Google is adjusting the standards of what contributes to a good user experience, and in turn has replaced one of their Core Web Vitals metrics that measures responsiveness (First Input Delay – FID) with a new metric, Interaction to Next Paint (INP) in March 2024 after two years of experimenting and gathering feedback from the community.
Measuring the INP metric is already possible on Grail, and a recent update has brought this functionality to Dynatrace Managed as well.
In this blog post, we want to dive deeper into how to utilize this new metric within Dynatrace and reap the benefits of the improved responsiveness metric.
FID vs. INP
Though FID is also a responsiveness metric, FID was only used to measure the input delay of the first interaction on a page, while INP observes all interactions on a page. That includes the input delay all the way up to the browser displaying the frame. So, while FID was an indicator to assess a user's first impression of the page, INP is designed to measure the overall responsiveness of a page.
The intended goal of introducing INP as a Core Web Vital metric is to “lead to a more responsive web over time”.
Since a majority of a user’s interactions take place after a page loads, we need to consider the responsiveness of the page throughout its entire lifecycle. Unlike measuring the input delay only on the first interaction on a page (FID), the new metric observes all interactions on a page until it is loaded (INP). The page’s INP is only calculated once the user leaves the page and is presented as a single value number to gauge the responsiveness of a page, indicating the longest duration of an interaction observed. Google includes actions like clicking, tapping, and keyboard interactions in the INP measurement, while actions like zooming, scrolling, and hovering are excluded. Since we leverage the Google API to ingest the INP value, these actions will also be reflected in Dynatrace accordingly. With this metric, low numbers mean that a page was reliably responsive to user input.
INP within Dynatrace
Customers are using Dynatrace to analyze Core Web Vitals in their web applications. Naturally, that includes INP as well. Because Google is ranking websites based on the results of their Web Vitals, specifically customers in the retail industry have a vested interest in keeping these scores above average and optimizing performance.
Tracking and visualizing this metric will enable users to find out the slowest interaction on a screen, figuring out where customer experience goes from excellent to bad.
With this update, users will be able to visualize INP on the following levels:
- with the standard web dimensions per frontend application in the data explorer and on dashboards (incl. percentiles)
- in central frontend application screen and page group multidimensional analysis
- on page level screens
Please keep in mind that INP is not a user action-based metric, so any filters applied for user actions or action type will not be supported for INP.
Selecting the data explorer will give you the option to visualize Interaction to next paint split by web application. In this example, we can see from the data provided, that in the last 72 hours there was an interaction that would not be considered satisfactory with a value of 536 ms. Based on this finding, we can go ahead and, for example, look for the corresponding user session to find out what went wrong.
Once you have your query configured as intended, you can pin it to your dashboards to keep an eye on everything in real time.
What's next?
Next, we are working on adding the additional “Browser” dimension to this metric to widen your scope to analyze your environment. This dimension will enable you to split this metric by different browsers, as well as synthetic monitors.
Additionally, we will add a drop-down menu to easily select the aggregation type.
Get started!
Track the overall responsiveness of your pages starting now! Query the metric in the data explorer and pin the graph to your dashboards for an easy overview, or check it in the frontend application and MDA screens.
Using Search Console and Google Analytics data for SEO
Featured Posts
