Thoughts on how to speed up my site? (Other site ideas.)
-
I was wondering if someone had a few minutes to talk a look at our site and see how to speed it up. I've ran some of the speed tests and I get different results, so I'm not sure if my site would be considered "slow" or not.
Also if there any other things that jump out at you (usability, side bar, things I should change / add / take away to make a users experience better, etc.) please let me know. Any feedback is good feedback to me.
I'm using WP Super Cache as well.
Thanks!
-
Paul,
Thank you very much for all the feedback. Such great info.
I've been working on improving many of these items over the weekend. I'm only using total cache, I set up a CDN, i took off on of the tabs on the activity widget and just let the popular posts. I'm trying to figure out how to resize the images better on the sidebar tab as you're right they are using css to rescale the images.
I'm actually saving up to have the site designed by a designer instad of my non-design self. But that's down the road as having something designed professionally isn't cheap (although it is worth it.)
I'm going to be working on your suggestions for the rest of the site. Thanks again for the feedback and for the kind words about our site. I'm really trying to help people through this site, and from the feedback we are getting it looks like it's been doing that.
Now I'm on the journey to make it an even better experience for people.
-
Be careful with WP Smushit. As mentioned it relies on Yahoo api that has been very unreliable, but it also uses memory mapping. If you exceed memory mapping (very easy to do on shared hosting) the plugin fails and can bring down the site with it.
-
Yea, I figured from some of the commentary and what I saw in his code he must have been acting on some of the suggestions right away. Just didn't want to assume, just in case
And yea, getting to the next level of performance may mean bigger changes like removing some items from the page altogether.
Paul
-
Looks like he's already implemented many changes since this question was posted. At the time, he was running just Supercache, now he's got W3TC, MaxCDN and he seems to have optimized some of the images.
Definitely agree with aiming for 1Mb or less per page, though.
-
First of all, just wanted to wish Noah a happy birthday!
You're right to be asking about speeding up your website, as there are a number of opportunities to improve page load times. I notice you also have questions posted asking about VPSs, and CDNs.The thing is, there is definitely an order in which these things should be tackled.
The first thing you must do is get your on page code as efficient as possible. The reality is that an inefficiently coded website can bring even the strongest of servers to its knees with just a little extra surge in traffic. It's never a good idea to try to make up for inefficient code by throwing additional hardware resources at the problem.
By default, WordPress is fairly easy to use and build, but that front-end simplicity comes at the cost of complexity and inefficiency of code on the back end. Every bit of functionality you add using plug-ins and widgets comes at the cost of additional files to load, more database calls to process, and more opportunities for someone else's inefficient code to get in the way.
In your case, your homepage is approx 2 MB in size, and some of your internal pages are over 4.5 MB. This translates into page load times anywhere from 10 to 25 seconds at typical cable Internet speeds. In practical terms, anything over 1 MB is awfully big.
Site speed optimization can be a fairly complex process, and needs an overall plan, rather than trying to implement one-off tactics and hoping for the best. I'd be happy to help you off-line to build up a plan, but in the meantime these are the kinds of things you should be taking into account:
- reduce the number of plug-ins and widgets adding content to your page to the bare minimum.
- Make sure the plug-ins/widgets you have chosen are the most efficient available in their class
- make certain photos are as efficiently compressed as possible, are being rescaled as opposed to resized in html, and be careful of using too many photos on any one page
- Ensure your caching plug-in is tweaked and tuned to be as efficient as possible for your server configuration.
Some specifics for you to think about:
- the Recommended sidebar widget (and to a lesser extent the Popular widget) are extremely inefficient in their use of images. Even though they're only showing 50x50px thumbnails, they're loading full-size images up to 550kb in the background.
- is there enough difference between Recommended and Popular for both to be of benefit to visitors?
- you're obviously using W3 Total Cache - hope you don't also have SuperCache enabled as well? If so, they'll compromise each other's performance. Only one caching plugin - ever.
- you have several different advertising networks providing ads to the page. Each one loads it's own set of backend scripts, many of which are VERY inefficient. Could you consolidate?
- you currently have over 1 megabyte of just Javascript loading on each page when target should be less than 1 MB for the whole page including content
- sidebar titles like "Like Us on FaceBook" have so little contrast from the background they're very hard to read (& likely disappear on dark monitors)
- you get so many comments on each post (a nice problem to have!) that it may be time to consider paginating the comments so that each page doesn't have hundreds of comments to load each time it displays
Overall, you're doing an absolutely kick-ass job with your website, from everything I can see. You're rocking the engagement and interest of your visitors, and you've tuned most of the things that are readily tuneable. The next level up in performance will require some tough decisions and perhaps even killing a few sacred cows. That's why I recommend you make a plan before randomly changing things.
