6.1. Exam Strategy

💡 First Principle: With 50–60 questions in 120 minutes, you have roughly two minutes per question on average — which means the biggest threat to a good score usually isn't knowledge gaps, it's time mismanagement on a handful of difficult questions that eats into time needed for easier ones later.

Flag-and-return is the core pacing tactic: the exam interface lets you flag any question and move on, then return to flagged questions at the end using whatever time remains. If a question's scenario is long or its answer isn't immediately clear, flag it and move forward rather than burning five minutes on one question while ten easier ones wait unanswered behind it.

Eliminate before you select. Nearly every question includes at least one obviously wrong distractor — a service that doesn't exist, or one that solves a completely different problem than the scenario describes. Removing those first narrows a four-option question into a two-option decision, which is a meaningfully easier problem than picking correctly from four options cold.

There's no penalty for guessing. An unanswered question scores the same as a wrong one, so never leave a question blank — even a pure guess after elimination has better odds than a guaranteed zero.

Read for the constraint, not the service name. The exam rarely asks "what does BigQuery do" directly — it describes a scenario with a specific constraint (global consistency, real-time sync, cost sensitivity, minimal management overhead) and expects you to work backward to the service built for that constraint. Every "Exam Trap" callout across Phases 1 through 5 exists because a scenario's phrasing is usually the whole clue.

✅ Best Practice: On your first pass through the exam, answer every question you're confident about immediately, flag anything uncertain, and do a full second pass focused entirely on flagged questions with whatever time remains — this guarantees your easy points are locked in before time pressure affects your judgment on harder ones.