What's new with GTmetrix
Here you can find details on the latest major updates and features for GTmetrix.
New Performance Timings Available
GTmetrix now tracks more advanced timings with regards to your page load. These metrics provide even more accurate data by tracking key events.
To generate performance timings, you must be logged in to your GTmetrix account when testing URLs.
These are for technical users - they're meant for providing even more insight on your page load activity. Much like our Resource Usage Graphs, they are not scores to chase unless you have a strong development background.
In the GTmetrix Report page, you'll a find new tab called Timings with our unique visualization of your page load events. These events include:
- Redirect duration
- Connect duration
- Backend duration
- Time to First Byte (TTFB)
- First paint time
- DOM interactive time
- RUM Speed Index
Our unique timeline visualization gives you the event occurrences in order so you can see at a glance where they happen. Events are in order, however they are not to scale. This is due to webpages potentially having massive or negligible differences between events. We opted for to display these timings in a general order to illustrate what's happening as your page loads.
These performance timings might be new to you - so we did our best to explain each one. Hovering over each event will generate a description so that you can better understand how they affect your page performance.
Tests now defaulted to stop at Fully loaded time
GTmetrix previously used Onload time as the point to stop the test. Due to network variances or the way the page was designed to load (eg. asynchronous loading), resources loaded after the Onload time and not make it into the report, resulting in some pages yielding inconsistent results. Instead, we now wait for 2 seconds of inactivity before stopping the test (what we now call the Fully loaded time).
To explain why, we'll need to clarify the difference between Onload time and Fully loaded time:
Onload time - is when the processing of the page is complete and all the resources on the page (images, CSS, etc.) have finished downloading.
- This used to be the point where GTmetrix would stop the test and mark as the Page Load Time in the GTmetrix report.
Fully loaded time - is the point after the Onload event fires and there has been no network activity for 2 seconds.
- Essentially, GTmetrix is now waiting until your page stops transferring data before completing a test, resulting in more consistent page load times
- Possible issues with using this event: This event fires only when a page completely stops loading content, including ads and below the fold elements. Your site might have loaded fast above the fold and be usable, however since the analysis is now waiting for the entire site to stop loading data, your reported Page Load Time might be longer.
Again, GTmetrix reports from now on will have Fully loaded time as the default metric, along side Total Page Size and Number of Requests for your Page Details. However, you can change this to Onload time in the Page Settings, or default to it in the User Settings.
Common Questions About Onload vs Fully Loaded
You might have some questions about Onload time vs Fully loaded time. Here are some basic ones covered:
Which setting should I choose?
If you're seeing inconsistent results (incomplete screenshots, varying total page sizes/requests) from using Onload time, the switch to Fully loaded time should help make your results more consistent. We don't recommend using Onload time unless you have a specific use case for it.
GTmetrix says my site is taking longer to load! (as of Feb 8, 2017)
We switched over our Page Load Time metric to Fully loaded time on Feb 8th, 2017. Previously, GTmetrix was using Onload time to determine the Page Load Time number.
Here's why this number is larger or "longer" than Onload time:
Previously, GTmetrix used Onload time as the point at which to stop the test. However, due to network variances or the way the page was designed to load (eg. asynchronous loading), resources loaded after the Onload time and not make it into the report. The result is a GTmetrix report with fewer requests for resources, smaller total page size, and as a result, a faster page load time. This was the key reason why some screenshots were incomplete or page sizes and requests were inconsistent.
To make things more consistent, we have switched to using Fully loaded time, which waits for 2 seconds of network inactivity before stopping the test.
Changes with the new performance timings and Fully loaded time
With these new features affecting the core of GTmetrix data and monitoring, there will be a few adjustments in what you see in your GTmetrix usage. Here is a recap of all affected areas of GTmetrix with this updated:
- Page Load times display Fully loaded time.
- Reports that have Stop onload enabled will show the Onload icon in the options field.
- Old reports that do not have Fully loaded time data will show Onload time, marked with an asterisk.
GTmetrix Report Page
- Fully loaded time is used by default. You will see a tool tip alerting you of this change upon first visiting a report page.
- Reports prior to this update will have Onload time displayed.
