AI Lead Scoring Quiz

Answers and explanations appear as you take the quiz.

1. How does AI lead scoring primarily differ from traditional rule-based scoring?
   • It learns weights from historical conversion data instead of using fixed points
   • It only counts email opens
   • It requires no data at all
   • It ignores company size and job title
2. Which machine learning task best describes predicting whether a lead will convert?
   • Image segmentation
   • Binary classification
   • Speech recognition
   • Reinforcement learning
3. Why is plain accuracy a poor metric for evaluating lead-scoring models?
   • Accuracy directly measures calibration
   • Accuracy tells sales representatives why a lead will convert
   • Converters may be rare, so a model can appear accurate by predicting “no conversion” for nearly every lead
   • Accuracy automatically adjusts decision thresholds
4. According to HubSpot’s lead-scoring documentation, which data can a custom score use?
   • Only a person’s name
   • Only the time of day
   • Unlabeled audio waveforms
   • Record actions or properties selected for the scoring model
5. What can a lead-scoring model return for each prospect?
   • A score or estimated conversion probability, which still needs evaluation
   • A probability that is automatically calibrated by definition
   • A guaranteed sales outcome
   • A legal determination of creditworthiness
6. What does feature importance summarize in a tree-based lead-scoring model?
   • That a feature causes conversion
   • How influential a feature is in the model’s decisions under that importance method
   • That a feature is legally required
   • The future sales quota
7. What does calibration mean for a lead-conversion score?
   • Every lead receives the same score
   • The model’s accuracy is always 100 percent
   • Predicted probabilities match observed conversion frequencies over comparable cases
   • Scores are sorted alphabetically by company
8. Why should a lead-scoring model be monitored after deployment?
   • The original threshold can never become outdated
   • Monitoring makes labels unnecessary
   • A model score permanently captures market behavior
   • Lead behavior and the meaning of historical patterns can drift