Skip to main content

Semantic Data Model

DeveloperIntermediate~16 mins

Introduction

The Unified Marketing Intelligence Semantic Data Model (SDM) combines paid media, owned-channel engagement, and CRM data in one queryable semantic layer. You can use the SDM to report across channels with consistent measurements, shared dimensions, and supported join paths.

The SDM brings together Marketing Intelligence paid media data, Marketing Cloud owned-channel objects, and CRM objects. For example, you can compare paid media impressions, owned-channel clicks, and influenced pipeline by campaign without building a custom reporting model.

In this workshop, you’ll extend the SDM with a custom calculated field, then create a metric and dimension based on that field for use by agents and in dashboards.

Create Calculated Field

In this exercise, you will add a custom field to the Marketing Intelligence Custom Semantic Model in Tableau Next.

  1. Search and select Tableau Next from App Launcher .
  2. Click Browse on the page menu.
  3. Enter Marketing Intelligence Custom Semantic Model in the Search field, then select the matching record to open it.
  4. Click Edit.
  5. Click the icon in the Calculated Fields section of the Objects panel to create a new calculated field.
  6. Click Draft with Einstein, then copy and paste the following prompt into the text area field.
Create a calculated field named `Marketing Impact Score` that measures overall performance across branding, engagement, and conversion. Calculate the score on a 0–100 scale using these weighted metrics: Delivery Impressions (30%), Delivery CTR (25%) and Delivery ROI (45%).
  1. Click Draft with Einstein.
  2. Confirm that the generated formula looks similar to the expression below.
IF WINDOW_MAX(
  SUM([Delivery].[Delivery Impressions])) > 0 THEN (
  SUM([Delivery].[Delivery Impressions]) / WINDOW_MAX(
    SUM([Delivery].[Delivery Impressions])
  ) * 0.3 + IFNULL([Delivery CTR], 0) * 0.25 + IFNULL([Delivery ROI], 0) * 0.45
) * 100 ELSE 0 END
  1. Click Validate, then click Save.
  2. Click the icon in the Metrics section of the Objects panel to create a new metric.
  3. Enter Marketing Impact Score in the Name field.
  4. Enter the text below in the Description field.
The Marketing Impact Score (MIS) measures the overall effectiveness of marketing activities across impressions, engagement, and return on investment.
  1. Click Next.
  2. Select Marketing Impact Score from the Measure field.
  3. Select the Engagement Date Time calculated dimension as the Time Dimension.
Configure values for new metric
Configure values for new metric
  1. Click Next.
  2. Select the following dimensions in the Additional Dimensions menu:
    • Marketing Channel.Marketing Channel Name
    • Market Segment.Market Segment Name
    • Campaign.Campaign Name
    • Ad.Ad Name
    • Product Name
  3. Click Next.
  4. Leave the Create Goal toggle off, then click Next.
  5. Click Save.

Test Agent

In this exercise, you will validate that the agent understands the context of the MIS calculated field that you previously created.

  1. Search and select Marketing Intelligence from App Launcher .
  2. Click the Marketing Analytics tab.
  3. Click the Agentforce icon icon in the page header to open the agent panel.
  4. Choose the Marketing Intelligence agent, then click Select.
  5. Enter the following question in the chat text area, then click the send icon.
What is my MIS by marketing channel?

The agent should understand what the Marketing Impact Score (MIS) represents and provide recommendations.

Response for prompt 'What is my MIS by marketing channel?'
Response for prompt 'What is my MIS by marketing channel?'