Page Speed And Conversion Optimization
When most clients come to me for conversion optimization, I generally do some basic investigation — like checking page load time — and I generally get the same response:
- “When are we going to do conversion optimization?”
- “I thought you did conversion optimization. Why are we doing this?”
- “You know, I think our site’s okay. Now, it’s time to focus on conversion optimization.”
- “So, I heard that button color can have an impact on conversion optimization …”
Thankfully, most clients are understanding, and not nearly as passive aggressive.
Nonetheless, I see a significant blind spot in most conversion optimization discussions.
The blind spot is this: page load time.
Page load time has a huge impact on conversion optimization. More details to come on that topic.
However, load time is mistakenly viewed as under the control of the developer or designer. Surely a conversion optimizer should be spending his time on more important things like CTA buttons, page length, button color and checkout process, right?
It is my view that page load time is one of the most overlooked areas of conversion optimization, yet also one of the most direct ways to improve your own website’s conversion rates.
Let’s dig into the data, and I’ll show you exactly why your page load time matters, big time, for conversion optimization.
What is page load time?
Let me start at the beginning and explain a few important details regarding page speed.
According to Google, page load time is “the average amount of time (in seconds) it takes that page to load, from initiation of the page view (e.g., click on a page link) to load completion in the browser.”
Let me simplify that: Page load time is how long it takes for a page to show up on your screen.
There’s a huge amount of information that I’m totally skipping over, however. Page load time is an extremely technical field, full of mind-numbing data and flush with indecipherable jargon.
Let me descend for a moment into the depths of techy, just to satisfy any CRO geeks with a curiosity about technicalities.
The average page load time is an amalgam of both the network and server time, plus the browser time. Onload time, also known as "document complete time," includes the total load time of all onpage resources, including all backend resources and analytics trackers.
Although the full load time may be lengthy, according to speed tracking tools, the page may have full functionality within just a few seconds. That’s a good thing.
As long as the page feels fast to the user, then you’ll have the conversion upside that you’re looking for. Generally speaking, however, page speed and user experience speed (how fast it feels) are positively correlated.
The bottom line is this: Your site needs to load fast in order to get more conversions.
So, how fast (or slow) is your site?
Before you read all the eye-popping, mind-blowing information below, I want you to do something.
I want you to test your site speed. Go ahead; I’ll wait.
There are two great places I recommend you check your website speed
I like Pingdom in a geeky, totally unromantic sort of way.
The data is remarkable. And it’s completely free.
But you don’t have to be a datahead to enjoy the raw benefits of Pingdom. Just open it up, plug in your URL, and let the magic flow.
Pingdom gives you the information in seconds. The most important metric above is the “load time.” Google.com clocks in at 957 milliseconds, or .95 seconds.
Pingdom breaks all the information down into little visual chunks so you can geek out over it a lot easier.
This is a good and basic way to start.
But there is data — oh, the data.
2. PageSpeed Insights
Another free and useful tool comes from the King of the Internet. (Not Elvis.)
Google’s PageSpeed Insights provides a clean visual interface and easy-to-follow instructions for total optimization.
My complaint with Google has to do with the way they score it. Instead of giving you the data in seconds, they provide a score from 1 to 100. This is slightly less useful for comparison purposes.
For purposes of this article, it’s helpful to remember the number, in seconds, of your site load time. You get that number from Pingdom, not Google.
How does page load time affect conversions?
Conversion optimization could more aptly be called “user optimization.” The better a website is for the user, the better it will produce conversions. It’s just that simple.
So, to get specific, the faster a website loads, the easier it is for the user, and therefore more enjoyable and conversion-ready.
Here’s the simple facts:
- A page that loads slowly will have lower conversion rates.
- A page that loads quickly will have higher conversion rates.
Let me share some stories.
WordStream constantly tests everything. They live and die by data, data, data. So, it’s no surprise that they religiously track conversion rates.
The goal of WordStream, as with most websites, is more conversions. When they score higher conversion rates, they’re able to pat a team member on the back and say “Good job. That was a nice move!”
But when conversion rates suddenly and unexpectedly rose without apparent intentionality, they were nonplussed. Why did their organic search traffic suddenly experience a major conversion uptick?
After some investigation, they focused on the awesome hero: page load speed.
According to author Chad Summerhill, “Milliseconds matter.” And for his website, those extra milliseconds gave their page load time a big push.”
One Leadpages study analyzed the website landing page of entrepreneur Juan Martitegui. Martitegui ran several tests to find out which of his landing pages would get higher conversion rates — one asking for someone’s first name on a form, or not asking for a first name.
