hardMultiple Choice
PMLE Practice Question: An e-commerce company uses a Vertex AI endpoint…
An e-commerce company uses a Vertex AI endpoint for product recommendations. Recently, the click-through rate (CTR) dropped significantly. Model monitoring shows no significant data drift or skew. Logs show increased latency but no errors. Which technique should the engineer use to diagnose the issue?
⚠ Common exam trap
Google Cloud often tests the distinction between data drift (input distribution changes) and prediction drift (output distribution changes), and candidates mistakenly assume that no data drift means the model is fine, overlooking that the model's predictions can still degrade due to concept drift.
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
✓
Analyze the prediction output distribution using Vertex AI Model Monitoring for prediction drift and compare to a baseline.
The drop in CTR despite no data drift or skew suggests that the model's predictions have shifted in distribution (prediction drift), even if the input features remain stable. Vertex AI Model Monitoring can compare the current prediction output distribution against a baseline to detect such drift, which directly explains the CTR decline. The increased latency is a symptom, not the root cause, and fixing latency alone would not restore CTR.
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 endpoint's request timeout value to accommodate the higher latency.
Why it's wrong here
Raising the request timeout only lets callers wait longer for slow responses; it does not identify why latency rose or why CTR fell. Timeouts are the correct fix when clients abort before a legitimately slow response completes, not when diagnosing an unexplained latency increase.
- ✗
Enable autoscaling on the endpoint to reduce latency by adding more nodes.
Why it's wrong here
Autoscaling adds nodes to serve more concurrent requests, but the stem reports increased latency without errors, so capacity is not proven to be the constraint. Autoscaling is correct when CPU or accelerator utilisation saturates under load; here the cause remains undiagnosed.
- ✗
Retrain the model with the most recent user interaction data.
Why it's wrong here
Retraining addresses data or concept drift, yet monitoring already shows no significant drift or skew, so it cannot explain the CTR drop. Retraining is the right response when validation metrics degrade from stale training data; here latency, not model quality, is the anomaly.
- ✓
Analyze the prediction output distribution using Vertex AI Model Monitoring for prediction drift and compare to a baseline.
Why this is correct
With drift and skew ruled out, prediction drift isolates changes in the model's output distribution versus baseline. Comparing score distributions reveals whether the model itself shifted behaviour, explaining the CTR drop despite stable inputs and no errors.
Go deeper
Related to this question
About these practice questions
Courseiva writes every PMLE question from scratch — 775 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or dumps. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This PMLE practice question is part of Courseiva's free Google Cloud 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 PMLE exam.