How to Improve Einstein Prediction Builder Model Performance
An admin is training an Einstein Prediction Builder model for binary classification (lead conversion). The model performance is poor. Which THREE actions should the admin take to improve it?
Quick Answer
Ensuring the prediction field's outcome values, like whether a lead converted or not, are well represented in the training data is one of the three fixes here because a binary classification model can only learn a pattern it has actually seen enough examples of; if converted leads are rare in the dataset, the model has little to learn from and will lean toward predicting the majority outcome regardless of the real signal in the data. This sits alongside the closely related issue of overall data volume: more training records generally give the model more examples of the patterns behind lead conversion, reducing variance and improving how well it generalizes to leads it hasn't seen before. Both point at the same underlying idea, that a poorly performing binary model is very often a data problem before it's a configuration problem, and Einstein Prediction Builder's accuracy depends heavily on having enough records, and enough examples of both possible outcomes, to learn from. That's different from tuning settings within the tool itself, which won't help if the underlying dataset simply doesn't contain enough signal. When a scenario describes weak binary classification performance without describing errors in setup, look first at whether the dataset has enough records and a reasonable balance between outcome classes.
⚠ Common exam trap
The AI Associate exam often tests the misconception that reducing data prevents overfitting, but in Einstein Prediction Builder, overfitting is more commonly caused by too many irrelevant features or insufficient regularization, not by having too many records.
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
✓
Increase the number of records in the training dataset
Increasing the number of records in the training dataset provides more examples for the model to learn patterns from, which is critical for binary classification tasks like lead conversion. In Einstein Prediction Builder, a larger dataset helps reduce variance and improves the model's ability to generalize, especially when the initial performance is poor due to insufficient data.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Increase the number of records in the training dataset
Why this is correct
More data generally improves model accuracy.
- ✗
Use fewer records to avoid overfitting
Why it's wrong here
Fewer records typically underfit, not improve performance.
- ✓
Remove features that have little correlation with conversion
Why this is correct
Irrelevant features add noise and reduce performance.
- ✗
Add more features, even if they are not related
Why it's wrong here
Unrelated features add noise and can degrade performance.
- ✓
Ensure the prediction field value (e.g., converted) is well-represented in the data
Why this is correct
Balanced or well-represented outcomes are important for binary classification.
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
3 more ways 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 Salesforce admin is building an Einstein Prediction Builder model to predict whether a support case will be escalated (binary: Yes/No). The dataset includes cases from the past two years. After selecting the prediction field and features, the admin notices that the model's training score is very high (0.99) but the prediction score field shows very low confidence for new cases. What is the MOST likely cause?
hard- A.The prediction field contains data leakage (e.g., a future value)
- B.The admin selected too few features, causing underfitting
- C.The training data is stale and no longer reflects current case patterns
- ✓ D.The model is overfitting because the number of features is too large relative to the number of training records
Why D: A training score of 0.99 combined with low confidence on new cases is a classic symptom of overfitting. In Einstein Prediction Builder, when the model has too many features relative to the number of training records, it memorizes the training data instead of learning generalizable patterns, leading to poor performance on unseen cases.
Variation 2. An admin is using Einstein Prediction Builder to predict whether a case will escalate. They have selected the prediction field (binary) and the dataset. After training, they notice the model uses all available fields. What should they do to improve model performance and reduce noise?
hard- A.Increase the dataset size
- B.Use a different algorithm by default
- ✓ C.Select relevant features (input fields) and exclude irrelevant ones
- D.Change the prediction field to a different binary field
Why C: Einstein Prediction Builder automatically includes all available fields by default during training, which can introduce noise and reduce model accuracy. By manually selecting only relevant features (input fields) and excluding irrelevant ones, the admin reduces dimensionality, minimizes overfitting, and improves the model's predictive performance. This feature selection step is a standard best practice in machine learning to ensure the model focuses on meaningful predictors.
Variation 3. An admin is training a new Einstein Prediction Builder model to predict whether a support case will be escalated (binary). They have selected the prediction field 'Escalated__c' and the data set of all cases from the past year. Which step is essential to ensure the model can distinguish between escalated and non-escalated cases?
medium- ✓ A.Ensure the 'Escalated__c' field has both 'True' and 'False' values in the training data.
- B.Set the prediction window to the next 30 days.
- C.Select at least 20 features from the case object.
- D.Ensure the data set contains at least 500 records.
Why A: Einstein Prediction Builder requires the target prediction field to contain both positive and negative examples (e.g., 'True' and 'False') in the training data. Without both values, the model cannot learn the decision boundary between escalated and non-escalated cases, making binary classification impossible.
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.