SiteLoadSpeed

Time Your Website Before Panic Or Complacency Sets In

Add Your Comments

With the rumblings from Google implying and then suggesting that the time it takes your website to load is going to be a factor in ranking, webmasters everywhere had kind of a dual reaction — panic because of concern about how they were going to speed up loading of huge sites with tons of background stuff going on versus high fives due to the realization that here was a ranking factor that they could do something about and over which they had control. Once the brouhaha died down, however, it was time to look at the situation realistically.

On April 9, 2010, Google officially announced that site load times were part of the algorithm in a post on the Official Google Webmaster Central blog. The most interesting part of that post, however, was not the official announcement of something that had been an open secret for a long time, but Google’s estimation that:

“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. We launched this change a few weeks back after rigorous testing. If you haven’t seen much change to your site rankings, then this site speed change possibly did not impact your site.”

Does this mean that you should stop panic mode, stop investigate mode, and just say to yourself, “1%? I’ve got better things to spend my time on!” and go back to harvesting your crops on Farmville? If investigating your site load time were a complex matter that ate up a lot of resources, required costly tools, took a ton of time, or was too complex a process for the average person to complete, you might be justified in bringing in the strawberry harvest instead, but it’s none of those things.

There are a number of free tools that will tell you what you need to know in a short amount of time with minimal effort. Google even tells you what some of them are. So it would be a mistake to look at the 1% stat and say it’s not worth investigating. In actual fact, this is a stat you should have been checking into anyway, long before Google started hinting about it becoming an official part of the algorithm. Why? Because load time is going to impact user satisfaction with your website. The longer it takes to load, the greater your risk that searchers will open a new window in their browser to search again or hit the back button to the SERP to try the next site on the list. And searchers are getting so used to having their choices appear instantaneously that even a delay of a couple of extra seconds will be enough to persuade some to abandon your site before they even see it.

This behavior is due to a bunch of different factors, including the increasing speed of computer equipment that the average person has access to, the expansion of broadband access and increasing options for ordinary people to access super high speed access, and a general emphasis placed on speed in all our online activities. Although I remember the days of modem dial-up and clicking on a URL, heading to the kitchen to brew a pot of coffee, only to come back and have the page still loading, I also expect sites to load quickly and I find myself hitting back buttons or closing windows if they don’t. I was quite happy with the time it took to access a site in the “good old days,” but we get used to quick access very quickly and we don’t want to settle for less once we are used to that level of service.

With a ton of tools available to check the load time, most will do the trick. One suggestion that I think is important, however, is to ensure that one of the tools you use is a Google one. Although the differences in the results you receive from any one of the available tools is likely to be small, it makes sense to make one of your benchmarks be a Google-endorsed tool, as it is likely to be closer in construction to the standards that Google is working into the algorithm. The way to this tool is through Google Webmaster Tools. Access your Google account and the site in question (it must be a Google-verified site, which is a simple process involving verifying that you are the owner of the site in question, achieved as quickly and easily as pasting the supplied code into the head section of the index page of the site and uploading the revised home page to your server), and then choose Labs, Site Performance. The tool tells you how quickly the site loads in seconds, how it compares to other sites, its load time over time, and suggestions for speeding up the load process based on analysis of the files in your website.

Here’s a sample of the load speed graph the Google site will show, graphing the last few months. Frankly, I’m not sure what caused the improvement in load time in the site in question, as I don’t recall anything specific being done to the files that comprise that site. However, it is a small site on a shared server, and an improvement at that level is probably responsible. If you have a similar situation, you’ll need to weigh all the factors to decide if upgrading to a dedicated server is part of the solution.

This area also has a link to the Page Speed browser add-on that, once installed, will provide a speed for each page, as well as detailed analysis and suggestions for improvements in much greater detail than the general Google analysis. Areas dealt with include optimizing caching, minimizing items such as round-trip load times, payload size and request overhead, and optimization of browser rendering. Each of the factors in the areas covered is given a score, so you can drill down and see specifics of what individual items might be able to be tweaked to improve your overall score and ultimately, the page load (and site load) time.

Of course, there are a variety of independent tools that do the same thing and will test the same factors — a topic for a future blog post. You certainly should check all of the factors uncovered via the Google tools in another, independent, tool. Then gather all the data, correct what you can, and look into any areas where the different tools have returned significantly different data (if any exist). You may need to dig further to discover which is more accurate, but if your main goal is your ranking on Google, you may have to settle for fulfilling Google’s expectations first and dealing with any inconsistencies or others later.

Whatever you do, don’t panic if your site load time falls into the “slow” category. Many of the fixes will be easily accomplished and any that you are able to take care of with little fuss will immediately confer some benefit to those accessing your site, especially if they are still using older equipment that magnifies differences in site load time by virtue of its relative slowness.

About the Author

Frances Krug has worked in market research since graduating from UCLA with an MA and CPhil in Latin American history. As an editor and online content provider for the last 7 years, she currently is Associate Editor at iNET Interactive, where she also directs Search Marketing Standard's email marketing program.

Add Your Comments

  • (will not be published)