Ignore Known Defects and Customize Results with Powerful New Optimization Features
We’re happy to announce several new enhancements to the Optimization platform to help you tailor results to show only the findings that you care about most.
Mute Irrelevant Defects
We listened to Optimization users and learned that different teams care about taking action on different defects. Your team may know about an ongoing issue, and wish to hide that known defect from a report.
The new Mute command allows you to hide defects that you deem irrelevant or low priority. Muted defects will not display in any tables, charts, or reports in snapshot results. If priorities change you can use the Unmute option to restore hidden defects.
To Mute a defect:
- Select the Defects tab for any snapshot
- Check the defects you wish to hide, and
- Use the ‘Mute’ command from the toolbar.
Muted defects will then be hidden from this snapshot and from all future snapshots for any test using the selected Defect Check Policy that was used when creating this test.
Customize Defect Severity
Optimization ranks defect results by severity. This severity is defined by our algorithms, where more severe defects are generally those that offer the greatest optimization benefit for the least effort.
But we know that each website is different. You may decide any image optimization is critical, while optimizations to your CSS are less important.
With our new defect severity customizations, you can change the assigned severity for any detected defect, and those settings will persist to all future snapshots for all tests using that check policy.
To change the severity of a check:
- Select the appropriate defect from the Defects view
- Select the Severity option
- Set a new Severity for the check
Any defects with modified severity will be tagged in your results with a reminder about the change, and you can easily undo any changes later by hitting Reset.
Change How Defects Flag with Configurable Defect Thresholds
Many of the Optimization defects alert only when a certain threshold is met. For example, we can alert when compression savings are larger than a certain number of bytes or when cache durations are longer than a certain number of seconds. Each check is different based on what is being examined.
Until now, alerting thresholds were pre-configured in the application based on the best practices we’ve encountered from experience and industry research. Again, we know that each website is different. Some website owners may care if even the smallest lossless image optimizations are possible, while others may only want to focus on the truly extreme cases.
To help with this, we now offer configurable Defect Alerting Thresholds. Defect Thresholds can be set at two levels: a general level across the entire Defect Check Policy, or a specific level for a specific Defect Check (for example, Content without HTTP Compression).
Default Policy Thresholds
To set at the defect check policy level, go to Settings, then select Defect Check Policies and edit the policy of interest. Remember, Defect Check Policies are assigned to a Performance Test when you create that test. Each test has exactly one policy, although the same policy can apply to multiple tests. (More information about Defect Check Policies).
The default level is Standard, which is most similar to the behavior you are already used to. To see more defects, you can reset this value to Low. This is useful if you want to see as much detail as possible so you can optimize every last millisecond of speed from your site. If you don’t have a lot of time to make performance optimizations, you may instead want to use the High level. High will only alert on the most severe defects, which reduces the noise but may cause you to miss some of the smaller optimizations.
Disclaimer: Changes made here will only apply to future snapshots. Existing snapshots, will not be updated.
Override Defect Check Thresholds
While configuring a broad policy level threshold is quick and easy, it may not give you the level of control you are looking for. Now Optimization gives us the ability to override thresholds at the Defect Check level.
Not all Defects have thresholds, for example a Resource Not Found error is either there or it’s not; there’s no middle ground. To spot defects with thresholds, use the Has Threshold filter in the defects view or look for those defect rows with the Threshold icon displayed.
To customize a defect check:
- Hit the Threshold icon on any defect row, and
- Choose the appropriate new value.
A preview window will be displayed that shows the current threshold levels, and you can even supply a Custom level with your own specific values.
You can always review all the customizations you’ve made later by visiting your Defect Check Policy settings page, and you can undo any changes you’ve made using the Reset command on and all checks.
Note that when you change a threshold, it’s possible some defects that alerted earlier will no longer display. For example, if an image can be optimized 200kb and your threshold changed from Savings > 100kb to Savings > 300kb, that opportunity for 200kb savings is no longer considered a defect.
Defects in this state are called unflagged. Unflagged defects are simply defects that were found by Optimization but that are not reported in your results due to your threshold configuration. Still, we think it is important that you be aware of those defects not reported (especially if you wanted to fix them later), so any view that hides defects due to the “unflagged” status will show a link on that view to the number of unflagged defects hidden from that view. If you follow that link, you can then choose to further customize the threshold level to a value that causes that defect to reappear in your results.
Improved and Updated Defect Checks
One of the Optimization’s key advantages is the huge repository of performance defect checks that are analyzed, as showcased in our Performance Knowledge Base.
Having a large depth of rules to check for defects helps us identify and correct problems that other performance tools would never find, and we must also continuously focus to deliver relevant check results.
Because technologies and priorities change and evolve over time, we recently audited Optimization’s repository of performance defect checks. We removed obsolete checks, consolidated a number of related checks, and added new checks (especially around video).
Now you will see new check rules and results in future snapshots. Some checks will be marked as Deprecated in historical snapshots.
Disclaimer: As a necessary result of the defect check improvements, you are likely to see different results when comparing performance snapshots made before this release to those after this release. The Optimization score and number of reported issues could be different. While we endeavor to keep results as consistent as possible over time, we hope you also understand the ongoing need to keep our results accurate and timely with the latest advances in website performance recommendations. All snapshots made after this release will show consistent and comparable results.
We hope you find these new customization tools helpful in personalizing your Optimization reports to the exact level of detail you expect. Over time we’ll continue to improve these tools based on user suggestions.