- Waterfall Charts now also display First paint (green line) and Fully loaded (purple line).
- "Stop test Onload" option added to Page Settings.
- Your monitored jobs will analyze using Fully loaded time now.
- History graphs will display a new line graph (Fully loaded time - red line).
Added new timings to conditionals for alert triggers:
- Redirection duration
- Connection duration
- Backend duration
- Time to First Byte (TTFB)
- DOM interactive time
- DOM content loaded time/duration
- Onload time/duration
- Fully loaded time
- First paint time
- RUM Speed Index
- Added new timings to conditionals for alert triggers:
- Onload time will continue to display until enough Fully loaded time data is collected.
- Switch over will occur seamlessly.
- This is done to keep consistent comparisons.
- Added new performance timings to the API
- See our API Request Response page for more details.
Log in and try out our new performance timings and Fully loaded timings today!
Resource Usage Graphs Now Available
This release brings a new feature to GTmetrix, as well as a few fixes and updates.
We're happy to announce another set of metrics to help in determining performance issues: The Resource Usage Graph.
This is a companion to the Waterfall Chart available in the Waterfall tab on your Report page. At the bottom of the Waterfall Chart, you'll see three new graphs plotting the performance of:
Similar to the CPU usage graph, the memory usage graph shows the amount of memory required by your browser to render the page. It also can be used to get a heads up on inefficient or buggy code (e.g. memory leaks) and can give you a starting point on where to look if your website is performing slowly.
This graph shows you how much data your site transfers when it is requested. Use it together with our throttling feature in order to better understand how your page loads on slower connection speeds, like 3G connection or DSL connection. If your site maxes out bandwidth on desired connections types, you may see long bars in your Waterfall Chart.
Again - these graphs are to be used in conjunction with the Waterfall Chart. For example, a gap in your Waterfall Chart occurring with 100% CPU load in the Resource Usage Graph could bring attention to faulty or inefficient code.
Common Questions About Resource Usage
You might have some questions about resource usage. Here are some basic ones covered:
I just anaylzed my site - should I be worried about my CPU/Memory/Bandwidth levels?
It depends. Resource Usage Graphs are used to provide a little more insight on your current Waterfall Chart. They are more technical in nature, and rely on a multitude of variables within the site build and connection type. They are by no means a "grade" or "ranking-scale" - don't think that they must be optimized like your PageSpeed or YSlow score.
Worry about them when you've hit a roadblock in your optimization efforts, and need to dig deeper into your site load behaviour. You may find some indication of what's going on if you cross reference your Waterfall Chart with your Resource Usage Graph.
How do I fix my CPU/Memory/Bandwidth issues?
As stated above, these graphs can be indications for more technical issues. Things like memory leaks, buggy code, or demanding scripts could all be the cause of high resource usage. A developer would be best suited to troubleshoot and tackle these more involved issues.
Resource Usage Graphs are available for Registered users
Resource Usage Graphs are found in the Waterfall tab in the GTmetrix report page. Log in to your GTmetrix account and try it out today! If you're looking for monitored resource usage tracking, upgrade to GTmetrix PRO.
Other Changelog Items
- Upgraded Chrome and Firefox (to Chrome 54.0.2840.71 and Firefox to 49.0.2).
- Upgraded Chrome on mobile device to 54.0.2840.68.
- The upgraded Firefox should fix issues analyzing certain HTTP/2 sites (your users running older Firefox will still have the same issue, so this should be fixed).
- Improved connection throttling (more accurate downstream throttling).
- Fixed broken timezone handling in History graphs for certain timezones (eg. America/Montreal).
- Filmstrip now available in API.
- HTTP auth and Cookie data is now available in HAR (only for new reports). HTTP auth and cookie data will be hidden if you don't own the report.
Internationalized Domain Names (IDNs) Fully Supported!
Our IDN support was limited prior to this update, but with more and more IDNs emerging we felt it best to ensure full support for the growing number of IDNs.
If you have a website with an IDN, give it a try!
