Predictive AI: scoring, frequency, send time and bots
How to pick Engagement Scoring, Engagement Frequency, Send Time Optimization or Metrics Guard from a requirement, with each feature's edition, setup and data needs, and how the Einstein Decision element uses scores in a flow.
Given a scenario, determine which predictive AI feature to use.
~7 min read · 9 practice questions ·
What the exam expects
- Choose among Send Time Optimization, Metrics Guard, Engagement Scoring and Engagement Frequency from a business requirement.
- State which features each edition includes.
- Plan each feature's Setup path, identity resolution ruleset and global model decision.
- Explain why a contact has no score, and what a model is not intended to do.
- Configure Einstein Decision models, paths and data graph requirements.
- Distinguish the real-time and batch versions of Metrics Guard.
Key terms. Predictive AI: features that provide insights to inform the content and timing of marketing activities. Global model: a model trained on anonymized, aggregated data across customers. Model card: the page stating a model's intended use, data and limits. Data graph: the related data objects a flow can read. Einstein Decision: the flow element that routes people by engagement level.
Key facts
- Detail
- Send Time Optimization and Metrics Guard
- Detail
- Engagement Scoring and Engagement Frequency
- Detail
- Individual as primary object, among the first two rulesets created
- Detail
- Past 90 days; email only
- Detail
- One email send for a contact to have scores
- Detail
- 10 subscribers; five variants over 28 days
- Detail
- Scoring and Frequency update weekly; the cadence varies by one to several days
- Detail
- Scores every Engagement+ event in the Email Engagement DMO
- Detail
- Unified Individual
Sources for this section (7)
- AI data usagehelp.salesforce.com (opens in a new tab)
- Enable AIhelp.salesforce.com (opens in a new tab)
- STO cardhelp.salesforce.com (opens in a new tab)
- Scoring cardhelp.salesforce.com (opens in a new tab)
- Frequency cardhelp.salesforce.com (opens in a new tab)
- Metrics Guard cardhelp.salesforce.com (opens in a new tab)
- Einstein Decisionhelp.salesforce.com (opens in a new tab)
- Why shouldn't a team expect Send Time Optimization results straight after enabling it?
- Why is it risky to withhold messages because of a low predicted engagement score?
- Which data do these model cards say the histories exclude?
The questions are free to read. Answers with their sources and self-grading are included with the pack.
Keep reading with the Marketing Cloud Next Consultant pack
- Core concepts
- Exam traps
- Worked scenario
The pack includes:
- All 24 lessons of the Marketing Cloud Next Consultant guide, with reading progress
- Interactive decision walks and step-by-step sequence steppers
- Practice exams purchase adds inline bank questions and topic practice (9 questions for this lesson)
- An official source cited for every claim
- An AI tutor that answers from the guide
- Choose guide-only, practice-only, or save $4 with both