Einstein Prediction Builder Setup: Steps to Create a Custom Model
A company wants to use Einstein Prediction Builder to predict whether a support case will escalate (binary: escalate or not). They have historical case data in Salesforce. Which THREE steps are required to set up this prediction?
Quick Answer
Selecting the prediction field, the binary flag that actually indicates whether a case escalated, is a required setup step because Einstein Prediction Builder needs to know exactly which field represents the outcome it's being asked to predict before it can do anything else with the data. That requirement goes hand in hand with another necessary piece: the underlying dataset has to include a real mix of both outcomes, escalated and non-escalated cases, since a model can only learn to distinguish between two classes if the historical data actually contains meaningful examples of each one. If the training data skewed heavily toward one outcome, or the field marking escalation weren't explicitly selected, the model would have no reliable target to learn against, no matter how much other case data was available. This is why setup for a binary prediction always starts with clearly identifying the target field and confirming the historical dataset represents both possible outcomes, rather than assuming the platform will infer the goal from context. The general lesson for similar Prediction Builder scenarios is to expect the correct setup steps to always include explicitly designating the outcome field being predicted and making sure the historical records actually contain a workable mix of both outcome values, since both are prerequisites the model cannot function without.
⚠ Common exam trap
Many candidates confuse optional post-setup activities like building dashboards or bots with the mandatory configuration steps required to create the prediction model itself.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Select the dataset (records to train on)
Selecting the dataset is a fundamental step in setting up an Einstein Prediction Builder model. You must specify which historical case records to use for training, ensuring the data includes both escalated and non-escalated cases so the model can learn patterns. Without a properly defined dataset, the prediction cannot be built.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Build an Einstein Bot for escalation handling
Why it's wrong here
Einstein Bot is a separate feature; not required for Prediction Builder.
- ✓
Select the dataset (records to train on)
Why this is correct
Yes, you need to specify which records are used for training.
- ✓
Select features (input fields)
Why this is correct
Yes, you must choose the fields that influence the prediction.
- ✗
Create a dashboard to monitor model performance
Why it's wrong here
Dashboards are not required for setting up the prediction; they are optional after deployment.
- ✓
Select the prediction field (escalation flag)
Why this is correct
Yes, you must choose the binary field that indicates escalation.
Go deeper
Related to this question
About these practice questions
One of 753 original AI Associate practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
Same concept, more angles
1 more way this is tested on AI Associate
These questions test the same concept from different angles. Work through them to make sure you can recognise it however the exam phrases it.
Variation 1. A company wants to use Einstein Prediction Builder to predict whether a support case will be escalated within the first 24 hours. Which field should be selected as the prediction field?
medium- A.A date field for the escalation date
- B.A text field with case escalation notes
- ✓ C.A checkbox field named 'Escalated within 24 hours'
- D.A numeric field representing escalation time in hours
Why C: Einstein Prediction Builder requires the prediction field to be a binary outcome (e.g., true/false, yes/no) that the model will learn to predict. A checkbox field named 'Escalated within 24 hours' directly represents the binary target (checked = escalated, unchecked = not escalated) needed for supervised classification.
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AI Associate practice question is part of Courseiva's free Salesforce certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AI Associate exam.