Training Data
In Salesforce Einstein and AI, the historical dataset used to train a machine learning model, which the model analyzes to learn patterns and relationships that it then uses to make predictions on new data.
Definition
In Salesforce Einstein and AI, the historical dataset used to train a machine learning model, which the model analyzes to learn patterns and relationships that it then uses to make predictions on new data.
In plain English
“Training Data in Salesforce Einstein and AI is the historical dataset used to train a machine learning model. The model learns patterns from this data, and the quality and relevance of training data directly determines how well the model performs on new data.”
Worked example
At Beacon AI Labs, the ML engineering team curates the Training Data that feeds their Einstein Prediction Builder models. The team's discipline - tagging every record's source, validating label quality through a sample-based audit, retiring stale records as the underlying business shifts - is what produces models that hold predictive accuracy across quarters instead of decaying within weeks.
Why Training Data matters
In Salesforce Einstein and AI, Training Data is the historical dataset used to train a machine learning model, which the model analyzes to learn patterns and relationships that enable predictions on new data. Training data quality directly affects model accuracy: garbage in, garbage out.
Salesforce Einstein uses CRM data as training data for features like Lead Scoring, Opportunity Scoring, and Prediction Builder. The better your CRM data quality (accurate, complete, consistent), the better Einstein's predictions. Mature AI programs invest in data quality as foundational to model effectiveness.
How organizations use Training Data
Treats CRM data quality as foundational to Einstein model effectiveness.
Cleaned historical data before training Einstein Lead Scoring for better accuracy.
Invests in data quality as a prerequisite for AI initiatives.
Test your knowledge
Q1. What is Training Data?
Q2. How does data quality affect models?
Q3. What does Einstein use as training data?
Discussion
Loading discussion…