Other Changelog Items
- "Save" button is now sticky in the menus
- Your Saved URL sort order and number displayed on the Dashboard are now remembered
- Added an 'X' to clear the URL on search and URL inputs
We get a lot of feedback on PageSpeed and YSlow's recommendations (eg. "Specify image dimensions" and "Serve scaled images") in regards to their relevance with newer responsive sites. We believe most of these recommendations are still valid, so we've been working on improving and fixing bugs with PageSpeed and YSlow's recommendations to work better with modern websites.
We've made several fixes on the backend which should result in more accurate results. A big one being that PageSpeed's "Enable gzip compression" recommendation would trigger incorrectly if you served a WOFF font as text/plain.
We're a couple years late on this (we missed these updates), but we've upgraded to the latest PageSpeed 1.15. The main changes in this upgrade are:
- Overall scoring algorithm has changed. It is now generated based on estimated impact of all the recommendations. You'll likely see a drop in your PageSpeed scores that align a little better with what PageSpeed Insights gives.
- PageSpeed dropped individual recommendation scores, but we've come up with a new algorithm that's based on how the overall score is calculated.
- The "Avoid landing page redirects" recommendation now allows a single redirect on the main page without penalty.
- More asynchronous scripts were added to the "Prefer asynchronous resources" recommendation.
We've also made our own customizations and bug fixes to PageSpeed code to modernize the existing recommendations. We've made these publicly available on GitHub.
Here's a summary of our recommendation changes:
- "Enable gzip compression": Add support for detecting brotli.
- "Specify image dimensions": Now also checks dimensions specified in CSS.
- "Serve scaled images": Allow up to 0.5x scaling with responsive pages.
- "Enable Keep-Alive": Don't require for http/2 and trigger if "close" directive specified.
- "Specify a Vary: Accept-Encoding header": Don't require header for SPDY.
- "Avoid CSS @import": Fix triggering even though page wasn't using @import.
- "Avoid landing page redirects": Ignore Chrome's HSTS internal redirect.
We've also made some similar changes to YSlow to fix issues with some of the recommendations. Note that the removal of the 4 recommendations has made the left over recommendations have a higher impact on the YSlow score.
- "Compress components with gzip": Add support for detecting brotli and ignore WOFFs served as text/plain.
We're excited to finally release the version 2 update to GTmetrix! This redesign was driven by user feedback and responses from thousands of participants in our survey. Read on to see what we've done:
Updating the look of GTmetrix was the main focus of this update. Our original site design had a width of 960px - now we've increased the width of our layout to take advantage of the wider displays available.
Modern, flatter design
We've flattened things out and focused on a cleaner look. In addition, we've added some visual icons to help with feature and option recognition.
Colors are brightened and high contrast buttons emphasize important actions.
All our icons and graphics have been converted to SVG, ensuring great looking visuals at any size.
GTmetrix is now fully responsive! This was quite difficult given all the data GTmetrix generates, but we managed to deliver the core functionality of GTmetrix even at smaller screen sizes. We focused on fluid design as well, maximizing the use of screen real estate as much as possible.
Some data and functionality will be limited at the reduced display sizes. Features like History graphs, Waterfall graphs and others simply won't be functional at a small screen size, so either switch your device to landscape mode or have a look at the report on a full desktop screen.
At the focal point of this redesign was the Report Page - it's the most viewed page aside from the Home page. Here are some highlights on the work we did:
We moved a few features and common actions into their own dedicated sections for higher visibility.
In our survey, we discovered that a lot of our users were unaware of the additional features GTmetrix had to offer, so we took them out of the general "Page Settings" screen and gave them their own space.
You'll see new buttons for:
- Set Up Alerts
- Compare (Combines URL and Region comparison)
We also did some reorganization on the Report Page tabs:
- Added "Video" tab - Gave video playbacks their own home to increase visibility of the feature
- Changed "Timeline" to "Waterfall" - More conventional name
We switched from lightboxes to a sliding sidedrawer for all Page Settings and options. This change will make navigating settings and options a lot more streamlined - especially down to mobile device sizes.
New Features and updates
GTmetrix version 2 is primarily a visual update, however we were able to implement some new features and updates to make GTmetrix more usable.
Now you can easily find your reports with filters and instant search. We've also added sorting, paging and "Show Amount" options to help accounts with numerous reports (30+) find what they need more efficiently.
