You have received a design review report. It contains findings, some of which are straightforward, some of which require judgement, and some of which may seem to conflict with decisions you made deliberately. This article explains how a report is structured, what the severity levels mean, and how to work through the findings systematically so that the review translates into a better design rather than a checklist exercise.
The cover page: the severity summary
A well-structured review report opens with a summary: total finding count and a breakdown by severity level. This gives you an immediate sense of the state of your design before you read a single finding.
A design with 2 critical and 5 advisory findings is in a very different position from one with 0 critical and 12 advisory findings. The first needs immediate attention on two blocking issues. The second is a clean design with room for polish. The summary tells you which situation you are in before you spend time on the details.
The summary also shows you the spread, if most findings are advisory, the design is fundamentally sound and the review is helping you refine it. If critical findings dominate, there are issues that must be resolved before fabrication.
Severity levels explained
Different reviewers use slightly different terminology, but the substance is consistent across the industry. Here is how we define each level:
| Severity | Meaning | Action |
|---|---|---|
| Critical | The design will not function as intended, or there is a safety concern. The board may fail immediately, fail under specific conditions, or pose a risk to the end user or connected equipment. | Fix before fabrication. No exceptions. |
| Minor / High | The design may work in some conditions but reliability or correctness is at risk. This includes marginal component ratings, missing protection, and practices that increase the probability of field failure. | Fix before fabrication unless you have a documented, specific reason not to. |
| Advisory | A best-practice deviation or a suggestion. The design will likely work, but the finding is worth considering. These often represent trade-offs the reviewer has identified where a different choice would reduce risk or improve performance. | Evaluate individually. Apply your engineering judgement. |
| Info | Context, background, or an observation that requires no action. Often includes datasheet references, clarifications of why a finding was assessed a certain way, or notes on the reviewer's assumptions. | Read and file. No action required. |
Anatomy of a finding
Each finding in the report follows the same structure. Understanding what each field is for helps you work through findings efficiently.
Reference
The specific component, net, or board area the finding applies to, for example, U3, pin 14, or NET: MAINS_L, or Connector J2 area. Use this to locate the relevant part of your schematic or layout before reading the rest of the finding. Trying to evaluate a finding without looking at the referenced location first is a common cause of misunderstanding a recommendation.
Analysis
What was found and why it is a concern. This is the most important part of the finding. The analysis explains the reviewer's reasoning, the electrical or mechanical mechanism by which the identified issue creates risk. Read this before reading the recommendation. If you understand the problem, you are better placed to evaluate whether the recommendation fits your specific situation or whether a different fix addresses the same root cause.
Recommendation
A specific, actionable fix. This is one valid solution to the identified problem, it is not always the only solution. If your design constraints make the recommendation impractical, or if you have a different approach that addresses the same root cause identified in the analysis, use your own solution. The recommendation is a starting point, not a prescription.
If a recommendation seems wrong for your application, re-read the analysis section first. The recommendation is derived from the analysis, understanding why the reviewer identified the issue often clarifies whether the recommendation applies to your specific operating conditions.
How to work through the report
The order in which you address findings matters. Here is a practical approach:
- Read the full report once before making any changes. Some findings interact. A critical finding in the power section may affect how you approach a minor finding in a downstream sub-circuit. A complete first read gives you the full picture before you start changing the design.
- Fix Critical findings first. These are blockers. Start here, confirm the fix addresses the root cause identified in the analysis, and document what you changed.
- Work through Minor findings. Most should be fixed. For any that you decide not to address, write a brief note explaining your reasoning, this is useful for your own records and for any future reviewer.
- Evaluate each Advisory finding individually. Some will be clearly applicable to your design and operating conditions. Others may be best practices that do not apply to your specific use case. Use your engineering judgement on each one.
- Read Info findings as background. They require no action but often contain datasheet references or context that is useful when making decisions about related findings.
What if you disagree with a finding?
Disagreement is a normal and useful part of the review process. Sometimes you have context the reviewer does not, a design constraint, an operating condition, a customer requirement, or a deliberate trade-off that was not visible in the submitted files.
When you disagree with a finding, the right approach is to reply with your reasoning. A good reviewer will engage with the pushback. They will either confirm the finding with additional technical detail that clarifies why the concern stands even given your context, or they will acknowledge that your context changes the assessment and update the finding accordingly.
What is not useful is silently closing a finding without addressing the underlying concern. If a critical finding identifies a real risk and you have a reason for not fixing it, that reason should be documented, for your own protection and for anyone who works on the design after you.
After the fixes
Once you have worked through the report, document which findings you fixed and which you accepted risk on, and why. This is particularly important for designs heading toward regulatory certification, a documented risk acceptance is defensible; an undocumented one is a liability.
If you made significant changes to the design in response to the review, not just component value adjustments, but architectural changes to power rails, topology changes, or major layout revisions, a follow-up review of the changed areas is good practice. The review process is most effective when it catches issues before fabrication, and significant redesigns introduce the possibility of new issues that were not present in the original submission.
A design review is not a pass/fail certification. It is a structured conversation between your design and a reviewer's experience. Used well, it is one of the highest-leverage steps you can take before committing to fabrication costs, assembly costs, and the time lost to a redesign spin.