Chrome ends support for First Input Delay

In May, the Chrome team officially replaced First Input Delay (FID) with Interaction to Next Paint (INP) as the Core Web Vitals metric for measuring interaction responsiveness. As part of that plan, in January, Chrome announced its intention to officially end support for FID. Well, that time has finally come, and as of today FID is no longer supported in Chrome tools.

What is not changing is Chromium's support for first-input entries in the PerformanceObserver API. Developers who want or need to continue measuring FID in their applications will be able to do so.

What is changing is the availability of this metric in Chrome performance tools like PageSpeed Insights and the Chrome UX Report (CrUX). Unless otherwise noted, the following tools will stop reporting the FID metric over the next few days:

• PageSpeed Insights will no longer report FID data in the real-user experiences section.
• The PageSpeed Insights API will stop serving data for the FIRST_INPUT_DELAY_MS metric.
• The CrUX API and CrUX History API will stop serving data for the first_input_delay metric.
• The CrUX Dashboard will remove the First Input Delay page.
• The CrUX BigQuery project will remove the first_input struct from the schema starting with the 202409 dataset, which will be published in October 2024. Materialized and experimental tables will stop reporting new FID data as of the 202409 dataset, but historical FID data will continue to be available.
• The web-vitals.js library will remove the onFID function. Given that this is a breaking change, it will be released in the next major version, 5.0. Subscribe to release notifications in the repository for updates on version 5.0.
• The Web Vitals extension will no longer report FID data in any of its surfaces, including the overlay panel and console logs.

In addition to the tooling changes, there will also be some changes to the FID documentation on web.dev. The canonical FID article will remain available for posterity, however there will be a warning message to ensure that readers understand that its use is discouraged in favor of INP. The Optimize FID article won't remain available, given that all of its best practices have been superseded by INP, therefore it will permanently redirect to the Optimize INP article.

While we're officially closing the FID chapter in the Web Vitals program, the INP chapter is just beginning. Developers are finally taking notice of responsiveness issues that FID just wasn't capturing. These tooling changes to remove FID help to ensure that everyone is fully transitioned off of it and using INP as the best available metric to improve interaction performance.