Modeling your product data faster with AI: Introducing the Smart Data Modeler

The challenges of modeling product data 

Structuring product data has always been a critical pillar of digital commerce success. Product data modeling, the strategic process of defining how product information is organized, stored and managed in an eCommerce or PIM (product information management) system, creates a standardized framework that supports product lifecycle management, marketing, and, naturally, enhanced customer experiences. 

In a nutshell, the product data model defines:

• How variants relate to base products.
• How attributes support search, filtering and personalization.
• How regional or channel-specific requirements are handled.
• How PIM, ERP and downstream systems integrate.

For enterprises that operate across multiple brands, regions and business models, the product data model establishes the foundation of scalability, localization, personalization and operational agility.

However, modeling product data has never been an easy endeavor, especially when enterprises migrate off a monolith like SAP, Adobe or Salesforce to a modular commerce platform, such as commercetools. For many businesses, adapting a product data model from a fundamentally rigid system to a fully customizable one is a task that requires months, data entry into spreadsheets, multiple workshops, manual corrections and, ultimately, nerves of steel. 

This Herculean effort also increases costs. It’s often the case that businesses hire specialists to run a project, which eats a considerable amount of IT budget, or run into the hidden costs associated with data inconsistencies, integration rework and missed revenue opportunities due to a prolonged time-to-market. 

The problem for enterprise IT and development teams starts with the beginning of the process: A blank schema and thousands of inconsistent SKUs that need to be adapted to a new model. And with the product data model being literally the foundation of everything, every decision weighs on product owners, solution architects and developers. Nerves of steel, indeed. 

So, what if AI could provide the initial legwork for product data modeling? 

A key strength of AI is its ability to recognize patterns in unstructured or inconsistent data that aren’t immediately visible to humans. It can process an enormous amount of data in record time, making spreadsheets and manual corrections a thing of the past. When applied to product catalogs, AI can spot: 

• Inconsistent naming conventions.
• Attribute sprawl.
• Legacy workarounds.
• Category drift over time. 

Instead of asking solution architects or developers to infer structure from chaos manually, AI can now ingest a sample dataset and propose:

• A normalized core product model.
• Suggested product types and attributes.
• Logical variant structures.
• A foundation aligned to how the business actually operates. 

Is this the final answer? It isn’t, but it provides a starting point from a blank canvas to an intelligent draft. For professionals restructuring a product data model, it means saving time, money and sanity. 

Introducing the Smart Data Modeler by commercetools

Smart Data Modeler is an AI-powered assistant that analyzes existing product data and proposes a core product data model in commercetools, aligned with your business needs. 

As a starting point, it analyzes your existing product catalog and suggests a structured set of Product Types and Attributes, detecting Product Attributes that can be defined at the Product or Variant level. 

It’s designed for enterprises that are:

• Migrating from monolithic systems to commercetools. 
• Expanding into new product segments or industries. 
• Evaluating whether commercetools can represent their catalog complexity. 
• Aligning PIM or ERP data for optimized storefront and downstream consumption. 

Using the Smart Data Modeler, teams can upload representative product data and receive a proposed model in minutes. Based on this initial draft, teams can refine and validate their product data model in days or weeks, rather than months. 

Using the Smart Data Modeler to prepare your catalog for AI channels

Each AI channel defines its own set of required and recommended product attributes. Products that don’t meet these requirements might not be displayed or recommended by AI agents.

Merchants can now convert their existing PIM export into a structured, LLM-optimized product type in minutes. The Smart Data Modeler prepares your product catalog for AI shopping platforms. 

When you use the Smart Data Modeler with agentic channels, it analyzes a sample of your existing catalog, maps your product data to the attribute structure your target platforms require, and generates a platform-ready Product Type for your Project. Where your catalog is missing mandatory fields, the mapping flags them so you can resolve the gaps before going live. You can then transform your full catalog to the new structure and import it into your Project.

This enables a faster setup, less dependency on technical resources and a cleaner path to being visible and shoppable on AI channels, directly reducing the time and cost of going live on agentic channels.

The benefits for technology leaders 

The Smart Data Modeler delivers on three core priorities of CTOs and CIOs:

1. Efficiency without compromising control

The Smart Data Modeler removes repetitive schema design work while keeping architects in control of final decisions. Teams get a structured, AI-generated baseline, but governance, refinement and architectural standards remain firmly in their hands.

Importantly, the Smart Data Modeler doesn’t use customer data to train models and files are automatically deleted after 30 days. Behind the scenes, the tool uses Google’s Vertex AI, which ensures data privacy. 

2. Reduced migration risk

When moving off platforms like SAP or Adobe, the largest hidden risk is mis-modeling data and discovering it late in the project lifecycle. AI-assisted modeling de-risks this foundational layer early, reducing costly rework and protecting delivery timelines.

3. Better developer experience

Instead of spending months designing and refactoring product types, developers start with a structured baseline and refine from there. This enables earlier API integrations, faster storefront progress and shorter feedback loops, accelerating the entire build phase.

4. Faster readiness for AI commerce channels

The Smart Data Modeler analyzes existing product data, maps it to platform-specific attribute structures and identifies missing mandatory fields before launch. This enables organizations to prepare catalogs for AI channels faster, reduce manual effort and accelerate time to market while minimizing costly rework during onboarding. 

The benefits for business operations leaders

While technical teams lead product data modeling, the organizational impact is felt across the entire business. Faster, more efficient modeling improves IT workflows and drives real operational advantages, such as:

• Faster go-live timelines: Projects launch sooner because the foundational product data model is structured and validated early, reducing delays caused by rework or misaligned attributes. AI-assisted modeling also prepares product catalogs for AI shopping and agentic commerce channels by mapping existing data to platform-specific requirements and identifying missing mandatory attributes before launch, accelerating time to market across both traditional and AI-driven commerce experiences.
  
  
• Reduced dependency on lengthy discovery workshops: Teams spend less time gathering and reconciling data manually, freeing operations and product teams to focus on strategy, marketing and expansion.
  
  
• Greater confidence when expanding into new regions or product lines: A standardized, well-modeled product data framework ensures regional requirements, localization and channel-specific needs are handled systematically.
  
  
• Improved alignment between PIM, ERP and commerce systems: With a consistent product model, downstream systems integrate more smoothly, reducing errors, data inconsistencies and operational friction.
  
  
• Enhanced organizational agility: Streamlined data modeling allows teams to adapt quickly to market changes, new business models or shifts in product strategy without starting from scratch.

How to get started with the Smart Data Modeler

The basic Smart Data Modeler is available for free  in our business tooling, the Merchant Center. You begin your product data modeling journey by uploading your product catalog data and answering a few questions. The initial response will include a structured set of Product Types and Attributes, and will detect Product Attributes that can be defined at the Product or Variant level. 

With that initial information, your team is poised to refine and validate the product data model quickly, align it with business requirements and accelerate API integrations and storefront development. 

What once took months of manual effort and workshops can now be achieved in days or weeks, giving both technical and operations teams the confidence and agility to scale commerce initiatives efficiently.

Ready to try it out? To enable the basic Smart Data Modeler for your Project, submit a request and include your Project key.