on product attributes. When every product carries complete, consistent tags, shoppers get more relevant results and similar items surface together. AI product categorization adds the category each product belongs to, which decides which tags apply.

Product image tagging uses AI to read each product photo and add attribute values, such as color, pattern, shape and style, to the product record. In Hypotenuse AI, every tag maps to your own attributes and list of values, so a color read from an image matches the value your filters already use.

Yes. When a supplier sends only a title and a photo, Hypotenuse AI still tags the visual attributes from the image. For details a photo can't show, such as exact dimensions, it researches manufacturer sites and spec sheets, and uses the UPC to match the right product when one is available. See product data enrichment.

Visual attributes such as color, pattern, shape and style. Take an eyewear line where the supplier row says only "Sunglasses 54mm". From one photo, the AI can tag frame shape, frame color, lens tint and rim type, each mapped to your list of values.

Every tag shows where it came from and how confident the AI is. Set a confidence threshold and your team reviews only the tags below it, then approves the rest in bulk. Hypotenuse AI also flags products whose data and image disagree, such as a color attribute that says red on a product that's clearly blue in its photo. The AI learns from your team's corrections over time.

Yes. In the Hypotenuse AI DAM, you can upload, organize and auto-tag every product image, and the AI links each image to the right SKU. When needed, tags read from an image are added to the product record as structured attributes, so a color the AI reads from a photo goes into your Color attribute.