Deploying predictions with components and writeback
Deploying an Einstein Discovery model to a prediction definition (object connection, variable mapping, supplemental datasets, segmentation filters, actionable variables), then surfacing scores with the Einstein Predictions component, automated prediction writeback and bulk scoring.
Enable prediction features on Lightning record pages across Salesforce and CRM Analytics.
~7 min read · 7 practice questions ·
What the exam expects
- Deploy a model through the Deploy Model wizard: target prediction definition, object connection, variable mapping, segmentation filters and actionable variables.
- Use a supplemental dataset when a model variable has no field on the Salesforce object.
- Add the Einstein Predictions component to a Lightning record page and configure it for the model type, with the right permissions for admins and viewers.
- Decide when automated prediction writeback fits, including its multiclass restriction and field-level security step.
- Refresh stored scores with bulk scoring after a model changes.
Key terms: prediction definition (a container for one or more deployed models; each model scores a different segment); segmentation filter (a condition that picks the records a model scores); actionable variable (a factor users can control, needed for suggested improvements); supplemental dataset (a dataset that supplies variables not found on the object); prediction field (a writeback field only Einstein can write to).
Key facts
- Detail
- Customize Application
- Detail
- View Einstein Discovery Recommendations
- Detail
- Up to 10 active, one per segment
- Detail
- Case-sensitive exact match
- Detail
- Numeric and binary; not multiclass
- Detail
- All variables mapped
- Detail
- Only Salesforce sharing rules apply
- Detail
- Need an actionable variable; binary and numeric only
- Detail
- 1,000,000 records per rolling 24 hours
Sources for this section (6)
- Lightning Pageshelp.salesforce.com (opens in a new tab)
- Capacities and Requirementshelp.salesforce.com (opens in a new tab)
- Segmentation Filtershelp.salesforce.com (opens in a new tab)
- Prediction Writebackhelp.salesforce.com (opens in a new tab)
- Map Model Variableshelp.salesforce.com (opens in a new tab)
- Actionable Variableshelp.salesforce.com (opens in a new tab)
- Why can a component show a score while the record's prediction field still shows an older one?
- When does a bulk scoring job pause on its own?
- Which sharing applies to data read from a supplemental dataset at scoring time?
- What does Show prediction warnings tell a user?
- What happens next in the wizard when you choose Update an existing model?
The questions are free to read. Answers with their sources and self-grading are included with the pack.
Keep reading with the CRM Analytics and Einstein Discovery Consultant pack
- Core concepts
- Exam traps
- Worked scenario
The pack includes:
- All 26 lessons of the CRM Analytics and Einstein Discovery Consultant guide, with reading progress
- Interactive decision walks and step-by-step sequence steppers
- Practice exams purchase adds inline bank questions and topic practice (7 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