CODEWITHSUNDEEP
core-web-vitals
Svalby
Svalby

Reputation: 1

Core web vitals difference in available metrics between Origin and this URL

In the process of rebuilding my site there now appears to be a difference between the scoring available for the origin versus the index.html URL.

Passing the origin through PageSpeed Insights returns four of the six core web vitals with FID and INP listed as not applicable.

A screenshot of PageSpeed Insights scoring on site origin

Passing the index file URL through PageSpeed Insights however returns five of the six core web vitals with FID listed as not applicable.

A screenshot of PageSpeed Insights scoring on site index URL

The subdomain shifted from a core web vitals fail to a pass across all six metrics roughly eight weeks ago, and remained a pass until one week ago. Now they have shifted back to a fail across three of the metrics. I am attempting to make the case for funding the engineering team responsilble for the hosting environment to look at the infrastructure to see whether there is any room for improvement on this front.

I have not experienced any of this previously - having trawled this website for corresponding questions, i would appreciate any insight from users who have experienced this, and would be willing to shed some light on a prospective solution.

Upvotes: 0

Views: 498

Answers (1)

Barry Pollard
Barry Pollard

Reputation: 46040

In theory it is not possible to have an INP value, without also having a FID value - the first interaction that FID takes into account should be counted for the INP value. So I can’t quite explain that.

However I do not think that is your main issue as an N/A value for FID does not prevent you passing Core Web Vitals as some pages do not have interactions (think static article pages or blogs).

You main issue for passing seems to be due to LCP, which can also be seen reflected in FCP and TTFB. This also appears to have regressed considerably recently, as you can see when accessing the CrUX history data via this tool:

Bar chart of historical TTFB data showing a drop off recently

TTFB is the starting point as, until you get those first bytes back, you cannot hope to start painting your first content (FCP), or your largest content (LCP).

When TTFB drops considerably like this it’s usually for one of two reasons:

  1. Your hosting provider has got worse (either due to issues, or because you have changed hosting provider)
  2. Your traffic patterns have changed.

The second can happen if for example:

  • you got a large influx of traffic from slower users (e.g. from another country and you are not using a CDN)
  • users are reaching you via many redirects (e.g. if running an advertisement campaign as those links often go through multiple redirects. Or if you’re getting traffic via a URL shorter redirects)
  • users are looking at lots of older pages that are not cached and so take longer to serve. Or URL parameters (e.g. utm params) are preventing cached content from being used.

Have a look at your site analytics to see if you’re seeing a shift in traffic sources or pages visited.

A CDN can often help with this if not using one. And ensuring users can get to your site with as few redirects as possible - i.e. try not to use multiple URL shortened for quick links, or sending people to http:// links that redirect to https://.

Upvotes: 2

Related Questions