Sunday, May 02, 2010 at 9:17 PM
Webmaster Level: Beginner to Intermediate...k, i, s, s, i, n, g! Perhaps you heard our announcement that speed is a signal in rankings, but didn’t know where to start. We’d like to help foster a lasting relationship between you and a responsive experience for your users. Last week I filmed my updated presentation from "The Need For Speed: Google Says It Matters" which includes three first steps to understanding site performance. So grab headphones and some popcorn, then verify ownership of your website and download a plugin, and we’ll all be comfy with site performance in no time.
Just curious about the Q&A? No problem! Here you go:
Is it possible to check my server response time from different areas around the world?
Yes. WebPagetest.org can test performance from the United States (both East and West Coast—go West Coast! :), United Kingdom, China, and New Zealand.What's a good response time to aim for?
First, if your competition is fast, they may provide a better user experience than your site for your same audience. In that case, you may want to make your site better, stronger, faster...Does progressive rendering help users?
Otherwise, studies by Akamai claim 2 seconds as the threshold for ecommerce site "acceptability." Just as an FYI, at Google we aim for under a half-second.
Definitely! Progressive rendering is when a browser can display content as it’s available incrementally rather than waiting for all the content to display at once. This provides users faster visual feedback and helps them feel more in control. Bing experimented with progressive rendering by sending users their visual header (like the logo and searchbox) quickly, then the results/ads once they were available. Bing found a 0.7% increase in satisfaction with progressive rendering. They commented that this improvement compared with full feature rollout.
How can you implement progressive rendering techniques on your site? Put stylesheets at the top of the page. This allows a browser to start displaying content ASAP.
Page speed plugin, videos, articles, and help forum are all found at code.google.com/speed/.
29 comments:
Does anyone know if Google WMTs Site Performance considers the site's geo preference settings, and report accordingly?
If you have a preferred geo location set (or you have a geo tld) do your Site Performance indications consider serving speeds to your primary audience.
I get the feeling that they don't consider this - I have a reasonably well-optimised .com.au site with Australian hosting, that serves images from subdomains (soon-to-be CDN'd), serves everything minifed, rates 100/100 on Google's Page Speed (when I remove gAnalytics from the page), scores 96 on Yslow, yet gWMTs scores it as slow, around the 3second mark?
uptrends.com w/ Sydney server says <1sec. webpagetest.org w/ NZ server says <2secs. but gWMT says >3 secs?
Please advise.
The slowest thing on my website costing me the last 7 points to the full 100 in Google's own PageSpeed-measurement tool is Google's AdSense-ads.
Agreed. Seriously Google ..... is it even possible to load a webpage in half a second with ANY of the Google tools included (analytics, adsense, site optimiser etc)?
Even a bare-as-bones Google homepage (google.co.nz) takes >3secs to serve from webpagetest.org's NZ server (http://www.webpagetest.org/result/100503_8AK2/).
And that's with NO google analytics. (apparently?)
Maile,
This is something that I asked you about at SES London in Feb. What are the typical causes/solutions around fixing long Time to First Byte metrics. In your example in the video TTFB are the green bars in the Webpagetest.org results for your blog's performance.
Could you have Steve Sounders reply with some links or another blogpost talking specifically about this issue.
For example, I have a site that scores over 90/100 on PageSpeed, but still takes 6+ seconds to load. Looking at webpagetest.org I see most of the time "lost" is in the TTFB green area.
Other than just reducing the # of requests what other optimizations are there?
On twitter: @spanishgringo
The Need for Speed, just like a PC game, well thanks
although i fully understand the drive to push content to users faster, and that a faster site, is (usually) one that's enjoyed more by a user, i have the following issue:
everything that is within my control loads nice and fast. i am doing all the required cashing, keepalives, compression, etc.
but all the 'external' items i DO NOT have control over, are slowing my site down, at least according to google webmaster tools (and various other testing methods...) :/
in particular the advertising networks i am using.
what (if any) is googles opinion on this?
Thanks for sharing such a nice informative information. Really very helpful.
By the way for more information check this link: Free tutorials
seo
Really interesting! I was afraid by one of the last S.Souder post "Speed Doesn’t Matter".
I have been trying to use the google webmaster tool but I don't seem to understand it, can you help out please!
The information given is very thought provoking but I still feel speed does not play much role in the webpage rankings.
I find the high cache setting not that OK, as a client can empty cache upon exit, or use a ramdisk (firefox) etc.
I would have expected webserver output compression, just a few simple settings and a lot of gains. The less data over the line, the faster.
@e
Yes, using the geo preference would be a good idea. Or, break down the data by country or such in some other way.
i have to agree with a lot of the comments here - Google's Code, Analytics and AdSense components and PalPal are things i have absolutely no control over.
i find it curious that you claim a goal of under 0.5 seconds yet webpagetest.org shows Google related resources adding over 1.6 seconds and PayPal adding another 0.8 seconds, contributing to almost 50% of the time to completely load my root page (http://johnbishopimages.com/)
also, in WMT I have seen this for months now --- "On average, pages in your site take 3.7 seconds to load (updated on Feb 19, 2010). This is slower than 56% of sites. These estimates are of low accuracy (fewer than 100 data points)." --- why has this not been updated in over 2.5 months? since 02.19 i have put a lot of time and effort into making changes to improve my sites performance but Google doesn't seem to be paying attention! does this number in any way equate to this new signal?
and finally, you have really taken to easy way out here. Apache mod_expires does contribute to the solution here, but you completely neglect to mention Javascript Minify and GZIP compression, two other fairly simple to implement, effective methods on improving page speed.
