Loading

Activate Einstein Predictive Intelligence for B2C Commerce

게시 일자: Nov 22, 2024
상세 설명

How does a site on a PIG instance (Development, Staging, and Production) get activated with Einstein for B2C Commerce?

Note: Einstein deployment is not available for sandbox instances.

솔루션

Prerequisite

In order for the deployment to be available, sites must be set to Online (Protected) or Online and the needed feeds must have been scheduled for Commerce Cloud Einstein Deployment.

Steps

Deployment Configuration
  1. Go to Business Manager | Administration | Operations | Einstein Status Dashboard
  2. On the right side of each site, use the dropdown arrow to select Configure
  3. The host field may be left as is, unless blank.
    • If blank, then populate with the current site's host URL (Ex: www.CommerceCloud.com. Do not include 'https://')
  4. Set Region to reflect the primary operating geography for that site (The region controls the Einstein data residency for your site)
  5. Select Out-of-Stock Products IF you want Einstein Product Recommendations to be generated on product detail pages of products that fall out of stock (This allows Einstein to still understand what products are out of stock)
  6. Select Variation Group Products IF your site's catalog leverages product type: Variation Group
  7. Select Multi-Locale IF your site has product catalog data beyond the default locale
  8. Define Export Orders Created After- (Note the configuration settings under Historical Order Configuration)
  9. Define Maximum Orders per Run- (Note the configuration settings under Historical Order Configuration)
  10. Click the switch/checkbox next to Schedule Catalog and Order Feed
  11. Use the new form fields to schedule the catalog and order feeds (Recommended to run daily at a time of low traffic, after any daily replication)  Note - Follow the steps outlined below under Catalog and Order Feed Configurations.
  12. Click Save


Historical Orders Configuration
  1. Export Orders Created After: This should be the lesser of the date your site went live on Commerce Cloud or 2 years back
  2.  Maximum orders per run: The anticipated number of orders on a peak day
    • The default is 10,000. 
    • If set to 100,000 orders and 100,000 orders are processed, the order feed processing time is 30 minutes
    • Important note: For new sites with large order history that you are looking to import into Einstein, it is recommended to increase the number of orders beyond 10,000, so that the order feed may import orders from the most recent date instead of taking many days to catch up to the current day. If the order feed is not up to date, you run the risk of Commerce Insights not generating, which displays shopper basket visualization for the latest 7 days. 


Catalog Feed Configuration
  1. Select the switch shown in the previous image to the right of 'Schedule Catalog and Order Feed' to begin feed setup
  2. Start Date and Time
    • Schedule the job to a time shortly AFTER your replication from staging to production job completes, so Commerce Cloud Einstein receives your most current product information or recent product launches
    • To view staging to production replication job completion time go to Staging Instance > Administration >  Replication >  Data Replication Processes > view End Time for Type: Transfer & Publication
  3. Time Unit: Day
  4. Frequency: 1
    • A daily catalog feed run is recommended, to allow any changes to get sent to Commerce Cloud Einstein
    • The entire catalog is only exported on the first run and deltas are sent after, which will keep the daily run time fairly low (Less than a minute for most catalogs)
 
Order Feed Configuration
  1. Start Date and Time
    • Schedule the job to a time that is off-peak ordering
  2. Time Unit: Day
  3. Frequency: 1
    • A daily order feed is important for keeping the feature: Commerce Insights up to date because the feature uses orders from the last 7 days. If a daily order feed is not used, then Commerce Insights in Einstein Configurator will not generate results
    • Also used for Product Recommendations strategy: Product Affinity Algorithm
  4. Click SAVE and your Commerce Cloud Einstein Engine start deployment automatically.

     

    If you don't have access to Einstein Configurator, please follow the steps in Provide Einstein Configurator Access for Salesforce B2C Commerce.

    Knowledge 기사 번호

    000391436

     
    로드 중
    Salesforce Help | Article