New Approaches To Interaction Latency
Over the past two years, the lines between technical SEO as a pure process and wider marketing practices and functions have become increasingly blurred.
Core Web Vitals is one of the more prominent, with obvious considerations nesting in the intersection between technical SEO (page speed) and user experience. You can read more about Core Web Vitals here.
At the time of writing, we know that Core Web Vitals consists of three metrics; CLS, LCP, and FID. But how often do we come across FID issues when auditing websites?
The answer, it seems, ties in with my own experience, and that’s not often.
The lack of FID issue has also been acknowledged in the most recent HTTPArchive Almanac, whichcovers content management systems (CMSs) and raises the point that most out-of-the-box platforms score well on FID in chapter 16.
In fact, WordPress, Wix, Squarespace, Drupal, TYPO3, Weebly, AEM, and Tilda all scored 90% or above, with only Joomla and 1C-Bitrix below 90% (but even then, they were above 80%).
This lack of FID issue was also addressed by Google in November 2021, when they published an article asking for feedback and for developers to try out the new responsiveness metrics on their production websites.
In order to understand how Google may change FID, or abandon it in favour of other metrics, we need to understand the objective of FID.
FID Objectives
The Objective of FID is to measure the interaction time between the browser and any given website. For example, when a user interacts with the webpage and clicks on a button, FID can quantify the visual response time that a user will experience between action and result.
Typically, the main contributors to a poor FID are:
- Heavy JS bundles
- Render-blocking JS
- Lengthy JS execution times
- Slow server response times
And to overcome these issues we:
- Minimize main thread work (where possible)
- Reduce the request count
- Reduce the transfer sizes
- Remove unnecessary third-party scripts and codes from loading
- Reduce JS execution time by streamlining and removing unnecessarily heavy styling or unused code
So this is great! As SEOs, we know when this issue does arise that we have methods to reduce and resolve it, providing a better user experience in the process.
But the objectives of FID ultimately have to align with the objectives of the Core Web Vitals initiative as a whole.
Going back to April 2020 and the Web Vitals release post, Philip Walton (Google Engineer) wrote:
Web Vitals is an initiative by Google to provide unified guidance for quality signals that are essential to delivering a great user experience on the web.
This means that to develop a great user experience on the web, these metrics need to remain useful to help keep pushing the boundaries and keep up to date with the frontrunners in user experience delivery.
In the post, Walton includes two key points that mean that this entire post shouldn’t be that much of a surprise:
- The metrics that constitute Core Web Vitals will evolve over time.
- The current set of metrics for 2020 (and subsequently 2021) focuses on three aspects of the user experience.
So it’s no surprise that as one metric has lost its usefulness in terms of developing a better web, it will be replaced.
New Measures For Interaction Latency
So what could Google replace it with? In their November 2021 article, they talk about Maximum Event Duration and Total Event Duration as potential candidates.
Both event duration metrics would record from the user interaction (click, keystroke – also known as a hardware timestamp) to when the next paint is completed, after the event itself has been handled.
In instances where the interaction (event) doesn’t cause a new paint update, the metrics will measure from the hardware timestamp to the point when Google is sure that no update will occur.
And for reference, a hardware timestamp is a click, tap, drag, or keyboard input.
Maximum Event Duration
The Maximum Event Duration (MED) metric would quantify and report on the least responsive event within an interaction group.
Total Event Duration
The proposed Total Event Duration (TED) would quantify the sum of all event durations within the interaction group.
Next Steps
Google is actively wanting feedback on these metrics, and provides a JavaScript testing code and feedback mechanism here.
Outside of this, and as general best practices, it’s good to reduce latency on your initial page load and on how it changes and adapts based on user interaction.
By working to reduce any latency delays as much as possible, and applying general site speed best practices, you can be in a more robust position forwhen the new responsiveness metrics come into force.