The State of Web Vitals

LCP loading attribute

The loading attribute on the LCP element: eager, lazy, or auto.

Field data PhoneDesktopAll Scope All sites Q1 2026 edition · All devices field outcomes
Metric LCP INP CLS
1

At a glance the headline numbers for LCP loading attribute

The loading attribute on the LCP element: eager, lazy, or auto.

3
Categories
39,055 sites
74.9%
Top share
auto
0.00
Top CLS
Median for auto

17.6% of LCP images are lazy-loaded, the one attribute they should never carry.

The State of Web Vitals · Q1 2026 · 94,910 sites · all devices field datacorewebvitals.io/state-of-cwv
2

LCP loading attribute who uses what, and how stable each group is

Median CLS
0
0.05
0.10
0.15
0.20
0.25
auto0.0075% of sites
lazy0.0018% of sites
eager0.017% of sites
VariantShare of sitesMedian
auto 29,263 sites
75%
0.00
lazy 6,884 sites
18%
0.00
eager 2,908 sites
7%
0.01

Little daylight between the groups: every variant sits near 0.00 median CLS. computed

The State of Web Vitals · Q1 2026 · 94,910 sites · all devices field datacorewebvitals.io/state-of-cwv
3

Passing CLS per LCP loading attribute which group passes the CLS most often

VariantSitesPassing CLSCLS
Eager 7.4% 2,908 91% 0.01
Lazy 17.6% 6,884 88% 0.00
Auto 74.9% 29,263 87% 0.00
Good Needs Improvement Poor Sorted best-passing first · median colored by its own rating · pass = good CLS (0.1 at p75) · one value per site

No variant stands out: pass rates sit between 87% and 91%. computed

The State of Web Vitals · Q1 2026 · 94,910 sites · all devices field datacorewebvitals.io/state-of-cwv
4

Why this matters for the Core Web Vitals, and where to start fixing it

loading=lazy on the LCP image is a self-inflicted delay. A lazy image waits for layout to prove it is in the viewport before it downloads. That puts the most important image of the page behind a step it never needed. The browser would have fetched it immediately.

Usually nobody decided this. Themes and plugins lazy-load every image by default, hero included. The fix is removing one attribute, and the audit is one search through the templates.

How does this affect the Core Web Vitals?

The choice barely moves the LCP: 87% pass at best, 85% at worst. This signal does not separate passing sites from failing ones.

Go fix it Fix: LCP image was lazily loaded → Fix slow hero images → Defer offscreen images →
Related signals LCP origin → LCP image format → LCP image size → LCP element type → Chrome field data from 94,910 sites, representing millions of real page loads · How we measured