Google’s PageSpeed Insights added two new metrics to the API and UI for Lighthouse:
- Interaction To Next Paint (INP)
- Time To First Byte (TTFB)
Currently, the two new metrics are categorized as experimental and are collected for field data. The corresponding lab data can be used in diagnostics.
Interaction To Next Paint (INP)
Unlike First Input Delay, which measures how long it takes for the page to respond to the first interaction by a visitor, this new metric points out how long it takes to interact with the entire page.
Google’s Web.dev explains the new Interaction to Next Paint (INP) metric:
“INP is a metric that aims to represent a page’s overall interaction latency by selecting one of the single longest interactions that occur when a user visits a page.
For pages with less than 50 interactions in total, INP is the interaction with the worst latency.
For pages with many interactions, INP is most often the 98th percentile of interaction latency.”
Google gave us a shopping page as an example. Imagine you’re looking at shirts but can’t decide on the colour. If the system is giving you feedback in a form of a graphic indicating it’s working on something as soon as you clicked a different colour, that would mean that the responsiveness is good.
Time To First Byte (TTFB)
This metric shows how fast the server responds to a request for a resource. It did exist before but is now available in the PageSpeed Insights result section for the first time. Also, Google will collect and report TTFB field data.
TTFB consists of the following request phases:
- Redirect time;
- Service worker startup time (if applicable);
- DNS lookup;
- Connection and TLS negotiation;
- Request, up until the point at which the first byte of the response has arrived.
If you’re wondering why it’s important, it’s because it helps isolate a particular factor that can negatively impact all other page loading metrics that follow.
Find out how to measure TTFB here.
Page Speed Insights Updates
Other than some design and layout changes, there are additional changes to the tool itself, one of which is the speed at which results are shown. Since field and lab data load in parallel, the results are visible sooner. Also, PageSpeed Insights will try to resolve HTTP redirects before sending them off to Lighthouse for analysis. In the past, Lighthouse would handle HTTP redirects when performing analysis.
By adding these new metrics, all page speed tools that use Lighthouse analysis will become much more valuable. How? The TTFB provides insight into server-related issues that could lead to poor performance elsewhere, while the INP offers a broader overview of page performance. Both metrics are available in third-party tools.