Lower perceived credibility (Fogg et al. 2001)
Lower perceived quality (Bouch, Kuchinsky, and Bhatti 2000)
Increased user frustration (Ceaparu et al. 2004)
Increased blood pressure (Scheirer et al. 2002)
Reduced flow rates (Novak, Hoffman, and Yung 2000)
Reduced conversion rates (Akamai 2007)
Increased exit rates (Nielsen 2000)
Are perceived as less interesting (Ramsay, Barbesi, and Preece 1998)
Are perceived as less attractive (Skadberg and Kimmel 2004)
We know that search engines and site visitors reward fast loading sites so it’s in our best interest to help our members achieve fast page load times. So how do we do it?
A LMS CASE STUDY
Let us give you an example. We recently had a new member sign up who has been selling their course online for the last 3 months, but ran into too many problems doing it on their own so they signed up with us.
However, before we transferred them over to our servers we wanted to do some tests. Essentially, we wanted to measure their “before and after” site speed details.
Just to give you an example of what’s average for more robust sites (like many eLearning sites), at the time of this writing Ebay had a site speed of 5:07 according to Pingdom.
Ebay is a great site with lots of dynamic features so it’s never going to win a speed race. There are many faster sites online. However, it’s Ebay’s dynamic features (the parts of the site that slow it down) that make eBay worth visiting in the first place.
It’s always a delicate balance between great website features and site speed. eCourse owners often find themselves in the same predicament. For example, they might have amazing features within their online course software, but those features might slow their site down to a crawl if too many users are using them at the same time. Which essentially was the issue in the case study we’re about to look at.
THE BEFORE PICTURE
Before this member joined us they had a site speed of 5:15. They were paying around $100 / month for a VPS server. However, their host didn’t publish much relevant information about the server’s specs. Which was a bit surprising since it’s more common for shared hosting providers to hide that information since most people who sign up for shared hosting aren’t deeply concerned with processing power if they are only willing to pay $20 / month. These people often just want something cheap that does the trick.
But like most things in the world, usually “cheapest” and “greatest” aren’t found in the same sentence together. That’s especially true when it comes to hosting. Since cheap shared hosting providers can’t sell their packages on server specs, they instead focus on gimmicky (and mostly untrue) statements like “unlimited bandwidth” or “unlimited space” or “99.9% up-time”. For example, Hostgator, one of the most popular shared hosting solutions online, touts a 99.9% up-time guarantee, but just this week they had over 15 hours of downtime in one day and tens of thousands of customer complaints. You don’t have to be a math guru to figure out that it doesn’t add up.
SO WHAT IMPACT DOES A SERVER HAVE ON SITE SPEED?
Okay, so let’s get to our case study. In this example we took our new member off their old server and we put them on one of our dedicated servers. Not surprisingly there was an immediate increase in site speed. They went from a page speed of 5:15 to 2:70 (a 47% increase in speed!)
THE JUICY STUFF! A CONTENT DELIVERY NETWORK (CDN)
Then we took it a step further by setting them up on our Content Delivery Network (CDN) which essentially takes the static elements of their website, uploads them to different “clouds” around the world, and then delivers those website elements to visitors based off that particular visitor’s location. So a visitor visits their site from NYC, that visitor would be retrieving their site details from a server in NYC. However, if a visitor came to their site from London, then that visitor would be served their site content from a server in London. Adding their site to our CDN account took their site speed from 2:70 to 2:39 (a further 11% decrease in page load time).
In total they had a 53.59% increase in site speed by becoming an Academy of Mine member.
OUR SERVER SPECS WHEN YOU FIRST SIGN UP
When you first sign up with Academy of Mine you’ll be put on our basic dedicated server with a limited number of other clients. It’s like a hybrid system between shared and dedicated. Our server specs are:
Intel Xeon E3-1230 V2
3.3GHz, Turbo 3.7GHz
4 Cores, 8 w/ HT
8,883 CPUMark Score
12 GB RAM
Dedicated servers are great because you don’t have to worry about the many pitfalls of shared hosting. If you’re selling courses online you’re probably running Learning Management System software which requires a substantial amount of processing power to make run smoothly. At Academy of Mine we’ve setup our hosting environment to be able to handle that bulky software efficiently. Similarly, we’re always monitoring our hosting environment so you don’t have to. And our up-time is great. Last month our up-time on our starter dedicated server was 99.916%, and the month before was 99.932%.
Best of all, since Academy of Mine manages our clients’ sites, our members don’t need to worry about someone uploading a rogue script that would bring down the entire system (which happens often in the world of shared hosting). We control everything on our end to make sure the hosting environment runs smoothly.
We hope you found this case study helpful!