The Data Architect Agent (DAA) empowers new customers to quickly establish their product data structure in Akeneo PIM.

By leveraging AI, it automatically generates an accurate data model based on your catalog extract, while streamlining the PIM initialisation and reducing the time to go live.
 

Check out our latest improvements to the feature!

 

December

Data Architect Chat

(December 19th)

The Data Architect Chat is a new tool designed to assist with Akeneo PIM data modeling. This feature is in an experimental stage for the moment and is intended to provide guidance and answer modeling questions.

After generating a data model, users can now consult the chat for guidance on their modeling and broader PIM concepts. Whether you are asking 'What is a family?' or seeking validation like 'Are my attributes well-defined?', the Chat is there to help.

Some things to keep in mind, to ensure a consistent experience, only the original creator of the data model can interact with the chat. All other users can view the conversation in read-only mode.

Also, each chat is unique to its specific data model, allowing you to keep your modeling discussions organized.

AI-Powered Mapping & Tailored Import

(December 15th)

The Data Architect Agent now features deeper integration with the Tailored Import, to significantly streamline your PIM setup and accelerating your time to go-live.

Here are the key benefits for our users:

• Maximum control: Users gain granular control to view and manually adjust the AI-generated mapping between source file columns and PIM attributes
• Reusability: The tailored mapping profile can be saved and reused to accelerate subsequent product imports.
• Scalability: The system effortlessly handles large-scale product imports, allowing you to manage massive data volumes without increasing manual workload or compromising performance.

This enhanced capability includes:

• - Automatic mapping: We’ve added some magic to automatically generate the mapping of attributes from the uploaded file using AI to save our users some precious time.
• - Generate Assets Mapping : From now on, users are able to map assets in addition to product attributes. This will allow them to directly create assets and automatically link them to products during your import.
• - Automatic Family Categorization for Product Mapping: If the uploaded file lacks a dedicated column for the PIM Family assignment, users can use this operation to let the AI determine the correct family. The AI will analyze the entire row of data to decide where the product belongs
   

• 

• 
  

  ‘Reset’ functionality on sandbox instances

  If your instance is a sandbox and already contains entities, you can leverage the 'reset' functionality to remove all previous entities.
  Please note that this feature is available only on sandbox instances when you apply a new data model to an environment that already contains entities (i.e. attributes, families, etc.), allowing you to easily facilitate data initialization.

   

October

Existing data model audit

(October 22nd)

We've enabled the ‘Audit’ tab to work for existing PIM configurations as well.This provides immediate value to current customers by allowing them to review and analyze the structure of their existing data model.
 

Locale selection for technical codes


(October 20th)
We now allow users to specify a dedicated locale for entity codes to ensure full control over technical codes that can now be generated in the language required to meet specific business and technical needs.

Attribute assignment

(October 15th)

We developed an ‘Attribute Assignment’ tab to manage attributes easily.This facilitates modifications on the data model and allows users to assign attributes to their families more easily.  

Product import with a sample of products

(October 14th)
The DAA now supports importing a sample of products during the apply phase. This provides users with an immediate, visualized first version of their catalog structure, accelerating feedback and project iteration.   
 

Source file

(October 13th)
We added a 'Source File' tab to display attribute mapping. This provides greater visibility by allowing users to instantly see which attributes from their input files were successfully mapped by the DAA and which were not.
 

September

Data model audit tab

(September 30th)
This new improvement provides a comprehensive overview of your generated data model with key metrics (e.g., attribute distribution, properties). It also gives you a detailed breakdown of shared and family-specific attributes.
 

Increase of the file size limit

(September 26th)
We've significantly upgraded the file size limit to 120MB. This allows users to successfully upload larger data files directly from their ERP, PLM, or e-commerce platforms. 
 

Support for assets families

(September 23rd)
We now support assets by identifying and creating asset collections for your files, such as images, based on the data provided in your input file.  
 

Brand-new user interface

(September 17th)
This new UI offers a cleaner and more straightforward experience, with an improved grid display and powerful filters that make it easier to iterate on your data model.  
 

August
 

Support for reference entities with records

(August 29th)
This improvement will allow the agent to suggest reference entities and their records directly from the input file that you have uploaded.
 

Shared database for the generated models

(August 29th)
Generated models are now stored in a shared database. Each user has the right to modify the models they have created themselves. However, models generated by other users are accessible in read-only and cannot be modified.
 

Contextual follow-up questions

(August 29th)
To ensure the model is as accurate as possible and to reduce any interpretation errors, we now ask a set of targeted follow-up questions before the model is generated. This helps us better understand your needs and propose a more relevant structure from the start.
 

Detailed source explanation

(August 29th)
We've enhanced the agent's ability to provide more precise information about your input data. It now shows you the exact location of the information within the original source document, making verification and understanding much easier.
 

Reasons for creation

(August 29th)
To help our users understand the logic behind the model's choices, we've integrated a reasoning for every entity creation. This explains the agent's decisions and adds transparency to the process.