Miros

Miros targets e-commerce retailers, especially visual categories such as fashion, furniture, footwear, and decor. The page says its AI reads behavioural signals like scroll velocity, clicks, dwell time, and ignored items to build a real-time preference profile per visitor, and that one intent profile powers search, browse, recommendations, and shop-the-look. It states that AI extracts visual and semantic attributes from product images without manual tagging. The page claims results from randomised A/B split tests, including +7-19% GMV uplift, +296% conversion versus Algolia, and +30% repeat customers, and says it can be deployed in about 2 weeks with JS widgets or a GraphQL API. Pricing is not stated on the page.

Key features

• Reads shopper intent from behavioural signals such as scroll velocity, clicks, dwell time, and ignored items
• Extracts visual and semantic attributes from product images without manual tagging
• Resolves contradictory signals into coherent preference profiles
• Single intent profile powers search, browse, recommendations, and shop-the-look
• Available as JS widgets or a GraphQL API, headless and traditional
• Modular, so retailers can start with one component and expand later
• Works alongside existing search such as Algolia or Elasticsearch
• Working demo on the retailer's own catalog before commitment

Best suited for

• Fashion retailers with visual product intent
• Furniture and decor retailers
• Footwear retailers
• E-commerce sites seeking to improve product discovery and conversion
• Retailers with large catalogs seeking personalised product discovery

Pricing

Not stated on the website