See how web platforms and businesses are actually built
1.1 million US websites. By platform, by business category. See what your peers use, from how many referring domains they have, down to the tools they install for analytics.
Pick your platforms. Pick your arena.
Head-to-head platform benchmarks filtered to businesses like yours, not the whole internet.
1Choose your arena
2Narrow the field
3Fill your corners
Tap a corner, then tap a platform. Tapping a chosen platform swaps corners.
Shopify vs WordPress, retail edition
n = 52,138 Shopify sites · 38,276 WordPress sites
The read
Shopify takes pass core web vitals for retail by 30 points.
real corpus cells, minimum 100 sites per platform per arena · CWV requires 30+ sites with Chrome field data · referring domains from the Common Crawl web graph
Want specific insights? Request a report.
We read every request and reply by email.
Pick a platform, business category, or subcategory above to see its benchmark profile.
How to read these numbers
- Sites
- How many US websites fall into the group in this dataset. Larger samples give steadier medians.
- CWV pass
- The share of sites passing all three Core Web Vitals, using real Chrome user field data from the Chrome UX Report. Groups with fewer than 30 field-data sites show no figure at all, because a rate from that few sites would be noise.
- Median LCP
- The middle Largest Contentful Paint across the group's sites: how long the main content takes to appear for a typical site. Lower is faster.
- Referring domains
- The typical number of distinct domains linking to a site in the group, computed as a geometric mean from the Common Crawl webgraph (April/May/June 2026).
- Top schema
- The structured data type (schema.org markup) found most often on the group's sites, with the share of sites carrying it.
- Scope and reading
- Data covers US websites only. Comparisons between groups are descriptive: they show how the groups differ as measured, not that a platform or category causes those differences.
Benchmarks derived from public web-performance datasets (HTTP Archive and the Chrome UX Report); cleaning, business-type classification, and analysis are Stackra's own.
What this dataset is
A snapshot of how real US websites are built and how they actually perform, grouped so you can compare any platform, industry, or niche against its peers. No guesswork, no vendor spin, just what is live on the web right now.
What this dataset is
A snapshot of how real US websites are built and how they actually perform, grouped so you can compare any platform, industry, or niche against its peers. No guesswork, no vendor spin, just what is live on the web right now.
What is inside
More than a million US websites. For each one we track the platform or CMS it runs on, its real-world Core Web Vitals speed, the structured data and schema markup it uses, and the tools it has installed such as analytics, chat, booking, and reviews.
Where it comes from
Built from the same public datasets Google relies on: the HTTP Archive, which crawls millions of sites every month, and the Chrome User Experience Report, the real speed data from Chrome users. We layer on US business categories and subcategories with AI classification, plus location detail from Overture Maps.
What you can learn
Which platforms a niche actually runs on, how fast a category's sites really load, and where a whole industry under-invests in SEO markup or speed. For sales and SEO teams, that is a ready-made way to benchmark a prospect and open with something specific.