This new filter functionality also lets you create custom search filters. Use keywords and/or options to define the reports you want ot see, then save the filters into their own tab. You'll readily have your filters available for faster reference.
Improved Report PDF
We've made some general improvements to our PDF reports, including updating the look and feel to the new GTmetrix version 2 design, and implementing better logo handling for our white-label GTmetrix PRO users (better scaling, transparency enabled).
SSL for everything
To ensure the security of our users we've pushed everything to SSL.
Give GTmetrix version 2.0 a try today!
We've put a lot of work into making GTmetrix better and with your feedback, we hope this iteration makes optimizing your websites easier and more efficient!
Now offering Chrome as a browser option!
We're excited to add Chrome to our list of browsers for site analysis - See how your site renders on Chrome!
PageSpeed and YSlow scores should remain the same, however page load times and waterfall graphs might differ. Each browser renders sites in it's own way (different rending engines), so your site components might load differently. In addition, if you've set any browser-dependent content delivery or scripting, page loads will differ across each browser.
Compare for yourself
Take a look at GTmetrix's site across Firefox, Chrome for Android, and Chrome for Desktop.
Whitelist or Blacklist URLs in your tests!
Blacklisting useful for excluding 3rd-party scripts such as Facebook like/share widgets, Google Analytics scripts, Youtube embeds and any other external widgets you might want to be not be part of your page load.
Conversely, whitelisting a URL will ONLY serve resources from that domain.
Both support a wildcard (*) that will designate all resources that begin with the URL inputted as blacklisted/whitelisted.
How do I do it?
The Whitelist/Blacklist URLs feature is located in the "Advanced Options" dropdown on the Dashboard in the "Advanced" tab, or the "Page Settings" modal on the Report page also in the "Advanced" tab.
To Whitelist a URL
This means that GTmetrix will ONLY grab resources from the URL you designated for analysis.
Example 1 - Single URL:
http://www.domain.com/foo/ will result in testing that ONLY includes
resources referencing that URL. References to other URLs, even on the same
http://www.domain.com/bam/), are NOT included in the analysis.
Example 2 - Wildcard:
http://www.domain.com* will result in testing that ONLY includes
resources referencing that entire domain. The wildcard(*) designates that all
resources that begin with the inputted URL will be whitelisted (eg.
http://domain.com/js/script.js" will be part of the
report, where as
http://anotherdomain.com/image.jpg will not be).
To Blacklist a URL
This means that GTmetrix will EXCLUDE resources from the URL you designated for analysis.
Example 1 - Single URL:
http://www.domain.com/foo/ will result in testing that excludes
resources referencing that URL. All references to other URLs will still be
Example 2 - Wildcard:
http://www.domain.com* will result in testing that
excludes all resources referencing that entire domain. The wildcard(*)
designates that all resources that begin with the inputted URL will be
http://domain.com/js/script.js" will not be part of
Mobile Testing Now Available
We're proud to offer page testing through Android-powered mobile devices.
Currently, two Samsung Galaxy Nexus phones comprise our mobile testing suite, allowing users to generate GTmetrix reports based off actual page loads on these devices.
In addition, GTmetrix uses Google's PageSpeed rules with their prioritization for mobile devices. This ensures that characteristics of mobile devices and their connections are considered first in the recommendations. Google changes these rules often however, and some are experimental at the moment.
YSlow does not have a specific ruleset for mobile devices, so your score will remain similar to the desktop analysis.
More devices are planned for the future as usage increases, as well as more device options (like iPhones or other Android devices) as we mature the service.
How to do it
Our new mobile testing features can be accessed on both our Dashboard and Page Settings modal on the GTmetrix reports page.
Click on the "Advanced Options" tab under the Analysis field. Under the "Test Format" options, change the default "Firefox (Desktop)" option, with "Chrome (Android, Galaxy Nexus)", then click "Go!" A mobile report will be generated for you.
Mobile testing is still in Beta!
It was quite a technical challenge for us to implement actual mobile devices for analyses so please bear with us as we work out all the kinks. We should also note a few things:
- It's currently only available in our default, Vancouver, BC test region.
- We may have to take things down to make changes.
- We currently only have two phones available, so you may experience queues.
