Do the poor page load times of your website effectively hold your business down – unable to deliver strong user experience, attract high search rankings, and grow? Specifically, how does poor page loading cut into your revenue? Speed up your website to regain control of your upward trajectory. Moving to a high performance infrastructure is one essential step in the process. **WARNING: This piece contains a major potential time-waster.
- Why should your website be as fast as possible? (Stats)
- High performance infrastructure “hidden” from PageSpeed tool
- Other handy page load tools
- More ecommerce sales with a faster site? Yes.
- SSAE-16-Type-2-audited high performance infrastructure
Why should your website be as fast as possible? (Stats)
There are plenty of studies out there that indicate how critical speed is for the average user. Two of the most eye-opening studies were published a few years ago, as detailed in Econsultancy. Each of them has been circulated heavily ever since (perhaps qualifying as “classic IT market research” given their continuing relevance to understanding user behavior):
- Forrester Consulting, survey of 1048 online shoppers, 2009
- “Why Web Performance Matters,” interviews of 1500 consumers, 2010.
Here are some of the most interesting statistics from the two studies, highlighted in tandem by Kissmetrics “Minister of Propaganda” Sean Work in 2011:
- Nearly three-quarters (73%) of people who regularly surf the Web on smartphones or tablets say that they have come across sites that had unacceptably slow page load times.
- Just over half (51%) of people who access via mobile say they have either had an experience with an error message due to slowness, or have experienced a site freezing or crashing.
- Well over a third of shoppers say that they came across a site that they could not reach (38%).
- Nearly half of consumers (47%) say that they think a website should load within 2 seconds (what would make them happy), and two in five say they will leave if it doesn’t load within 3 seconds (goodbye, customers).
- If your page load times become 1 second slower, you will see your conversion rate drop as much as 7%.
- In terms of actual dollars, how much does one second of slower loading cost you? Just as an example, if your site generates $100,000 per day, one second of additional load time means you could be leaving behind $2.5 million of revenue annually.
Although these studies are both a few years old, the fact is, human behavior hasn’t changed all that much since the dawn of the information era, the early 1990s. Just look at the 1993 book referenced by Website Magazine in 2014. Penned by IT design consultant Jacob Nielsen, Usability Engineering suggests that tiny slices of time have major impacts on user perception. There are three time limits listed by Nielsen that relate to UX, in terms of basic psychology. “If the application responds instantaneously to the user’s actions, it gives an appearance of direct manipulation,” he wrote – referring to a limit of 0.1 seconds. “This phenomenon of direct manipulation is a great key to increase user engagement.” If loading instead takes 1 second, even at that point the person becomes more aware that the system is in control rather than them; they will have a second to think, but won’t become immediately disengaged.
Keep in mind, in today’s world, one second could be considered an eternity. In fact, the New York Times reported in 2012 on Google findings that even 400 milliseconds is too long for users.
Why were so many analyses being conducted on speed between 2009 and 2012? It was top news: Google officially announced that they were building speed into their algorithm as a determining SEO factor in 2010. Roger Dooley posited in Forbes, “While [Google SEO quality chief Matt] Cutts noted at the time that initially only a small percentage of sites would see a significant change in ranking or traffic due to page speed factors, I find it likely that the emphasis will increase over time.”
Dooley suggests that Google typically does not want to swing things drastically and suddenly in a way that makes it difficult for credible websites to be able to keep pace. However, he thinks the statement indicates that high performance infrastructure would be increasingly represented by the top results.
Interestingly, Dooley also thinks that the 2010 “Why Web Performance Matters” study – the one that interviewed 1500 consumers about Internet speed and that listed 2 seconds as the expected load time – had too high of timeframes. As indicated above, Google agrees, and even Nielsen does from 1993.
We can debate what the specific period of time is that a user will stick around, and obsess over that hard number, but the most important fact to take away from all these studies is that: 1. Potential buyers have time expectations; and 2. They will leave if those expectations aren’t met.
High performance infrastructure “hidden” from PageSpeed tool
As everyone knows, one of the most important tools out there, that’s very widely used, is Google’s PageSpeed Insights. Here’s a good pro tip: Dooley advises that PageSpeed does not include the speed of your network in its number. That means you could have a great score with that tool, but actually be suffering in search nonetheless because of factors having to do with your network connection and server. In other words, it’s an intentional blind-spot of that tool that could lead to many ecommerce companies and others feeling overly confident about their speed.
Other handy page load tools
Two other tools, offering different information related to page speed, are ones that offer:
- Geographical diversity: Using the Neustar Website Load Testing Platform (which you can access via a free 30-day trial), you can look at load times from different locations worldwide – so you can get a sense of geographical locations where user experience is stronger and weaker.
- Direct comparison: To look at it from a different angle, there’s a tool out there called (**Warning: major potential time waster directly follows.) Which Loads Faster? that allows you to basically race sites against one another. You can check the load time once or multiple times per site. The average milliseconds per page load is then listed, and it will tell you how many times faster the winning site is than the slower one.
More ecommerce sales with a faster site? Yes.
The various studies listed above were building on similar research from IT research firm the Aberdeen Group that was conducted in 2008 and republished by popular demand in 2015. “A 1-second delay in page load time equals 11% fewer page views, a 16% decrease in customer satisfaction, and 7% loss in conversions,” reported the Aberdeen researchers.
To better understand the need for speed directly in terms of ecommerce, let’s look again at the two studies we initially discussed that were featured in Econsultancy/Kissmetrics from a different perspective. Blogging and conversion author Sherice Jacob, in her analysis of these studies, focuses on different statistics from the studies than those mentioned above. She cites these two stats:
- More than three-quarters (79%) of visitors say that if an online store is slow to load, they won’t come back.
- Close to half (44%) say that they would mention an instance of annoyingly slow ecommerce performance to a friend.
“This means you’re not just losing conversions from visitors currently on your site, but that loss is magnified to their friends and colleagues as well,” notes Jacob. “The end result – lots of potential sales down the drain because of a few seconds difference.”
SSAE-16-Type-2-audited high performance infrastructure
We will look at some specific ways to speed up your site in the second part of this series (linked below), but one central focus must be your server and network. At Total Server Solutions, we offer an array of high performance infrastructure solutions, backed by our world-class technicians. Let’s do this!