This topic contains 4 replies, has 3 voices, and was last updated by Profile photo of Hin L Hin L 2 years, 11 months ago.

Exp #13: Site Speeds and Rankings

  • Profile photo of Hin L

    The Need for Speed – true or false?

    site speed

    One of the questions I keep getting asked is where you should be hosting your money site.

    Typically I put everything on BlueHost/Hostgator but having faster servers obviously has a tremendous benefit, including:

    1. Improved User Experience
    2. Improved Rankings

    I don’t think anyone will refute the first point, but how important #2 is is up for debate.

    In this experiment, we’ll be testing:

    • The fastest host
    • The significance of site speed and rankings

    1. Site Speed and Rankings

    Does site speed have a direct correlation to SERP rankings?

    We will take 6 sites that are currently ranking and transfer their hosting to either a faster/slower host (all sites are currently hosted with companies similar to bluehost, hostgator, godaddy etc.).

    The theory is that if there is a correlation, the sites moved to the slower host will drop in rankings, and vice versa.

    • 2 sites moved to a slow, $1 host
    • 2 sites moved to A2 Hosting (known for their speed)
    • 2 sites moved to Traffic Planet Hosting (also known for their speed)

    2. Which is the Faster Host

    We polled a couple people in private FB groups and the 2 hosts that kept getting mentions for their speed were A2 Hosting and Traffic Planet Hosting. We’ll be testing to see which one is faster.

    We will do this by hosting the same sites on both A2 Hosting and Traffic Planet Hosting to ensure a fair test.

    The 2 metrics we’ll be measuring are:

    • Document load time (total time to load the full page)
    • Time to First Byte (the amount of time to receive 1 byte of data)

    We’ll be using http://www.webpagetest.org/ for our testing.

    Profile photo of Alston

    looking foward for this :)

    Profile photo of Hin L

    Alright guys, apologies for the delay here. We finally had a chance to put everything together so here we go (this will be long! If you want to skip the details, jump to the bottom).

    Before we start, let’s see whether the site speeds increased or decreased.

    Remember: We put 4 sites on a faster host (A2 Hosting and TPH), and 1 site on a slower host (was supposed to put 2, but couldn’t because of issues with the site).

    Speed Changes

    So like I mentioned before, we’ll be looking at load time and first byte time (TTFB). According to a study by Moz, a faster TTFB correlated to higher rankings.

    Overall Speed Changes2

    There was little variation in the Load Time, so we won’t focus on that.

    In terms of TTFB, we saw speed increases across all sites except Site 4 (A2 Hosting). Interestingly enough, it had the higher TTFB increase AND decrease. I’ll talk more about that below.

    In the meantime, just know that all sites increased in TTFB speed (which is what is important for rankings according to Moz) except for site 4.

    Rankings

    Now that we see site speed changes, let’s see if this correlated to noticeable ranking increases.

    Site 1 – Pre-host change

    Site 1 - pre-change - modified

    Site 1 – POST-host change

    Site 1 - Post Host Change - modified

    Site 3 – Pre-host change

    (NOTE: we had to remove site 2, because of some internal issues)

    Site 3 - pre-change - modified

    Site 3 – POST-host change

    Site 3 - Post Host Change - modified

    Site 4 – Pre-host change

    Site 4 - pre-change - modified

     

    Site 4 – POST-host change

    Site 4 - Post Host Change - modified

    Site 5 – Pre-host change

    Site 5 - pre-change - modified

    Site 5 – POST-host change

    Site 5 - Post Host Change - modified

    Site 6 – Pre-host change

    Site 6 - pre-change - modified

     

    Site 6 – POST-host change

    Site 6 - Post Host Change - modified

     

    As you can see, the ranking increases are very negligible. There is very little noticeable change that you wouldn’t see from any 2 week period in the SERPs.

    Why did site 4 decrease in speed

    Personally, I’m not sure, but A2 Hosting has a very good plugin that optimizes the sites for you (compresses images, minifies HTML, javascript etc.). Here’s what it looks like:

    A2 Optimizer plugin

     

    As you can see, very easy to use for the non-techie.

    We enabled everything and put it through the test again, and these are the results we got:

    Speed Change after A2 Plugin

    As you can see, BOTH load pages and TTFB increased significantly. Not to mention, very little work was involved.

    Conclusion

    Overall, the speed increases from A2 Hosting were just so much more significant than traffic planet, that they’re the ones I’d have to go with. They have a built in plugin to optimize your site, which makes it incredibly simple for the non-techie.

    In terms of ranking increases, we noticed very minor fluctuations (normal with what you’d see in the SERPs). If you’re increasing your site speed with the goal of ranking higher, then you’re probably wasting your time. There is however a significant benefit for your end users, just not for Google’s sake.

     

    Profile photo of rogerke

    Thanks for sharing this. It completey complements my own experiences.

    Matt Cutts also said you only see decreases in rankings if your site is really slow. So there seems to be a rather high threshold you have to cross to get your site penalized. As long as you’re in the “safe zone” improving your website’s speed is no worthwhile effort.

    Profile photo of Hin L

    @rogerke I dont’ always trust Matt Cutts, but he does seem to be on point with this one

You must be logged in to reply to this topic.