Overview

Ask about this Page
Copy for LLM
View as Markdown

Make your product catalog discoverable and purchasable through AI commerce channels without replatforming your existing commerce stack.

AgenticLift connects a commercetools project to one or more AI shopping platforms through three core components. These components manage the lifecycle of an agentic commerce journey:

  • Agentic Activation Studio: Handles onboarding, including AI-powered data modeling via the Smart Data Modeler, catalog mapping, and platform authorization via the Agentic Channel Manager.
  • Universal AI Feed: The synchronization layer that extracts data from commercetools, transforms it using configurable JSONata mapping templates, and pushes it to AI platform endpoints. Price and availability updates propagate in near real-time.
  • Agentic Checkout: The transaction layer that exposes API endpoints compliant with the Universal Commerce Protocol (UCP). Orders are created in commercetools and can be written back to an existing OMS or ERP via webhook.
If your setup involves a product catalog already hosted on commercetools, use AI Hub instead of AgenticLift. Confirm the correct product with your Customer Success Manager (CSM) before starting.

Supported setup scenarios

The setup process applies to three different merchant scenarios:

ScenarioKey difference
No existing commercetools projectFull greenfield setup including all phases.
Existing commercetools contract (catalog not on commercetools)Requires a new, isolated project for AgenticLift.
Existing commercetools contract (new region or brand)Requires a new project or isolated Store configuration.

Operational constraints

The following technical limits apply to all AgenticLift implementations:

  • Full catalog sync: Limited to one synchronization per day.
  • Price and availability updates: Propagate in near real-time, up to every 15 minutes.
  • SKU-platform units: Usage is calculated by multiplying the number of SKUs by the number of connected AI platforms.
  • Single Product Type recommended: Multiple Product Types are supported but increase mapping complexity and catalog governance overhead.