Honest Take: Does heatmap tool slow down my website?

Yes, most behavior analytics tools out there will in fact slow down your website by around ~200ms. But, you shouldn't be stressed about it and I will explain every aspect that slows down your site.

In this article, I will be writing a clear and detailed honest take on how heatmap can slow down your website and whether you should be worried about it.

You may be wondering. If the impact is so minuscule, why would there be so many questions asking about slowing down my website for almost every single behavior tracking tool out there:

• Will Hotjar slow down my site?
• Will Crazy Egg slow down my website?
• Does Smartlook slow down my website?
• Does _______ slow down my website?

You can find one of these questions in the Google search for each and every one of the heatmap tools out there. Does that in any way incline that there is indeed a slowing down problem in the fundamental of heatmap tools?

To understand why are there so many questions revolving around this topic, you need to first understand the fundamental of the tracking codes.

How tracking code works?

Usually, how it works is you just install a seemingly simple piece of tracking code into your website whether it is Google Tag Manager, Hotjar, or anything of that nature.

It will then expand itself and fetch more codes into your website to serve its purpose.

Whether to track your website visitor's behavior, collect insights, or load more third-party scripts which is what Google Tag Manager is for, a script to load more scripts.

So, if there are too many scripts loading and taking up 1 to 2 seconds to load up everything, your website will be blanking until everything is completely loaded up.

Now, it is NOT a huge issue nowadays given that most tracking scripts have already got this covered by utilizing this technique called, asynchronous loading.

What is asynchronous loading?

Asynchronous loading is when a script loads and executes without blocking other processes, this means that scripts can load and execute at the same time without waiting for each other to finish.

So, if the external scripts take up to 1-2 seconds to load, your website will be loaded at its own pace without the need to wait for the external scripts to finish loading.

So, does asynchronous loading solves all the slowing down of your website problems?

The honest answer is partially.

However, most heatmap platforms are dismissing it by trying to lead you to think that asynchronous can solve all the problems but it is actually not.

Here are some of the lies that the other analytics platform wants you to believe.

Asynchronous loading with javascript only solves half the equation, and what is the other problem, and what is the truth that was hidden from the light?

Heavy data processing

For a user behavior analytics tool to work, you need to track everything that is happening on the website in real time.

To help you understand the complication when it comes to tracking behavior.

The easy part: the mouse movement coordination X and Y. Page scrolled positioning offset X and Y. Elements clicked with elements attribute and its relative position X and Y.

The complicated part: Every change in the website elements (DOM manipulation) such as after you clicked something, some other things pop up on the screen or fly across the screen.

All of this information will be tracked and recorded as it played out before your eye, everything happens simultaneously and your browser is processing all of it.

So, if your website has a banner that is hovering around the screen in an endless loop.

The tracking logic will read its position from X=0 to X=1000 to X=0, repeatedly for thousands and thousands of records.

Processing and running in the background of your web browser.

What it ultimately means is that the more moving part in your website, the heavier it will take to process all the website and slow down your website respectively.

Will Heatmap tool slow down my website?

Well, it is not that simple.

It ultimately depends on your service provider that does the recording and how they process your website information.

At Howuku, we pride ourselves on our in-house built recording technology that minimizes the impact of recording on our customers' websites by only tracking and recording what is necessary.

By keeping the DOM mutation process at a moderate level and finding the balance between performance and information gathered.

So far only certain providers that are fairly established have this issue where it will slow down your website and that is because they tried to collect very complete information and forgo the performance.

You can probably look at the review sites to find out if your service providers are slowing down other people's websites, but so far we didn't get any complaints about slowing down our customer's websites. 🤞*fingercross*

To prove that, we will do a performance test below with a before and after Howuku tracking script.

How to measure the performance impact?

One way to do it is by using the free website performance tool called lighthouse.

All you need to do is to measure the website performance WITHOUT the analytics tool on your website and write down the performance score.

Then, you add the analytics tool to your website and measure the website performance score again.

Now, calculate the Performance Impact with the following formula:

[1 - ( AFTER_SCORE / BEFORE_SCORE )] * 100

If the Performance Impact is less than 10%, that is an acceptable impact. Anything more than 10%, you probably should consider another tool.

Here is a performance impact test on Howuku Tracking Script on our website howuku.com.

WITHOUT Howuku Script, the website performance is 58.

WITH Howuku Script, the website performance is 56.

Quick maff
[(1 - ( 56 / 58 )) x 100 ]
= [ ( 1 - .9655 ) x 100 ]
= 3.45%

As you can see there is a performance impact but it is very minimal and that gives us a performance impact of roughly 3.45% only.

This is a popular way to measure third-party performance impact and was used by Shopify as a guideline to approve new third-party apps/plugins.

Conclusion

Does Heatmap tool (hotjar, crazyegg, smartlook, howuku, etc) slow down my website?

Yes, it does but you shouldn't worry about it as there is always a tradeoff when it comes to user behavior analytics tools.

What really matters is that can you take a small tradeoff of 3% performance impact and use it to improve your overall user experience, fix unforeseen bugs that are causing problems to your users, and increase your e-commerce site conversion rates.

I think the answer is very obvious. :)