AWS Solutions Architect Associate Passing Score

The passing score is 720 scaled points

The minimum passing score for Solutions Architect Associate is 720 on a scaled range from 100 to 1,000. AWS uses scaled scoring so results can be interpreted consistently across exam forms that may differ slightly in difficulty. A score of 720 is not 72 percent and does not mean a candidate must answer 72 percent of the displayed questions correctly.

There are 65 presented items, but only 50 affect the score. The remaining 15 are unscored questions used for evaluation and are not identified. AWS does not publish a fixed raw number of correct answers that always maps to 720. Do not infer a raw threshold from the scale or from one practice provider’s percentage.

What scaled scoring means

AWS explains that score equating helps account for differences among exam forms. For example, correctly answering the same number of easier questions and harder questions does not necessarily demonstrate the same level of knowledge. The raw performance is statistically adjusted and reported on the common scale.

This means a practice result is useful for learning but cannot be converted directly into the official score unless AWS publishes a valid conversion for that product, which it does not provide here. A percentage on a homemade set is not a reliable pass prediction. Its domain coverage, difficulty, and item construction may differ from the exam.

The exam uses compensatory scoring

AWS says candidates do not need to pass every section. The exam uses a compensatory model, so the overall exam score determines pass or fail. A score report may classify performance by section to highlight strengths and weaknesses, but section results are informational and do not create separate passing thresholds.

The four domains have different weights: Secure Architectures 30%, Resilient Architectures 26%, High-Performing Architectures 24%, and Cost-Optimized Architectures 20% of scored content. These percentages describe the blueprint distribution. They do not say exactly how many questions an individual will see in each section, and they are not minimum domain scores.

If your score report shows a weaker cost section but a pass overall, that section is a development signal rather than a separate failure. If you do not pass, the feedback can help organize a new study plan. Do not treat it as a complete answer key or assume it reveals each missed question.

How to use practice results

Use practice scores to compare your performance across sets from the same resource and to identify recurring gaps. Review whether you chose a design that met the exact requirement, overlooked a failure mode, or selected a service that was unnecessarily expensive. Track domain and task patterns rather than converting the percentage into a guessed scaled result.

For example, if you answer correctly whenever the question asks about storage classes but miss scenarios about recovery, study the difference among durability, backup, replication, and high availability. A high total score may conceal that pattern. Conversely, one low score on a poorly matched practice set does not by itself prove you will fail the official exam.

AWS offers an official practice question set and a full practice exam through its preparation resources. Use them to learn the style and assess readiness, while keeping in mind that no practice score changes the official 720 threshold or provides a guaranteed outcome.

Result status and next steps

AWS reports a pass or fail outcome and a scaled score. Final results are posted in the AWS Certification Account within five business days. If you pass, the certification is valid for three years. If you fail, you may retake after the 14-calendar-day waiting period and pay the full registration fee again.

Do not confuse the passing score with certification eligibility. There is no separate experience application after the exam. The exam result earns the certification; AWS’s recommended experience describes preparation context, not a post-pass condition.

Treat the scaled score as the official measure A 720 is a score threshold on AWS's scaled range from 100 to 1,000. It does not mean 72 percent correct. The exam includes 50 scored questions and 15 unscored items, and the candidate cannot identify which items are unscored. AWS does not publish a raw-answer formula that turns a question count into an official 720 estimate. Answer each item seriously and avoid building a personal scoring model from remembered questions. Section-level feedback can help direct learning, but it is not a second pass standard. A lower result in one area does not prove that every question there was wrong, and stronger performance elsewhere does not reveal how many points it offset. Use the feedback to choose what to study next. If it points to resilience, practise availability and recovery decisions; if it points to cost, compare pricing patterns and resource use against the scenario. Question bank scores are useful when interpreted within the bank. If you answer 80 of 100 practice items correctly, that is 80 percent on that set, not an official scaled score. Compare similar fresh sets over time and review whether you can explain the best option. A familiar bank can overstate readiness when repeated questions reward memory. Examples of correct score interpretation A result of 720 meets the pass threshold. It does not show that 72 percent of 65 presented items were correct, because the score is scaled and some questions are unscored. It does show that the reported overall result meets AWS's stated standard. A result of 719 falls below the threshold. It is not necessarily one raw answer away from passing. The candidate should read the report and focus on domain feedback instead of trying to calculate which item caused the outcome. A practice test percentage rises after two weeks. That may show real progress if the candidate used new questions and can explain why wrong choices fail. If the second attempt reused the same items, the score may reflect recall. A stronger readiness check uses mixed scenarios under the same time pressure and tracks whether each choice follows the stated availability, security, performance, or cost requirement. Do not use section feedback as a precise score decoder. Turn a broad area into tasks and test those tasks with new examples. If the candidate repeatedly selects a multi-region design when a multi-zone pattern meets the requirement, the gap may be overbuilding rather than lack of service knowledge. Correcting that reasoning can improve preparation without guessing at a raw cutoff.

Treat the scaled score as the official measure A 720 is a score threshold on AWS's scaled range from 100 to 1,000. It does not mean 72 percent correct. The exam includes 50 scored questions and 15 unscored items, and the candidate cannot identify which items are unscored. AWS does not publish a raw-answer formula that turns a question count into an official 720 estimate. Answer each item seriously and avoid building a personal scoring model from remembered questions. Section-level feedback can help direct learning, but it is not a second pass standard. A lower result in one area does not prove that every question there was wrong, and stronger performance elsewhere does not reveal how many points it offset. Use the feedback to choose what to study next. If it points to resilience, practise availability and recovery decisions; if it points to cost, compare pricing patterns and resource use against the scenario. Question bank scores are useful when interpreted within the bank. If you answer 80 of 100 practice items correctly, that is 80 percent on that set, not an official scaled score. Compare similar fresh sets over time and review whether you can explain the best option. A familiar bank can overstate readiness when repeated questions reward memory. Examples of correct score interpretation A result of 720 meets the pass threshold. It does not show that 72 percent of 65 presented items were correct, because the score is scaled and some questions are unscored. It does show that the reported overall result meets AWS's stated standard. A result of 719 falls below the threshold. It is not necessarily one raw answer away from passing. The candidate should read the report and focus on domain feedback instead of trying to calculate which item caused the outcome. A practice test percentage rises after two weeks. That may show real progress if the candidate used new questions and can explain why wrong choices fail. If the second attempt reused the same items, the score may reflect recall. A stronger readiness check uses mixed scenarios under the same time pressure and tracks whether each choice follows the stated availability, security, performance, or cost requirement. Do not use section feedback as a precise score decoder. Turn a broad area into tasks and test those tasks with new examples. If the candidate repeatedly selects a multi-region design when a multi-zone pattern meets the requirement, the gap may be overbuilding rather than lack of service knowledge. Correcting that reasoning can improve preparation without guessing at a raw cutoff.