It’s a well-known fact that site speed is a critical ranking factor for organic search. The big question has been how exactly Google does this. This is probably one of the most-asked questions I receive, and the answers aren’t easy to find.
Over the last year and a bit, I’ve done quote a bit of digging to get the answers, and I thought it would be useful to create an FAQ-style repository for the answers.
(Note: Google is, understandably, not 100% forthcoming with how it works. I’ve tried my best to fact-check my answers with Google employees and outside sources. If you think my answers are incorrect, let me know in the comments.)
1. Does the Google search bot track page load time?
No. The Google search bot has nothing to do with speed.
2. Does Google use synthetic test or real end user monitoring to gather its data?
I’ve talked in the past about how misleading speed metrics can be. Google actually uses real end user monitoring (RUM) to check site speed. This is the right thing to do. They’re measuring from users’ actual web browsers from real bandwidths — no simulations.
3. How does Google gather the data?
Google crowdsources page measurement by measuring site speed using the Google toolbar with Pagerank checking activated on the public’s computers. The results are “radioed” back to the mothership.
4. What browsers does the Google toolbar use?
The toolbar is available on Internet Explorer and Firefox only. More specifically, the toolbar is available on IE 6+ and Firefox 2-4. It is not supported on Firefox 5 which has led to speculation that Google has another plan for capturing this data but no other details have emerged.
5. What does the Google toolbar measure?
It measures onload time. This also includes third-party display ads, third-party scripts, etc.
6. What pages does Google measure?
They measure all pages visited by users on your site.
7. Do they measure pages marked as non-crawlable?
Yes. They measure pages your users use, not what you have told Google is crawlable
8. What if my page is personalized and has very different content for authenticated users but the same URL?
The measurement makes no distinction for personalized content if the URL remains the same. The results will be averaged together.
9. Does Google use its new Google Analytics page speed feature?
No, to the best of my knowledge they do not use any of the new data collected in Google Analytics for this purpose, but they should as it would allow them to sample more modern browsers.
10. Will pre-loading content on a page hurt my ranking?
No, because the results are based on the onload time measurement.
11. Will deferring a call help my rankings?
Yes. Anything that helps the page get to the onload event faster will help.
12. Will having pages start render faster help my rankings?
Unfortunately, no. It would be great to augment the system with some way to benefit pages that start loading faster.
Do you have any new/different/conflicting answers? Questions that aren’t answered here? Let me know in the comments.
For more reading on this subject, check out these articles:
- SEOmoz: Site Speed – Are you fast? Does it matter for SEO?
- Google: Using site speed in web search ranking
- Matt Cutts: Gadgets, Google and SEO
- Tim Morrow: How does Google measure site speed?
- Is the J.C. Penney SEO scandal relevant to the web performance industry?
- Google changes the game… again
- O’Reilly interview: Web performance for mortal companies