As often happens in conversion optimization, he ended up getting results that had nothing to do with first name vs. no first name test. The tests were not delivering consistent results. Curious, he dug in deeper, and decided to test the pages on different servers, with different load times.
That’s when the magic happened.
Faster pages delivered more conversions, regardless of the capture form variables.
Mozilla gets a lot of downloads, but they wanted more. Clearly, speed (or lack thereof) was putting the kibosh on a considerable amount of download potential. This abysmal page load speed was painfully obvious in this timed video comparing three browsers.
They figured out, no big surprise, that they needed to make faster pages. So the developers scurried off to their tri-screen cubicles to make faster pages.
And then they released the results. Pages got faster. And visitors started downloading more Firefoxes.
When the results came in, they had to hold on to their socks, due to their socks nearly being blown off. The results shocked even the most aggressive page speed promoter among them.
To put that into nice fat numbers with lots of zeros, Mozilla could potentially gain an additional 60,000,000 Firefox downloads by chopping a little bitty dos segundos of their page load time.
Once upon a time, Shopzilla decided to redo their website.
Good thing they did. Not only were the holidays coming, but they were also suffering under the curse of 6-9-second page load times. By anyone’s standards, 6-9 seconds is awful.
Once the shiny new website was up and purring away, pages were loading at the Formula One speed of 1.2 seconds.
Fast pages. Happy customers?
Apparently so. Although Shopzilla implemented other improvements, page speed provided the biggest statistical and measurable increase on the user side. The results were nothing short of remarkable. Those users started spending more money on the site — a lot more.
After the new website launched, users were shelling out cash 12% higher than before.
But enough stories. Let’s go for more data and numbers — the kind that very well might blow your socks off.
Now that you’ve heard about how awesome page speed upticks are for other people, do a little noodling on the kind of improvements that your page might enjoy.
Long load time? You’ll lose tons of users.
A Kissmetrics infographic traces out the percentage of users who abandon a page based on rising load times. By the time your page load time hits 4 seconds, you’ve lost a fourth of all visitors — a potential quarter of conversions!
Wow. Finicky users, huh? Yep, and those are your customers. Don’t gripe. This is what you have to deal with.
Kissmetrics dug in a little deeper, and discovered a few more gems. Website performance, i.e., page load time, is a key predictor of shopper behavior.
Here’s the brass tacks, taken from the infographic excerpt below.
- Nearly half of your customers expect your site to load in 2 seconds or less. What’s your page load time?
- 40% of your customers will leave your website after 3 seconds.
- 79% of your customers, if they are dissatisfied with the site, will never come back.
- More than half of all customers assert that a fast load time is something that will improve their loyalty to the site.
- Every second that goes by, your customers will lose 16% of their satisfaction.
- 44% of customers will gripe about your awful website to their friends.
One-Second Delay: Lose 2.5 million in sales.
If your website typically earns $100,000 per day, a 1-second page delay could cost you a few mil in annual revenue.
This data is based on accurate measurements of user behavior and hypothetical site sales.
I don’t know about you, but I kind of like an extra 2.5 million bucks.
Check out Walmart’s exploration of page speed reduction. The conversions nosedived as the site speed grew from 2-4 seconds.
Website optimizers spiffed up one website, dropping the load time from .6 seconds (that’s fast), to only .23 seconds (that’s wicked fast).
Just milliseconds of improvements helped users stay 26% longer, bounce 8% less, and view more pages per visit.
What’s a good load time?
The faster the better.
- 1-2 seconds: Good.
- 3-6 seconds: Average, but try to improve it.
- 7-10 seconds: Poor. Get to work.
- 10 seconds or more: I’m very, very sorry.
Remember, speed affects conversions. Long delays, fewer conversions:
A lot of people, 47% of people if you need to know, expect a website to load in 2 seconds. After 3 seconds, 40% of site visitors are gone forever.
That conversion optimization advice about “2 seconds to make an impression” never sounded so true.
But even if you can shave a mere 1 second off your load time, you’re doing alright. Two seconds? Even more. Four, six, eight? You’ll see stuff like this happen:
So let’s speed up your site.
The best place to go for speed improvement advice is as follows:
- Google PageSpeed Insights has the most succinct and actionable advice for improving your page speed. Have your developer start working on things.
- Kinsta provides a helpful tutorial on page speed and how to make it better.
- Kissmetrics has seven brief points on improving page speed. It’s simple, but practical and pragmatic.
- ConversionXL provides an easy-to-understand guide for improving website speed.
Page speed isn’t just about latency, server response time and other technical details. The science of psychology affirms the need for users to have an optimum chronological experience when they visit your website.
Nobody likes to wait. The people who least deserve to wait are your customers. So give them speed. Give them service.
And give yourself some major conversion improvement.