Loading
Salesforce now sends email only from verified domains. Read More
Set Up and Maintain Retail Execution
Table of Contents
Filter by (0)Add
Select Filters

          No results
          No results
          Here are some search tips

          Check the spelling of your keywords.
          Use more general search terms.
          Select fewer filters to broaden your search.

            Search all of Salesforce Help
            Search all of Salesforce Help
            Create Prompt Template: Recommend Products to Upsell

            You are here:

            1. Salesforce Help
            2. Docs
            3. Set Up and Maintain Retail Execution

            Create Prompt Template: Recommend Products to Upsell

            The Provide Upselling Assistance prompt template recommends top 10 products that a telesales rep can upsell. The recommendation also includes sales pitch for each product.

            Required Editions

            Available in: Lightning Experience

            Available in: Agentforce 1, Einstein 1, Enterprise, and Unlimited editions with Consumer Goods Cloud Retail Execution and Agentforce for Consumer Goods Cloud add-on licenses.
            User Permissions Needed
            To create a prompt template: Prompt Template Manager

            Make sure to turn on Einstein generative AI in Setup. See Set Up Einstein Generative AI.

            To create a prompt template for product upselling assisstance:

            1. From Setup, in the Quick Find box, enter Prompt Builder, and then select Prompt Builder.
            2. Click New Prompt Template.
            3. Select Record Summary as the prompt template type.
            4. Enter name and description for the prompt template.
              • Name: Recommend Products to Upsell
              • Description: Recommends the top 10 products, which includes the product sales pitch.
            5. For Object Type, select Account.
            6. Click Next.
            7. In the Prompt Template Workspace section,
              1. Enter your prompt. 
                You're a CRM assistant whose goal is to recommend products to help a tele-sales rep upsell during order taking. Recommend the top 10 products based on these criteria:
- Utilize historical order data to identify the most ordered products. 
Historical Data: {!$Apex:cgcloud__GenAIAccountSummarizationOrderData.Prompt}
- Use product survey data to recommend products that’re reported as out-of-stock often during visits.
Product Survey data: {!$Apex:cgcloud__GenAIAccountSummaryComplianceData.Prompt}
- Look at past visit notes to find problems or needs that were mentioned before but not fixed. This will let the tele-sales reps offer solutions or products that meet those needs.
- Use the past visit notes with product-related issues or feedback to recommend alternatives or improved products that address these concerns.
- Use the previous order delivery or invoice notes with issues or dissatisfaction to offer upgraded or alternative products that address those concerns and improve customer satisfaction.
Visit notes and order delivery and invoice notes data: {!$Apex:cgcloud__GenAIAccountSummaryHealthAssessmentData.Prompt}

- Include a sales pitch for each recommended products based on the results from the product survey, visit notes, order delivery, or invoice notes. A few examples of sales pitch are:
 Example 1: We've noted from recent surveys that your preferred product is currently out of stock. We have a new shipment arriving soon and can offer you a priority order to make sure you get it first.
 Example 2: We noticed from our recent delivery notes that there were some concerns with the previous shipment. We’ve arranged for an early delivery of your next order. 


Context:

* Account Name: Name of a customer.
* Order Name: Unique identification of the order record.
* Order Date: The date on which the customer commits to the sales document. For order entries, it’s the order date.
* Order Delivery Date: Delivery date of an order. By default, it’s the next day. Delivery date calculation is customized frequently.
* Delivery Recipient: Reference to the delivery recipient of the account. 
* Order Total Value: Gross total value (domestic currency) of the sales document including relevant sales taxes. The system calculates the value based on the sales document item template settings.
* Order Total Receipt Value: Total value of the order entry in the order or receipt currency.
* Order Name: Unique identification of the record.
* Product Description: A text description of the product.
* Product Category: Indicates the product category of the product. Possible values are:
 * Beverages
 * Chips
 * Detergent
 * Frozen—Frozen Goods
 * Hygiene
 * Snacks
* Product Brand: Indicates the brand of the product. Possible values for for the Beverages category are: 
 * Empower Cola
 * Empower Cherry
 * Dash Cranberry
* Product Quantity: Quantity of the product.
* Unit of Measure: Logistical unit of a product. Possible values are:
 * ConsumerUnit
 * Layer
 * Pallet
 * SalesUnit
* Base Price: Product base price per unit of measure.
* Discount: Manual discount by the user.
* Special Price: Specifies a price reduction for an item that's used to calculate the item's value. 


Rules:

* Format of the summary should be in bullet points.
              2. Select the resource as Apex.
              3. In the prompt description, make sure these invocable actions are selected:
                • GenAIAccountSummaryComplianceData
                • GenAIAccountSummarizationOrderData
                • GenAIAccountSummaryHealthAssessmentData
            8. We recommend to select Model Type as Standard and Models as OpenAI GPT4 Omni Mini as in the Template Properties section.
            9. Select an account for the Related Record lookup field.
            10. Select Save & Preview.
              In the Resolution section, you can view the data fetched by the Apex classes, and in the Response section, you can view the response from the LLM.
            11. Activate the prompt template.
             
            Loading
            Salesforce Help | Article