Pre-rendering Tester

Why compare two visitors?

Pre-rendering services, edge workers and some caching setups decide what HTML to return based on the user agent. A crawler might get a fully rendered snapshot while a browser gets an empty shell that JavaScript fills in. Done well, both contain the same content and the same signals. Done badly, crawlers get a stale snapshot, a different canonical, a missing noindex, or an error page that nobody notices because people never see it.

This tool requests the same URL twice, once per profile, and compares the parts that decide indexing and ranking. It is the quickest way to catch a misconfigured pre-render cache, a firewall that challenges bots, or a server that redirects crawlers somewhere unexpected.

Dynamic rendering and what Google says about it

Google describes dynamic rendering as a workaround rather than a long-term solution and recommends server-side rendering, static rendering or hydration instead. It is not treated as cloaking as long as the content is similar. Cloaking means showing crawlers substantially different content from users with the intent to manipulate rankings, which this tool cannot judge. What it can show is whether the responses differ, where, and by how much.

Reading the comparison

Status and final URL

If one profile gets a 200 and the other a 403, 429 or 503, a bot-protection layer is probably challenging that user agent. If the final URLs differ, one profile is being redirected, often to a country or device version.

Canonical and robots

These must match. A pre-rendered copy with a different canonical points Google to another URL, and a noindex left in a snapshot removes the page from search even though people can see it.

Content similarity

The score compares overlapping three-word sequences of the main content. Small differences from dates, counters or rotating promotions are normal. A score well below 90% means whole sections differ, which is worth explaining.

Headers

Vary: User-Agent tells caches that the response depends on the user agent. Headers such as x-prerender, a cache status or a different server value usually identify which layer answered.

Choosing profiles

The default compares Googlebot Smartphone with Chrome on Android, which is the pairing that matters most for mobile-first indexing. Compare an AI crawler such as GPTBot or ClaudeBot with Chrome to see whether AI systems are blocked or served something different, or use a custom user agent to test a specific rule.

Fixing what you find

If crawlers get a stale snapshot, shorten the pre-render cache lifetime or purge it when content changes. If a firewall challenges crawlers, allow verified crawler traffic rather than whitelisting the user agent string, which anyone can copy. If the canonical or robots tag differs, generate it from the same source for every visitor. The long-term fix Google recommends is to stop serving crawlers something different at all, by rendering the page on the server for everyone.