The State of Web Vitals

Requests via CDN

How many of the page requests are served via a CDN, and which one.

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

At a glance the headline numbers for Requests via CDN

How many of the page requests are served via a CDN, and which one.

7
Categories
In the distribution
66.5%
Fleet share
Top: origin
99.4%
Sites with any
Of origin

66.5% of requests come straight from origin servers, with no CDN in front.

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

The requests via CDN mix who uses what, and how fast each group loads

Median INP (sites using feature)
0
100ms
200ms
300ms
400ms
500ms
Origin49ms67% of sites
Cloudflare53ms22% of sites
Fastly55ms6% of sites
Cloudfront58ms5% of sites
Vercel66ms1% of sites
Akamai57ms0% of sites
Netlify63ms0% of sites
VariantShare of requestsMedian
Origin
67%
49ms
Cloudflare
22%
53ms
Fastly
6%
55ms
Cloudfront
5%
58ms
Vercel
1%
66ms
Akamai
0%
57ms
Netlify
0%
63ms

Requests via CDN. On the fleet: 66.5% origin, 22.3% cloudflare, 5.6% fastly. 99.4% of sites use at least one origin.

Origin leads by count (66.5%) and by bytes (75.2%). computed

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

Passing INP per bucket every category and count level at once - color is the pass rate

1
2
3
4
5
6
7
8
9
10
11
Origin 66.5%
98
98
98
99
99
98
99
99
99
99
98
Cloudflare 22.3%
99
99
99
99
98
98
99
99
98
99
99
Fastly 5.6%
98
99
99
98
99
98
99
99
98
99
99
Cloudfront 4.8%
99
98
99
99
99
98
97
97
97
97
93
Vercel 0.5%
100
97
100
100
93
97
92
100
96
92
Akamai 0.2%
99
100
100
100
99
96
98
93
89
99
Netlify 0.1%
100
91
60
80
← few of this category on the pagemany →
60%95%+ of sites passing INP Faded cells: under 100 sites

Each row is a category, each column its own count bucket (few on the left, many on the right); the cell is the share of those sites passing INP.

Cloudfront swings the hardest: 99% of sites pass INP with few, 93% with many. computed

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

Few vs many - does quantity cost INP? the pass rate with few vs many of each category

60%70%80%90%100% few → many
Akamai 0.2% 99% → 93%
Cloudfront 4.8% 99% → 93%
Vercel 0.5% 100% → 96%
Origin 66.5% 98% → 98%
Cloudflare 22.3% 99% → 99%
Fastly 5.6% 98% → 99%
% of sites passing INP · hollow ring = pages with few, solid dot = pages with many

Per category: the pass rate among pages with FEW of it (hollow ring) against pages with MANY (solid dot), worst trend first. Thin buckets are excluded from the endpoints.

More Akamai costs the most: the INP pass rate falls from 99% with few to 93% with many. computed

The State of Web Vitals · Q1 2026 · 94,910 sites · desktop field datacorewebvitals.io/state-of-cwv
5

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

A CDN moves your files closer to the visitor. Distance is round-trip time, and every request pays it: the connection setup, the request, the response. Serving static files from an edge node cuts that time for every image, script and stylesheet on the page, and those are the resources the LCP waits for.

The share matters more than the logo. A site that serves its HTML through a CDN but loads images from a slow origin still pays origin latency for its LCP image. Check what the CDN actually fronts, not just whether one is present.

How does this affect the Core Web Vitals?

Requests via CDN correlates with the LCP. With Vercel, 88% of sites pass the LCP. With Netlify, 81% do.

Go fix it Configure Cloudflare for passing the Core Web Vitals → Time to First Byte →
Related signals Last-Modified present → Uncompressed text responses → Cacheable responses → HTML size (kB) → Chrome field data from 94,910 sites, representing millions of real page loads · How we measured