Actual page load time vs. perceived one

Question

You know how all around the web, people are pushing for smaller and smaller page load times ? They do so with good reason, especially with Google now factoring it in in their way to rank pages. It's good business practice, good business ethics, and everybody is happier when pages load faster.

Now, something's been bothering me as of late. I had this page that loaded reasonably fast (under a second) but in it was a facebook like button. This little thingy cranked up the page load time to 2 seconds. But just the actual one, not the perceived one. For users, everything loaded fast enough on the page except for that iFrame.

So, for Google, even though it loaded fine for users (in term of usability etc), did it penalize us ? I've decided to remove it, and follow the results in analytics. Will this help with SEO ?

I feel like the concept of page load time itself shouldn't be restricted to the combined loading time of all elements on the page as calculated by a tool. I hope I am posting this in the right area, I wasn't too sure.

Thanks in advance for your answers !

Answer 1

The official word from Google on page load speed is that it makes little difference to your rankings:

"...we still put much more weight on factors like relevance, topicality, reputation, value-add, etc. — all the factors that you probably think about all the time. Compared to those signals, site speed will carry much less weight."

and:

"...fewer than 1% of search queries will change as a result of incorporating site speed into our ranking..."

As such, I suggest you optimise for people first and Google second. Removing Facebook buttons diminishes the user experience for people wanting to share your content and may actually harm your rankings, because people are less likely to share, blog, and link back to you. It seems counterintuitive to remove them in the hope of boosting your rankings.

Use Google's speed tools
It's still worth running your site through Google's Page Speed Online tool to see where improvements could be made, if only for the benefit of your users. Chances are there are other things you can do beyond stripping the Facebook button that might improve page load speed.

Out of interest, I ran a page that uses Facebook's 'Like' button through the Page Speed tool (like this), and got warnings mentioning that the API requests to Facebook's server have a short caching time of 2 minutes; Google recommends extending it to one week to take advantage of browser caching. Since you can't control the caching time, though, it's not worth worrying about.

In short, I'd put the buttons back or use a static link instead, as documented in this answer.

Answer 2

So the Facebook button is loaded in an iFrame? if so, then Google won't "see" that or attribute it to your site.

If you're loading in about 1-2 secs in terms of total loading time, then you're still operating within the top percentile of all websites. It's only when sites take much longer - 5s+ and beyond when it may start to have an impact and even then only a very minor part of the total ranking algorithm.

Short answer - I wouldn't worry about that level of performance.