With the way you're going, you may very well find that even the highly tuned site is going to outgrow shared hosting as a result of increasing traffic. If/when this does happen, you'll still benefit from having leaned out the website so it can take better advantage of whatever server hardware it's hosted on.
I'll close with a couple of really useful tools for visualizing exactly what's happening in the performance of your site.
First is webpagetest.org (if you haven't already found it) which allows you to run tests that directly mimic the experience a user and their browser would have. You can even have it test from different areas of the country, and using different typical Internet speeds. I usually test with the two lowest speeds, as they will cover the widest range of typical users including mobile.
The second is a plug-in called P3 Performance Profiler. It may seem counterintuitive to be adding another plug-in when what you're trying to do is reduce the complexity, but what P3 does is give you an actual readout of how much of your server's resources are being used by each of your site's plug-ins. You you can enable it for testing purposes, then turn it off again for normal running of your website.
No doubt this has created as many questions as it may have answered, so by all means fire away if you need additional clarification.
Paul
-
Pingdom says your page is 5.5mb big. But it loads pretty quick for me.
You can do alot better. Here's my recommendations.
1. Optimize your images before uploading. I would recommend the WP-Smushit plugin which does it on the fly, but lately the yahoo api hasn't been working.
2. Switch the W3 Total Cache. It does the same thing as Super cache, but there's some additional goodies.
3. Using W3TC, you should enable minify and combine for all the css and js. The plugin allows you to combine all your css and js in two files (from both your theme and other plugins).
4. Consider using a CDN. Again W3TC makes it easy. I use amazon cloudfront, and it's amazingly cheap. Essentially, you keep your current hosting, but actually serve all css, js, and images from the CDN. It makes an enormous difference.
-
Thanks. The host I'm using is really good, so I'm not sure switching hosts would be that different.
In regard to number of posts on home page, how many would you suggest?
-
After taking a quick look I would just utilize Cloud Flare (http://www.cloudflare.com). It is a free tool where they not only optimize your site with an increase of speed around 60%, but they also provide a CDN.
I tested this out a few weeks ago with a blog of mine and it took it down from 8 - 10 seconds to load, down to 1.8 seconds. Very impressive.
They also provide a Wordpress plugin.
No need to change hosting companies.
Two tools to help you as well.
-
Consider
- Moving to WPengine.com for hosting.
- Less posts on the home page.
- Compressing your CSS.
- Better optimization of images.
It was pretty quick for me. The slowest load was the externals, videos from you tube as it calls out for each one to "load" then in the browser.
Browse Questions
Explore more categories
-
Moz Tools
Chat with the community about the Moz tools.
-
SEO Tactics
Discuss the SEO process with fellow marketers
-
Community
Discuss industry events, jobs, and news!
-
Digital Marketing
Chat about tactics outside of SEO
-
Research & Trends
Dive into research and trends in the search industry.
-
Support
Connect on product support and feature requests.
Related Questions
-
With Google's new Speed Update, what does that mean for AMP pages?
Hey everyone! I wanted to get the other Mozzers opinions on this. With Google announcing a new Speed Update that will affect mobile rankings, I wanted to ask: How will AMP pages play into this? Let me know what you think!
Web Design | | TaylorRHawkins
Thanks!2 -
New To SEO Management, I just want to double check that my idea will work.
I am new to SEO management. I had a 3 month SEO copy writing internship and a 5 month SEO temp job. In both I mostly wrote copy, but I've been teaching myself SEO on the side, I became Google certified. I ended up getting a telemarketing job and somehow the conversation of SEO came up and I winded up managing their SEO for 12 dollars an hour. They say that every lead generated from the website that turns into a sale will be worth 10 dollars and if and when the sales exceed my paycheck I will starting making commission so long as it stays above my hourly. SEO is very fun and this is like my dream job. They are leaving the planning 100% up to me and I want to make sure that what I am doing will work. My plan is as follows: Part 1: Page Authority via backlinks and social media We are health care brokers and my boss, the owner has a lot of contact. He is talking with large unions like, "The Teamsters," and large company retirment groups like, "Blue flame," which is apparently in some way connected to DTE or GE. Long story short, I am trying to get him to convince them to give us a back link to our main page. He also has a ton of clients that own companies. This is good because they may be persuaded to give us backlinks too. In addition, the tech guy thinks he can implement something where we can get a google +1, facebooks likes/shares, twitter likes and shares and pintrest pin it's that would be a part of an email that we send to people within the list of 12,000 clients. From what I can see, from the client base and the people we are working with we should be able to raise the page authority substantially despite the fact that the site is only a few months old and is not yet out of the sand box. I have been slowly picking off each error with SEO MOZ's website crawling. Part 2: Making a Insurance Jargon Dictionary Guide For The Tri-purpose of gathering traffic, proving our professionalism and helping people understand semi-complex insurance jargon. I could build these 2-3 keywords would be addressed per page and they would be defined in a way to help people looking for terms understand them, while simultaneously netting a strong keyword density and a strong page. I think as far as I can tell there are no issues. Part 3: The dictionary pages will pull in new traffic and the home page will receive links and distribute link juice to the sub-pages. This subpages will guide traffic back to the main page with no-follow links to direct people from the unique termed landing pages to the home page for insurance processing. As far as I can tell my logic is solid and on paper this should work. Am I missing anything (like key details, flaws in my plan)?
