April 3, 2013 at 9:42 pm #32995
Ok.. So I’m in the process (or so I thought) of moving one of my sites from thesis over to genesis.
I set-up a practice site just so I could tweak genesis and the eleven40 theme before I installed it on my site. After I loaded Genesis and eleven40 up to my brand new domain with zero content and zero images, I checked the site in Google Page Speed and it scored a 73. And this on a site with absolutely no plugins or content whatsoever..compare that with my thesis site (the one I’m considering switching) that scores a 95 loaded down with content and images..
I don’t understand. what am I missing here? Shouldn’t the Genesis site with nothing on it (not even a logo) load faster than what it did? I really want to move over to genesis, but not at the expense of site performance.
I know Genesis isn’t slow, as there are many examples of that.. I’m just trying to figure out how to get there, without hurting the rankings when I switch from thesis.
Thanks.April 3, 2013 at 10:09 pm #32996
It’s impossible for anyone to provide meaningful comments on why your site is slow because you didn’t provide a link to it
Page speed scores assign weights to certain criteria according to who does the scoring. That may or may not be relevant to your site. You’d be better off focusing more on the absolute time it takes to load your site and less on any particular score. If you have a 500 Kb site that loads in 200 ms, that is blazingly fast. If a scorer gives that a low score because it fails one of their automated measures, they’re just plain wrong.
April 4, 2013 at 10:38 am #33106
Hey Bill.. Thanks for the insight. I think I understand what you’re saying, but doesn’t it support exactly what I’m saying? The fact that the site in question should be a very small load site but still loads slowly?
Sorry for not dropping the link – http://darkmarketing.net/
Like I said, site is completely barren, so I just assumed it would score faster on Googles Page Speed tester, which I thought mattered a decent amount.
Actually, it scored a 64… I just couldn’t remember exactly when I made that post last night and bumped up the score a little in case I was wrong.. but yeah.. a 64 for an empty site just seems really slow to my novice brain.
I’m not arguing, in fact, I have no clue what I’m talking about with this stuff.. Just trying to get a grip on it all.
Thanks again Bill.April 4, 2013 at 12:09 pm #33126
Given the small total page weight, your site is slow, which I’ll cover in a bit. My point on scores was that the score can be misleading – with the emphasis on “can”. It’s important to understand that a score is based on how weights are assigned to certain characteristics. For example, on our network (we do performance oriented managed WP hosting), we do not use a CDN. Many scoring algorithms check for the existence of a CDN and lower the score because a site lacks that characteristic. But that’s only because the person who wrote the scoring formula placed emphasis on that characteristic. A low score MAY not be indicative of a slow site. Slow is based on how big your site is and how long it takes to load in absolute terms, not based on a score. To illustrate this point, the Google report to which you linked marked your score down because you used an @import rule in your CSS to load a Google font. You can use a different method to load the same font – so there is virtually no change in what your site loads or how long it takes to load – and your score will go up because you drop the @import rule. That makes focusing on the scoring mechanism less efficient; you want to focus on the important thing – the time to load your site.
For your site, running 10 tests from Dulles, VA, your site took nearly 2 seconds to load @ the median, and for a site your size, in absolute terms, that is very slow.
A couple of things stand out:
1) In my tests, repeat views were slower than first views, and that should be flipped. Good caching will narrow this gap, but you’re limited in what you can do on a shared server.
2) Much of your content is not gzip’d. Do you have gzip enabled on your server? This will help, but it won’t solve the core problem.
3) Are you on a shared Hostgator server? My results are all over the place, ranging from a relatively fast page load of 597 ms (0.6 sec) to 2131 ms (2.1 secs). That fluctuation is most likely due to the variation in traffic on that shared server that might have nothing to do with you. If the server doesn’t have other things to worry about, your site is quick; if somebody else is consuming the server resources, then you are stuck waiting. Of my 10 tests, only 1 was relatively fast; 9 were relatively slow.
4) For comparison, you can load our eleven40 demo site. That’s a copy of the StudioPress demo site on our network. Our first views averaged 1224 ms and our repeat views came in at 245 ms – something that you can’t likely touch on your HG shared server, and the reason goes a little deeper than just server load. Tests on your site show that of the 2 seconds to load your site, about 75% of that is wait time, and that usually is waiting for a response from the MySQL server (since WP stores info in a MySQL database). To know for sure, you’d have to have server monitoring software installed that you can’t install on a HG shared server. DB connection times at shared hosts are slower than they need to be; it’s just a fact of life.
5) Note that on the Google report to which you linked, the only high priority item listed is server response – which confirms everything I’ve stated above.
With all that said, while I think your site is slow, I don’t know that there is much you can do about, other than changing hosts. Depending on your plan, we might be slightly cheaper than HG or slightly more. Your site is not likely to get a lot slower as you add basic content, though. If your fully built out site is faster, it’s not because of Thesis vs Genesis; it’s likely due to different server loads and the variations in testing. At the end of the day, if performance is important to you (and it does impact site traffic so it should be important to you), you have to change your server model, since it sets a lot of boundaries on the performance you can achieve.
April 4, 2013 at 9:27 pm #33239
OK Bill, I think I “get it”.
The score is just that, a score, NOT how fast the site actually is. But the score is relative to your own site and can provide insight into what actions could be taken to speed the site up.
My “fast” thesis site loads on average of 4 secs a page according to G Analytics, but scores a 94-96 on G’s page speed score. Whereas the Genesis site has page speed “score” of 64 but is twice as fast with a page load time of 2 seconds.
I get it… but I gotta say that’s pretty darn confusing, not to mention I put a good but of time to get my thesis site to a 90+ score only to find out my overall load time is merely average at best. Hmm.
Thanks again Bill. You gained a twitter follower.April 4, 2013 at 10:02 pm #33245
4 seconds to load a page is way to slow, no matter what score you get. You need first page load times under 2 seconds for just about any site, and under 1 second for sites that are not image or script heavy. And note that I am saying first page load times, because on our network, we have aggressive caching, so the 2nd page load within a time interval completely bypasses PHP and the DB, and loads typically in under 1/2 second (500 ms). As I said, that’s difficult if not impossible to accomplish on a shared server. I’ve seen quotes that every extra 1/2 second of page load can cost you 20% of your organic search traffic, so if your traffic translates to $, at some point you’ll want to get that 4 seconds under control.
April 4, 2013 at 10:09 pm #33246
Will do..and when I do get to that point.. I’ll keep you in mind. Although, I’ve gotta be honest, there’s some pretty stiff competition.April 5, 2013 at 7:00 am #33278
You must be logged in to reply to this topic.