Currently, we're using our beta phase to experiment, test, debug and give our users a glimpse of how their sites perform on mobile devices. However, depending on popularity, mobile analyses may become a PRO only feature.
Simulate Different Connection Speeds for Loading Your Page
Currently, GTmetrix loads your pages at the full speed of our datacentres' Internet connections. This is fine with the generation of the PageSpeed and YSlow reports - as they currently do not have any rules that are affected by connection speed and latency - but the page timings and waterfall graphs are hugely affected by these factors.
You can now change the connection speed GTmetrix uses to analyze your page to simulate up to 5 different connections:
- Cable (5/1 Mbps, 30ms)
- DSL (1.5 Mbps/384 Kbps, 50ms)
- 3G Mobile (1.6 Mbps/768 Kbps, 200ms)
- 2G Mobile (240/200 Kbps, 400ms)
- 56K Dial-Up (50/30 Kbps, 125ms)
You will be able to see the effects of the connection speed in the page load times and waterfall graphs. Enable the video capture option to see these slower speeds affect your users!
How to do this
Connection speeds can be modified on both our Dashboard and Page Settings modal on the GTmetrix reports page.
Click on the "Advanced Options" tab under the Analysis field. Under the "Test Format" options, change the default "Unthrottled Connection" option, with your desired connection speed, then click "Go!" A report will be generated with your connection throttled to simulate a Dial-Up/2G/3G/DSL/Cable connection.
Get an Alternate View of Your Page Load With Filmstrip
Visualizing how your page loads is a difficult task. HAR files are too technical for some, and video playback is difficult to dissect (often a page goes from a blank screen to a complete page in a split second).
Filmstrip gives you another way to visualize your page load by generating snapshots of your site load - from the first request to page completion. and can reveal blocks, bottlenecks and any other loading issues.How to Generate a Filmstrip
View a GTmetrix report with a video created, then click on "Timeline". Click "Generate Filmstrip", and GTmetrix will create a frame-by-frame view of your page load.
Filmstrips can only be generated on Reports that have videos created on them - enable the "Create video" option in the Page Settings and "re-test" to create a video.
If a re-test is performed on a report with an existing filmstrip, you'll need to regenerate the filmstrip.
We've done a bit of clean up to our settings and options pages to make it easier and more logical to find the features you want.
You'll notice our Advanced Options tab has been improved on the Dashboard.
The introduction of tabs in the top right corner let's us put popular or most used settings in the foreground, and move more advanced and technical features into the background.
Setting up an Alert is now housed in the Page Settings modal, and is now an easier process with direct links to pre-requisite functions if you've not met the requirements for setting one up.
We hope these new settings and option layouts improve your GTmetrix experience. Happy optimizing!
New Video Playback Feature
GTmetrix now offers video playback of page loads! You can see how your page loads and pinpoint exactly where bottlenecks or stoppages occur.
Slow pages up to 4 times their original speed to determine issues, and jump to major page load milestones such as DNS Resolved, First Byte and DOM Loaded (only with Chrome, Safari or IE9+).
Other uses for your page load videos:
- View the difference latency has on your page loads (eg. loading from our different test locations) - this includes ads
- View the different ads or other regional based content
- View why the page was slow (when used with the monitored alerts option - note that this is a GTmetrix PRO only option)
When used with our monitored alerts feature, GTmetrix will save a video of the page load that triggered the alert for viewing (only available to GTmetrix PRO users).
API users can also add video playback to their feature list. It costs 5 API credits to generate a video recording of a page load.
Video playback is only available to registered GTmetrix users. Log in to your account and give it a try today!
Announcing GTmetrix PRO
We're proud to announce the release of GTmetrix PRO, a set of paid plans which give you more of all of the tools and services GTmetrix offers.
GTmetrix PRO lets you
- Monitor more URLs and track them hourly
- Save more performance reports for deeper, long-term analysis of your sites' optimizations
- Monitor your sites from a variety of global test locations to better understand how your sites perform for specific users.
Great for for developers and marketers
If you're using GTmetrix as part of your own branded solutions or applications GTmetrix PRO also gives you more API credits, so you can take more advantage of our tools.
White-labeled PDF reports are also available, allowing you to offer your clients GTmetrix insights under your own brand.
