Skip to main content
Build the future with Agentforce at TDX in San Francisco or on Salesforce+ on March 5–6. Register now.

Deploy and Manage Commerce Einstein

Learning Objectives

After completing this unit, you’ll be able to:

  • Describe the Commerce Einstein tools.
  • List what you need to do to deploy and enable Commerce Einstein.
  • Explain how Commerce Einstein tracks activity.
  • Explain how you configure Einstein Recommendations.
  • Explain what happens when a shopper opts out of activity tracking.

What Is Commerce Einstein?

Commerce Einstein is a set of artificial intelligence-powered merchandising tools that can help enhance the shopper experience. Taylor Givens, as a senior merchandiser, couldn’t be happier. They are all about enhancing the shopper experience and can’t wait to use these tools to automate some personalization tasks. They are especially excited about using Commerce Einstein with their Salesforce D2C Commerce store.

Taylor Givens, Ursa Major Solar senior merchandiser

Commerce Einstein comes with two features: Activity Tracking and Einstein Recommendations.

  • Activity tracking, which Taylor can opt out of (but why would they?!) enables Commerce Einstein to collect anonymous metrics from shopper activity on the store and use these metrics as real-time data for Commerce Einstein features.
  • Einstein Recommendations enables Taylor to personalize the shopping experience by automating product discovery in the store. This is an Experience Builder component.

Here’s Taylor’s handy checklist to configure and deploy these tools.


Task

1

Make sure product data is ready.

2

Deploy Commerce Einstein.

3

Enable Activity Tracking.

4

Rebuild the Commerce Einstein Index.

Deploy Commerce Einstein

In the last unit, Taylor made sure the product data was ready. It turns out that this is really important when it comes to deploying Commerce Einstein. Before Taylor deploys Commerce Einstein, they make sure to:

  • Assign a catalog to the store.
  • Assign categories to the products.
  • Ensure that all of the products are active.

During deployment, Commerce Einstein progresses through three stages.


Stage

Description

1

Started

Commerce Einstein reads your catalog.

2

Processing catalog and Einstein data

Commerce Einstein processes product catalog data and runs Commerce Einstein automation.

3

Completed

Commerce Einstein was successfully deployed or redeployed using the latest catalog changes.

 During the processing catalog and Einstein data stage, you might see these errors.

Error Type

Resolution

Catalog error

Confirm your store catalog assignment and product activation, then click Reindex.  


If the issue persists, create a case and submit it to Salesforce Customer Support.

Internal error

Click Reindex to reprocess the data. 


If the issue persists, create a case and submit it to Salesforce Customer Support.

 Taylor, as the Commerce Admin, enables Commerce Einstein. Here’s how to do it.

  1. Open the Commerce App.
  2. Go to the store home page.
  3. Select Administration.
  4. Select Commerce Einstein.
    Deploy Commerce Einstein.
  5. In the Deploy Commerce Einstein section, click Enable Commerce Einstein.
    The Status bar highlights Commerce Einstein deployment status as Started.
Note

This process can take up to 24 hours to complete.

Activity Tracking

Activity tracking captures browser-based, real-time shopper activities on the store, unless a shopper explicitly changes a setting, not wanting to be tracked. This is important for European shoppers, whose privacy is protected by the General Data Protection Regulation (GDPR). This regulation impacts every retailer conducting business in the European Union (EU).

Commerce Einstein uses certain browser cookies and session storage objects to store and track browser-based activities and tailor shopper experiences. Some cookies are required, while others are functional. Functional cookies and tracked activities are used to personalize shopper data only when those shoppers have not explicitly disabled functional cookies for a site. Once accepted, cookies remain in the browser until they expire, or until the shopper deletes them.

If a shopper disables functional cookies, the store doesn’t use the cookies or tracked activities for that shopper. However, activity tracking continues behind the scenes and that data is used anonymously by Einstein for all shoppers. Disabled cookies might remain on a shopper’s browser, but merchants are unable to associate the tracked data with the shoppers who disabled the functional cookies.

Here are the cookies and activities that Commerce Einstein uses during activity tracking.

Data Object

Lifespan

Activity

Cookie Type

Description

cqcui

365 days


Functional

Provides a unique value for an unregistered shopper.

userId

Until deleted


Essential

Provides a unique value for a registered shopper.

viewProduct

7 days


How Commerce Einstein tracks the ID of products that shoppers view.

viewReco

365 days


How Commerce Einstein tracks when a recommendation displays on a page.