Web Design | | Tediscool0 -
New Mobile Site Traffic Drop
With all the talk about how much mobile is important and how it is going to return its own search results, we finally decided to make a mobile site for one of our smaller websites to test the water. We put it up about two weeks ago and did Vary HTTP header method to serve the site. Before the change, on the average week we would get 270-300 mobile visitors from organic search results and we converted 0.78% to sales. Since the change, we are now getting about 70 mobile organic visitors per week but converting 2.47% So what can I say but WOW. We are converting way way better but our organic mobile search traffic has dropped off a ton. Luckily our desktop and tablet traffic(we serve the desktop version of the site to tablets) has stayed the same and has not dipped. Do any of you guys have experience or gone through launching a mobile site before? Did you see the immediate drop in organic mobile traffic and did you recover your traffic back to previous levels? If so, do you know how long it takes to recover? I am thinking it is a big change and will take time for Google to adjust but I am not sure since the mobile version has so much less text now on the home page and on category or product list pages or whatever you guys want to call them.
Web Design | | KurtL0 -
What's the point of an EU site?
Buongiorno from 18 degrees C Wetherby UK 🙂 On this site http://www.milwaukeetool.eu/ the client wants to hold on to the EU site despite there being multiple standalone country sittes e.g. http://www.milwaukeetool.fr & http://www.milwaukeetool.co.uk Why would you ever need an EU site? I mean who ever searches for an EU site? If the client holds on to the eu site despite my position it's a waiste of time from a search perspective is the folowing the best appeasment? When a user enters the eu url or redirects to country the detected, eg I'm in Paris I enter www.milwaukeetool.eu it redirects to http://www.milwaukeetool.fr. My felling this would be the most pragmatic thing to do? Any ideas please,
Web Design | | Nightwing
Cioa,
David0 -
Site down for more than a month - lost rankings
Hello, We have run into a situtation where we had multiple pages setup for different keywords but didn't realize that we had a name server issue that has caused the pages to be down for the last month or so (2-3 weeks on the low side.) The rank finder was still working fine, but the offline page was never reported. We realized the situation recently and have since gotten the sites back online under the new nameservers. Most of these sites were ranking 1 and 2 spots in their keywords, and now are no where to be found in the Google Index. Should I do anything differently, or just put the sites back online and wait it out? I have seen in different places that it may only take 2 weeks to come back, but it's possible that Google has marked the sites as 'not quality' because of their downtime and it will be even harder to get them to rank again. Can anyone shed any light on this situation? Any information is appreciated. Thanks in advance.
Web Design | | EQ-Richie0 -
Critique a site I'm working on please
Hi guys, I have a list of things to do on this website that is longer than my arm but thought I would gather a few more thoughts on the site before moving on further with it. The site is here. It went live a couple of days ago for live testing. I have a news section coming very soon at theworkplacedepot.co.uk/news - it is in live testing now. One glaring issue for me is that some of the products simply aren't "buyable". What I mean by that is that the options are too vague and too vast - as an outsider I wouldnt have a clue what I was buying! Other issues: The banner at the bottom of the HP is a test - its design will be changing. The sitemap has some issues and will be addressed. There is no area yet to "log-in". However, critique for me pretty please! 🙂
Web Design | | MattJanaway0 -
Given the lastest Google update, should I rewrite my Flash site or try to present an alternative HTML/CSS site?
I have a site that was created using Flash. The reasoning behind this was, at the time, that I didn't care if the site ranked or not (portfolio site). Now I would like to drive traffic to the site from SE's. Given the Penguin update, should I rewrite my Flash site in HTML/CSS or present an alternative site for bots and browsers that don't support Flash? My concern is that by presenting an alternative site to bots and non Flash supporting browsers that the SE's will see potentially see this as cloaking. Thoughts and advice would be much appreciated.
Web Design | | mj7750 -
Testing your code and site
I’ve got various WordPress websites with the Share This social plugin for WordPress. I have been using Firebug and http://analyze.websiteoptimization.com/wso to do general checks on the site and the code. And used W3C validator too. Due to the way WordPress appears to work we never seem to be able to get all the firebug/ website optimization tests to pass and the W3C validator passes everything on HTML 5 apart from 7 errors with the Share This social plugin. How do you test your code/websites? Should I stop be a perfectionist and just be happy?
Web Design | | JohnW-UK0