Skip to main content
Enrich Products using AI
Updated over 2 months ago

Pre-set up

  • Add custom attributes as the base of your product data schema.


Enrich Product Data with AI


First, go to the AI Product Hub to visualize your products.

You may use the search box and filters to find the products you need to work with.

  1. Select the products you want to enrich.

    Alternatively, click on the 'More' dropdown under Actions for a specific product you wish to enrich, and then select 'Enrich Data' to trigger the AI Enrichment Service for a single product.

  2. Click 'Enrich Data' to trigger the AI Enrichment Service for the selected products.

  3. A pop-up window with user options will show up, for you to refine the enrichment task.

    1. The first option is whether you want the service to also look for your products across the web to use as a data sources for enrichment.

      This could be helpful if the product is widely available and its information exists publicly.

    2. If you choose Yes, you may refine further the web search to a selection of countries.

      This ensures local sources are preferred over international sources.

    3. Additionally, you may enter up to 10 preferred sources and up to 10 excluded sources.

      This is particularly helpful if you have previously identified good quality product data sources and you prefer the AI Enrichment Service to specifically look for your product on these websites.

      Similarly, if you have previously identified low-quality product data sources and you prefer the service to skip these pages, you may enter the websites as excluded.

    4. Once ready, click on Continue Enrichment to queue the task.

      You will be notified by email once the task is completed, else you may check after a few minutes on the All Products > Enrichment Tasks page.

Track Enrichment Progress

Use the left navigation panel to click on All Products > Enrichment Tasks to view the full list of enrichment tasks and their statuses.

  • There are 3 statuses possible under the Status header:

    • Pending: The task triggered is still in process or queued.

    • Finished: The task was successfully done.

    • Incomplete: The task fully/partially failed. The number count of failed products will show together. It is possible to retry the enrichment for the failed products for a specific task.

Check Enriched Result

  • Upon the enrichment process done (Finished or Incomplete), an email notification will be auto-sent to the registered email address of your account.


    Clicking on the email notification button will bring you to the AI Product Hub with the task ID applied as a filter, so enriched products for the task will be shortlisted.

    You may click on the Product ID to view the Product Detail page and the enrichment output for the selected product.

  • Alternatively, from the Enrichment Tasks you may click 'More' on the most right column under Actions header, where a dropdown of options will show up. Here you may click View Products.

    This will bring you to the AI Product Hub with the task ID applied as a filter, so enriched products for the task will be shortlisted.

Review Enriched data on the Product Detail Page

From the previous shortlist of products on AI Product Hub, click on the Product ID link of the product you wish to review. This will open up the product detail page for that product.

On the Product Detail page:

  1. Look for the Attribute tab look for the attribute filters at the top, so you can find empty attributes, with value, or produced by AI Enrichment (4)

    This will change the attributes visible across all the Display Groups to only contain those changed by the AI Enrichment Service.

    They are tagged with the star iconin between the attribute name and the value.


  2. Once you have hidden all the non-AI edited attributes, you may proceed to review each value and select Approve or Reject, depending on your acceptance criteria.

    Simply hover on the AI-edited value and a pop-up bubble will show up for your to review the options.


    Approve: if the AI value was satisfactory, click on Approve. This will make sure any future re-enrichment by AI does not change the approved value.

    The icon will change to green.
    This indicates to your fellow team members that this value is acceptable and indicates it should not be changed by any other user or the AI Enrichment Service.


    Reject: if the AI value is not acceptable, click on Reject. This will ensure future re-enrichment of the product retries to get a new value for this attribute.

    The icon will change to grey.
    This indicates to your fellow team members that this value was disapproved and will be updated by the next enrichment.


Did this answer your question?