Collecting Big Data: How To Know If Your Site Is Too Slow?

New Series: Collecting Big Data – how to know if your site is too slow?

How to know if your site is too slow? Here’s an uncomfortable truth: if your site is slow, you’re flushing burning money down the drain.

Maybe I’m just old-school at Big Data Social Media, but I still predominantly use my desktop for browsing and for work. But I’m an anomaly (tear).

The majority of the world is on mobile. The majority of all searches, of all website traffic, and all video views is on MOBILE.

So why collect big data? To get answers and evolve. Your visitors won’t visit you out of charity.

Why Is Mobile So Important?

I know that we are moving into a mobile dominant future and economy, and having a fast, responsive, EASY TO NAVIGATE, mobile experience, is no longer optional.

In fact I would go so far as to say that I’ve designed first and foremost for mobile, and the fact it works well on desktop or tablet is a nice bonus.

Slow websites with poor website performance are the bain of, well, everyone. Get your website speed right, and rest will take care of itself.

Why Site Speed Is Important:

Firstly, let me just say that I practice what I preach, and that I’ve made an active effort to make sure my website is lighting fast. Is there room for improvement? Sure there always is, but I’m well ahead of the majority of sites out there.

Here are screenshots from recent tests I did on Google’s Page Speed Insights and GT Metrix (both free tools). Additionally Google Analytics can be setup to track page speed performance across your whole website, and how each page compares relative to other pages for speed.

How To Know If Your Site Is Too Slow:

The page speed test above, are just that: tests.

But how can you prove definitively whether page load speed is a problem or not? Afterall, it’s somewhat irrelevant if you scored a 93% or a 39%, if the performance of your site is unaffected. So how do you know?

Page Load Speed Engagement Level Events

Okay, stick with me here as this gets a little complicated if you’re not familiar with Google Tag Manager (GTM). In future posts I’ll go into more detail about how this all works, but for now let me show you the basics.

In GTM, you can create custom events. Essential, these will fire to Google Analytics once specific criteria have been met.

So here’s what I did at Big Data Social Media. I setup 4 custom events I use to track “Engagement Level”, which are:

  1. Page View
  2. Window Loaded
  3. 5 Seconds on Page
  4. 30 Seconds on Page OR Link Click

Huh? This Looks Confusing

Here’s how I break it down:

  • Page View – this is fired the very instant the website is requested to load. This essentially says “watch out world, I’m loading!”
  • Window Loaded – this is fired when the page has fully finished rendering. As in, you can see everything on the page and can start to interact with it. If there is a big dropoff between page view and window loaded it suggests big load speed problems (people saw a white screen and just smashed the back button!)
  • 5 Seconds on Page – as the name implies, this fires 5 seconds after the Page View fired (not after the Window Loaded fired). Again, a massive dropoff here suggests page load speed issues, OR your website is visually offensive to your visitor, or perhaps just wasn’t what they were looking for.
  • 30 Seconds on Page OR Link Click – this is an important one, if people are still around after 30 seconds the great! You’ve probably built a good and useful site your visitor is enjoying. However I added the OR statement, as many people may click around and go to another page before 30 seconds comes. If this happens, again, great! You’ve got a site people are curious to explore.

Brining It Together In Analytics:

All together these metrics help to give a very quick snapshot of whether your visitors are happy coming to your site, or whether they very quickly want to run away as fast as humanly (or computerly?) possible. That’s why we collect big data.

Google Analytics gives great analytics tools and includes the bounce rate as a default metric, but in truth it’s blunt and doesn’t really tell you much. All it says it that someone left without clicking on something.

The difference is a butcher’s cleaver vs a surgical incision.

Cool, but how long were they there? Was it an IMMEDIATE bounce or did they stick around a while? Those subtle differences tell you a whole world about how your website load speed is performing, and give you actionable insights to drastically make things better.

Afterall, people form meaningful first impressions within the first 50 milliseconds. That’s fast.

The Big Data Social Media Way:

Want to go deeper?

We can break down page load speed by device, by data source or by social media platform. If you want to see whether mobile performs better than desktop, or whether your organic traffic sticks around longer than your paid traffic, you can break it all down.

This helps to debug and figure out where the real problems are happening and really how to know if your site is too slow.

Remember, just give your users a great experience, and they might actually stick around a while and read what you have to say, or buy your stuff.

Best of luck.

Love & Power,



Big Data Social Media

Leave a Reply

Your email address will not be published.