Commerce Einstein does not track whether a shopper scrolls to view a recommendation.

clickReco

365 days


How Commerce Einstein tracks when a shopper clicks on an Einstein recommendation.

Activity tracking captures browser-based, real-time shopper activities on your storefront, unless the shopper explicitly declines tracking. Remember to deploy Commerce Einstein before you enable it. Here’s how to enable it.

  1. Open the Commerce App.
  2. Go to the store home page.
  3. Select Administration.
  4. Click Commerce Einstein.
  5. Click to expand the Storefront Activity Tracking section.
    Enable the Storefront Activity Tracker.
  6. Select Enabled.

Einstein Recommendations

The Einstein Recommendations component personalizes the shopping experience and reduces the manual effort required to drive product discovery on your storefront. Taylor adds this component to the store page in Experience Builder. Here’s how they do it.

  1. Open the Store: Ursa Major Solar
  2. Click the Experience Builder tile.
  3. Click the Home tile and select the Product page.
  4. Click Components, then drag and drop the Einstein Recommendations component to below the Product Information component.
    Within the Experience Builder, drag and drop the Einstein Recommendations component to below the Product Information component.

Taylor can define the layout and the content. Here's how they do it.

  1. In Experience Builder, on the Product page, click the Einstein Recommendations component.
    In Experience Builder, configure Einstein Recommendations.
  2. Expand the Layout section.
    • Enter the Header Text: Product Recommendations
    • Enter the Maximum Products Visible: 4
    • Enter the value for Hide component if less than: 2
  3. Expand the Content section and select the Recommendation Type (strategy): Recently Viewed

Here are the recommendation types Taylor can select.

  • Recently Viewed
  • Similar Products: If the shopper views any product, then Einstein Commerce recommends products that match the category of the product viewed. For example, they view a solar powered power bank and are shown a solar hat charger.
  • Complimentary Products: If the shopper views any product, then Einstein Commerce recommends products that don’t match the category of the product viewed. For example, they view a solar powered power bank, and are shown a solar powered LED garden lamp.

Commerce Einstein uses strategies to determine which products to recommend. Taylor uses this table to decide the strategy to use. The number in the Strategies column represents the order in which Commerce Einstein uses a strategy.

Recommendation Type

Maximum visible products

Strategies

Recently Viewed

10

Last Browsed Products

Similar Products

20

  • Customers who Viewed also Viewed: Generates recommendations by analyzing the viewing behavior of other customers who viewed the same product.
  • Product Affinity Algorithm: Generates recommendations by analyzing the product's similarity to other products.
  • Text Similarity: Uses natural language processing to generate product recommendations based on similarity between products in product details.

Complimentary Products

20

  • Real-Time Personalized Recommendations: Generates recommendations by analyzing the customer's current viewing and past viewing behavior.
  • Customers who Viewed also Viewed: Generates recommendations by analyzing the viewing behavior of other customers who viewed the same product.
  • Text Similarity: Uses natural language processing to generate product recommendations based on similarity between products in product details.

With the Similar Products recommendation type, for example, Einstein Recommendations first looks for recommendations by checking what other customers look at who viewed the same product. If it comes up blank, it generates recommendations by analyzing the product’s similarity to other products. If it comes up blank again, it uses natural language processing to find recommendations.

Note

Text Similarity provides recommendations when there is no shopper activity data.

Rebuild the Commerce Einstein Index

After Taylor makes changes to their store's product catalog, they always rebuild the Commerce Einstein Index.

When Taylor deployed Commerce Einstein, they configured a weekly schedule to automatically update the catalog. You can manually update the catalog. Here’s how to do it.

  1. Open the Commerce App.
  2. Go to the store home page.
  3. Select Administration.
  4. Select Commerce Einstein.
  5. In the Deploy Commerce Einstein section, click Reindex.
    A message appears under the status bar to indicate that the reindex process has started.
Note

Reindexing can take up to 24 hours to complete.

Next Steps

In this unit, you learned how to deploy and enable Commerce Einstein, configure Einstein Recommendations, and rebuild the Commerce Einstein Index. Next, learn how to customize the store.

Resources

Comparta sus comentarios sobre Trailhead en la Ayuda de Salesforce.

Nos encantaría conocer su experiencia con Trailhead. Ahora puede acceder al nuevo formulario de comentarios cuando quiera desde el sitio de la Ayuda de Salesforce.

Más información Continuar para compartir comentarios