Google Analytics Site Speed is a bit Unreliable

Google Analytics will now allow site owners to track the performance of their websites with real live traffic. This is a nice feature that lets you understand just how long it takes for the average visitor to your site to see the fully complete version of your website. While this sounds like a great tool that will give you an accurate view of yoru website’s performance, it does not tell the full story. You can access the Site Speed section of Google Analyics under the Behavior>Site Speed>Overview section. Please be aware that not all visitors will send this information, so you will have to have a certain amount of traffic to your site before you will get any information in this section at all.

There are many factors that come into play when you measure the performance of a website. These include but are not limited to:

  1. Network Connection
  2. Web Browser
  3. Computer/Device Performance
  4. Screen size/resolution
  5. Internet Congestion
  6. Web Server Traffic Load

While this is not an exhaustive listing of the factors that come into play, they are a good starting place to outline how they can affect these numbers.

Network Connection

A User’s network connection is the single most important factor to consider when measuring or reading measurements of the performance of your website. If a user is on non-3G wireless or the equivalent of dial-up internet access, it doesn’t much matter what you do to optimize your site, if there are any images or any other large assets on the site, it will load slowly. If Google Analytics allowed you to filter the site speed results by connection speed, it would allow you to get a more accurate picture of how the download size affects the visitors to the site, also indicating if your visitors can handle a more image-intensive site, or if a lighter-weight site would be preferred.

Web Browser

Assuming that we are only discussing the latest and greatest browsers from each of the major vendors, this has very little impact on the performance of your site. However, while Firefox, Chrome, Safari, and Opera have all had adequate performance in their older versions of the browsers, older versions of Internet Explorer, however, had significant performance limitations that would cause websites to load more slowly than the other browsers. Fortunately, Google Analytics shows the differences in performance between Web Browsers.

Computer/Device Performance

This is another factor that is inconsequential to users of your site that have relatively current computers or devices they use to access your site. However, if a user is using an old and outdated computer or mobile device, they will likely have a worse experience with the performance of your site than what the average user would.

Screen Size/Resolution

While this may not seem like a big deal, there are many ways that this can impact the performance of a website. The first is that users that have lower screen sizes or resolutions tend to either be using older mobile devices or older desktop/laptop computers, resulting it lower processing capability to display the site. In addition, if you have a device with a higher resolution, then some newer sites will download higher resolution images to ensure the device gets images that are optimized for the higher resolution. However, this also will increase the total downloaded size for the site, so this will slow the site down as well. While Google Analytics overall allows you to view the screen resolutions of visitors to your site, they do not provide the same information for the site speed portion.

Internet Congestion

This has become more of a factor with the increase in the amount of streaming media that is available out there with YouTube and Netflix and others. During peak viewing times, it is not unusual for certain sections of the internet to become congested, causing internet traffic to take longer to get to the user than is typically expected. As this is something that happens outside the realm the web browser or your web server, there is not really a way to measure this impact.

Web Server Traffic Load

While you can hopefully optimize your site so that is able to be served as quickly as possible, there is always a point at which your web servers can take no more traffic without slowing all of the traffic down overall. Once you reach this point, the only way to address the slow down is to add additional servers or a more beefy caching solution. Again, this is something that Google Analytics is unable to show you. However, the reporting and analytics tools on your web server should allow you to detect this scenario. If you are not using something to do this with, a tool like New Relic will allow you to detect this scenario.

comments powered by Disqus

Related Posts

Unintended Consequences with Magento Observers on Sales Orders

Anyone that uses Magento to place orders will be hard-pressed to consider this process a speedy one. While it takes a while to process the order under the best of circumstances, there are a few things that you can do that actually make it worse. One of those things that can make it worse is creating an observer that runs in the middle of the saving of the order processes that is always slow-running, or continues to get slower over time as the data that the Magento site grows.

Read More

Responsive Images with Picturefill 2.0

Responsive Web Design seems to be the way that the majority of websites will be developed in the near future. For a while, everyone was creating a separate website that catered to mobile devices in addition to the main website that desktop browsers were able to access. Web Developers and UX Designers quickly discovered that this was a less than ideal approach as it required maintaining two separate websites, and the mobile website tended to remove data that was visible on the desktop version of the site.

Read More

Avoid jQuery.bind()

When chasing down performance issues, you never know what kind of problems you will find. I was looking for something that would cause jitter when scrolling on the page. After looking at the custom code that runs on every scroll event, I still had not found a reason for the jitter. Looking at the JavaScript CPU profile when scrolling in Chrome showed that there was an overwhelming majority of the time spent in a function in the Prototype JS library.

Read More