Marqo AI features and pricing (October 2026)

Marqo is an AI product discovery platform for online retailers. It replaces keyword-based store search with a model trained on the retailer's own catalog and shopper behavior, and adds merchandising controls, recommendations and a conversational shopping agent.

Marqo is headquartered in San Francisco, with offices in London and Melbourne, and targets enterprise brands. A one-line pixel records clicks, carts and purchases, and Marqo trains a dedicated model on the catalog and that behavior. Modules cover site search, merchandising with ranking rules and A/B tests, category pages, recommendations and the conversational agent Sibbi. It connects through an API and one-click integrations for Shopify, Adobe Commerce and Salesforce Commerce Cloud. Case studies name KICKS CREW, Mejuri and MUJI. The open-source search engine is published on GitHub under Apache 2.0.

• Trains a search model on your catalogMarqTune fine-tunes a dedicated model on the retailer's product data and real shopper signals such as purchases and add-to-cart events. The vendor says it uses no shared model. Search then reflects the store's own assortment and vocabulary, which reduces manual synonym work.
• Understands natural and visual queriesThe model interprets intent, synonyms and typos, handles long natural-language queries and supports image-plus-text search. When a query would return nothing, Marqo suggests relevant products. Shoppers reach items faster and search teams maintain fewer rules.
• Automates merchandisingAI sets ranking, boosts, filters and collections from conversion data, while merchandisers keep rules for margin, stock and brand priority. Teams spend their time on strategy and review rather than on hand-tuning result pages across a large catalog.
• Guides shoppers in conversationThe agent Sibbi asks clarifying questions in natural language, suggests products from the catalog and can search by image. Shoppers describe a need instead of browsing many pages, and the store gets cross-sell suggestions grounded in real product relations.

• Trains a dedicated model per retailer on its own catalog and shopper data, which the vendor contrasts with shared models.
• Search, merchandising, category pages, recommendations and a chat agent come from one vendor and share one model.
• Integrations for Shopify, Adobe Commerce and Salesforce Commerce Cloud, plus an API for custom storefronts.
• The search engine is also published as open source under Apache 2.0 on GitHub.

Marqo names San Francisco as its headquarters, with further offices in London and Melbourne. That documents the United States as its home market; the support policy names an Australian company as the contracting party.

Prices are not public, so request a quote. The pixel collects shopper behavior, so check cookie consent and US state privacy rules such as the CCPA for your storefront. The support policy names S2Search Australia Pty Ltd as the contracting party, and the data processing agreement lists AWS, Google, Zendesk and Stripe as sub-processors. Region selection may be offered for some services, so confirm where your data is processed. The SLA states 99.9 percent monthly uptime.

No reviews yet. Be the first to share how your team uses it.

Every review shows its verification level. We read each one before it goes live. How we check reviews