Go for PRO!
Take a look at the range of plans to find out which one suits your needs.
GTmetrix uses Jan Odvarko's HAR Viewer to give our users a way to visualize page loads via elegant waterfall graphs. Today, we're happy to announce HAR Diff, an add-on we wrote for HAR Viewer to allow the comparison of 2 waterfall graphs side-by-side!
Differences in loading elements are classified by colours:
- White: The resource status, size, or timings haven't changed significantly (small timing changes are ignored).
- Orange: The resource status, size, or timings have changed. Bolded orange text indiciates what has changed.
- Green: The resource was added in the second waterfall.
- Red: The resource was missing in the second waterfall.
- Blue: The resource is present in both waterfalls, but was loaded in a different order.
You can enlarge the HAR Diff by clicking on "View Larger Version", or open it in a new window to view the waterfalls in a full window.
We have submitted some patches to the HAR Viewer project (which have already been accepted) and will be releasing the rest of the HAR Diff code soon, so you can also add HAR Diffs to your project!
New Test Server Regions
We're pleased to announce new test server regions to analyze your URLs. In addition to our main Vancouver test region, you can now test your URL in Dallas, London or Sydney. Like the other test options, log in to access this new functionality.
Once in a report, you can compare the URL's performance to another region (using the same page settings).
A comparison will be generated with the 2 reports.
You can also change the default Test Server Region in your User Settings on the Dashboard. GTmetrix will use this selection whenever you analyze a new page.
Performance History graphs improvements
In this round of updates, we've added 2 major features to GTmetrix and fixed some bugs.
Complete History Viewing
Previously, you could only view the last 30 days of your reports in the "History" tab. We've now made it so you can view your entire report history. Some new UI elements are added to facilitate that process:
This is the date range selector. Click on the calendar to select the start date and end date of your range, and the graphs will regenerate to display all the reports within that range.
This is the zoom panel. You can select how close you would like to see your range of reports. From 1 day, to 1 year, to All. When zoomed in, you are also able to pan left and right using the arrows.
Another way to zoom is to click and drag a range on your graph. You'll zoom in on your defined range; use the Pan arrows to move left and right.
Averaged graph points
When there are several reports in close proximity within a range, they are averaged into a single point. You can zoom into that range by clicking on the "Zoom in" option in the tooltip.
You can now add notes and comments on your graph to help track changes.
In the History tab, click on the "Add Note" button, then move the cursor over the graph. You'll notice the pointer replaced with a crosshair. Click where you want to add the note. After a note has been added, it is indicated at the top of the graph with a speech bubble and a vertical blue line.
- Report data points are now aligned vertically across graphs
- Report times displayed on tooltip now correspond to the reports
Added Session Cookies and IPv6 support
Advanced users can now analyze logged-in pages (like Facebook profiles) by specifying cookies. This will simulate a logged in session and generate a GTmetrix report for that page. Read more in the FAQ
GTmetrix can now analyze IPv6 enabled websites:
Let's make the future of the web faster, too. Test your IPv6 enabled site today!
Added HTTP Authentication support
Page that are protected using an authentication mechanism (eg. Apache's htpasswd) can now be analyzed using GTmetrix! Just enter the username and password in the Advanced Options box and analyze the page as usual. Read more in the FAQ
New "History" Tab and Waterfall Graph Functionality
In this latest update, we've made some changes to the Breakdown section of the reports page to allow for a more streamlined user experience.
- The "Performance History" tab and "Report History" tab have been combined under a new tab called "History"
- A new tab has been added called "Timeline." This contains tab contains the waterfall graph.
The new "History" tab
The "History" tab contains all of the graphs that were previously under the "Performance History" tab. We've added tool tips to the nodes that allow you to:
- View the highlighted report (if you're not already viewing it)
- Compare the highlighted report to the currently viewed report
- Delete the highlighted report
This provides our users with a more visual way of keeping track of past reports.
The new "Timeline" tab
GTmetrix now generates a waterfall graph for each report. You can easily see how your site elements load and troubleshoot any problem areas that you may see. You can view a full version of the graph by clicking "View Larger Version" on the top right-hand corner of the graph.
Read more about how to read waterfall graphs here.