I Tested 50 Webflow Sites on PageSpeed Insights. Here's What I Found.

The Test

We collected 50 real Webflow sites from Webflow's showcase, Made in Webflow gallery, and agency portfolios. The selection included e-commerce sites, agency sites, SaaS marketing pages, portfolios, and small business sites. All were actively maintained, live, production sites.

Each site was tested 3 times on Google PageSpeed Insights using the mobile preset (which simulates a mid-range phone on a 4G connection, the same environment Google uses for ranking signals). We averaged the three scores for each site to reduce noise from server variability.

The Results

Metric	Average	Median	Best	Worst
Performance Score	61	59	84	28
LCP	3.4s	3.2s	1.6s	8.1s
TBT	240ms	210ms	40ms	890ms
CLS	0.09	0.06	0.00	0.42
Speed Index	3.8s	3.5s	1.8s	9.2s
Page Weight	2.1MB	1.8MB	0.6MB	6.4MB

Score Distribution

• 90-100 (green): 0 sites (0%)
• 80-89: 3 sites (6%)
• 70-79: 8 sites (16%)
• 60-69: 14 sites (28%)
• 50-59: 15 sites (30%)
• Below 50: 10 sites (20%)

Zero Webflow sites scored 90 or above. Only 3 out of 50 (6%) scored above 80. Half the sites scored below 60. The median score of 59 means the typical Webflow site is in the "needs improvement" range on mobile.

What Dragged Scores Down

Three issues appeared consistently across the 50 sites:

1. Webflow runtime overhead (all 50 sites). Every site loaded webflow.js (70-120KB) and platform CSS (150-400KB). Even the best-scoring sites carried this overhead. It is baked into the platform and cannot be removed.

2. Unoptimized images (38 of 50 sites). 76% of sites had hero images over 200KB, often in JPEG or PNG format instead of WebP. Several sites loaded 4000px-wide images on mobile viewports. This is a user error, not a Webflow limitation, but Webflow does not enforce image optimization the way a build system can.

3. Excessive interactions (27 of 50 sites). 54% of sites used scroll-triggered animations, page transitions, or complex hover effects. Each interaction adds JavaScript payload and execution time. Sites with 10+ interactions scored 8-15 points lower than comparable sites with minimal interactions.

The platform overhead alone (issue 1) limits even a perfectly optimized Webflow site to approximately 85 on mobile. Issues 2 and 3 push the average down to 61.

Comparison: 50 Static HTML Sites

For context, we ran the same test on 50 static HTML sites (our own client sites and sites from other static-first agencies):

Metric	Webflow (avg)	Static HTML (avg)
Performance Score	61	94
LCP	3.4s	0.9s
TBT	240ms	18ms
CLS	0.09	0.01
Page Weight	2.1MB	220KB

• 90-100: 43 static sites (86%) vs 0 Webflow sites (0%)
• 80-89: 6 static sites (12%) vs 3 Webflow sites (6%)
• Below 80: 1 static site (2%) vs 47 Webflow sites (94%)

The gap is architectural. Static HTML sites do not carry platform overhead. Every byte is intentional. The performance floor is 85+ with minimal effort. Webflow's performance ceiling is 85 with significant effort.

What This Means for Your Webflow Site

If your Webflow site scores 55-75 on mobile (where most Webflow sites land), you are in the majority but also at a disadvantage. Your competitors on faster platforms have a ranking edge you cannot close without changing platforms.

Image optimization and interaction reduction can improve your score by 5-15 points. But the platform ceiling remains. If you need 90+ to compete in your search category, Webflow cannot get you there.

The Alternative

Migrating from Webflow to static HTML moves you from the 55-75 range to the 90-98 range. Same design, same content, same URLs. The difference is architectural: no webflow.js, no platform CSS, no interaction runtime.

Webflow migrations start at $3,000. Test your Webflow site to see where you land in this data, or read our technical breakdown of why Webflow is slow.