A reverse proxy is a server that sits in front of your website. Visitors connect to the proxy, the proxy fetches the page from your real server, and it returns the response, with the option to change it along the way. “Reverse” distinguishes it from a forward proxy: a forward proxy acts on behalf of the client, hiding the visitor, while a reverse proxy acts on behalf of the server, standing in front of your site.
Why it matters for speed
Because every response passes through it, a reverse proxy is the one place you can optimise a page without changing the site itself. It can rewrite the HTML, inline critical CSS, defer scripts, transcode images and cache results, all in delivery, before the bytes reach the browser. Your CMS, theme and content stay exactly as they are.
That is a different model from a plugin or a rebuild:
- Nothing to install. No plugin to add, no snippet to maintain, no build step to change. The proxy works at the HTML level, so it is engine-agnostic.
- Nothing to break. Your source is untouched. Disable the proxy and you are back to the original page instantly.
- Works where you have no access. On locked-down platforms such as Tilda and hosted builders, fronting the site is often the only way to change how it loads.
The failure mode nobody warns you about
A reverse proxy is also the easiest way to measure nothing at all, and the mistake is invisible because everything reports success.
If your origin redirects, say from the apex domain to www, a faithful proxy passes that
redirect straight through. The measuring tool follows it, lands on your original site, and
measures the original twice. The result is a page that returns 200, an audit that completes,
and a small plausible improvement that is entirely the run-to-run noise of the measurement.
We found this in our own base, and the leak was not evenly spread: it affected effectively all
of the Wix audits we had run, most of the Squarespace and Webflow ones, about half of Duda,
and roughly a quarter of WordPress, because apex-to-www is a builder default while on
WordPress somebody configures it deliberately. It had already produced a confident conclusion
about which platforms we help least, drawn from a comparison where some platforms contributed
no valid observation whatsoever.
The trade-offs
A reverse proxy adds one hop between the visitor and your server, so it has to earn its keep: the optimisations and caching it applies need to save more time than the extra hop costs. It also has to handle dynamic behaviour faithfully, including cookies, forms and third-party widgets, which is why a speed proxy caches carefully and passes through what it should not touch.
How WebSpeed uses it
WebSpeed is a smart reverse proxy. It serves an optimised copy of your existing site by running the response through a filter pipeline, critical CSS, deferred scripts, lazy media, WebP, resource hints, and caching the result. Your site stays untouched, and you can measure the before and after on your own URL.