OK - i think this is even more telling.
in this video Maile rates the repeat view results - that is what the Expires directives affect.
repeat view on my site takes 1.3 seconds and 100% of that is due to Google and PayPal.
if you (Google) are going to use page speed to rank my pages, please subtract the time spent loading data from your own sites!
YES i Agree!!
<a href="http://www.newbusiness-solutions.com>New Business Solutions</a>
Hello fellow webmasters! First, I wanted to reiterate from our original blog post:
[ While site speed is a new signal, it doesn't carry as much weight as the relevance of a page. Currently, fewer than 1% of search queries are affected by the site speed signal in our implementation and the signal for site speed only applies for visitors searching in English on Google.com at this point. ]
Now let's cover some of your questions:
Does anyone know if Google WMTs Site Performance considers the site's geo preference settings, and report accordingly?
@e:
Some of our speed statistics come from real user data (opted-in toolbar users). If your site targets an Australian audience then our numbers should reflect their usage. The stats you mentioned:
"uptrends.com w/ Sydney server says <1sec. webpagetest.org w/ NZ server says <2secs"
and
"soon-to-be CDN'd"
sound like a great experience for your users. Congrats! :)
The slowest thing on my website costing me the last 7 points to the full 100 in Google's own PageSpeed-measurement tool is Google's AdSense-ads.
@Michael, @e, @Riddla, @john bishop images:
Steve Souders, Alex Russell, along with several of our co-workers and many outside developers, are looking into improving the speed of external factors like ads, etc. There are some promising things to keep an eye out for: html5 and its iframe attributes (seamless and srcdoc) and the FRAG tag.
Also, as I'm sure you're all aware, asynchronous loading would be a terrific improvement in the ads space. In fact, companies like BuySellAds.com are already using this technique to improve performance for their publishers.
What are the typical causes/solutions around fixing long Time to First Byte metrics? Other than just reducing the # of requests what other optimizations are there?
@Spanishgringo:
Steve and I chatted yesterday: can you flush the document early? It's covered in Chapter 12 of "Even Faster Web Sites."
I was afraid by one of the last S.Souder post "Speed Doesn’t Matter".
@Romuald:
Lol. I was too, but it was an April Fool's joke. Steve got both of us.
About the video:
"I would have expected webserver output compression, just a few simple settings and a lot of gains. The less data over the line, the faster."
and:
"You have really taken to easy way out here. Apache mod_expires does contribute to the solution here, but you completely neglect to mention Javascript Minify and GZIP compression, two other fairly simple to implement, effective methods on improving page speed."
@ Sebas, @john bishop images:
Hi there, thanks for your feedback. I choose caching because it was the first Page Speed recommendation listed, which made for easy screenshots for the video. I only filmed the presentation because I thought it might be helpful for new webmasters to see a simple speed improvement done on the fly. Lol, believe me, I know the vid's not all that! I tried to warn advanced webmasters not to waste their time with the label "Webmaster Level: Beginner to Intermediate."
But anyway, yes, I hear what you're both saying. There are dozens of site improvements to make. Thanks for listing other very high priority actions. :)
@sreeram, great to see you on the blog! Thanks for building Site Performance in Webmaster Tools.
owh! what a trap! I can breath again…
You explained so well how to leverage browser caching - at last I succeeded! Please go on and explain 'Parallelize downloads across hostnames' 'Serve static content from a cookieless domain'. Thanks a bunch!
@Maile (& Steve S.)
Thanks for the tip. I'll pass the flush suggestion on to my team to investigate (Our environment is J2EE so any J2EE flush examples would be appreciated.... Ch. 12 only uses PHP examples).
However, that does not seem to address the issue I see on TTFB for all of our resource files.
This is an example from webpagetest.org
http://www.webpagetest.org/results/10/05/05/1410a7e5b395b63fc4df6d4146f42076/1_waterfall.png
I would really like to know what the likely cause is of those green bars in the waterfall for all of the images. We are making sprites of what we can, but most we will not be able to touch (just like most e-commerce sites)
Any thoughts?
I have two problems with the available Google tools
1) Site performance was last updated on Feb 27, 2010. Which as today is 6th May, was a couple of days ago now so no longer relevant
2) Page speed for Firebug reports I should Leverage browser caching
on the following resources:
* http://www.google-analytics.com/ga.js
Both of which I cant resolve...
This is very good for all website users
This is really cool, Maile, thanks. I especially appreciate knowing how to test my server response time from multiple locations.
I am looking Page Speed plug in for Google Chrome browser too :-)
I am looking Page Speed plug in for Google Chrome browser too :-)
The video and the resulting comment threads are really helpful but what am I to make of a site performance diagnostic check that says my site is slower than ~80% of the rest of the sites? Is that a red alarm? What if my site is asynchrononously loading content/ads and we are moving file to external CSS. What is the norm with that percentile?
why google advises to combine javascripts which are served from their own servers? how can i combine them at all?
Interesting stuff, just wondering:
Does fast DNS have an impact on Search Rankings (i.e: Google checking that DNS is stable) - What impact would moving to a faster DNS provider have compared to moving to a faster web host?
Does moving DNS providers have an impact?
Thanks!
In one of my sites over 3/4 of the time load is due to adsense and (especially) google maps API, this results in "your website is slower than 70% of the websites".
My google maps mashup provides a complete new and really useful user experience, not considered in page rank factors. On the contrary you penalize it by comparing page load speed vs old style websites!
I'm very disappointed from Google: the rank algorithm doesn't keep pace with the evolution of the Internet. Now AJAX and rich web application are de facto disadvantaged against old style web sites.
Hi everyone,
Since over a year has passed since we published this post, we're closing the comments to help us focus on the work ahead. If you still have a question or comment you'd like to discuss, free to visit and/or post your topic in our Webmaster Central Help Forum.
Thanks and take care,
The Webmaster Central Team
Post a Comment