Analytics tools should be able to differentiate between real traffic vs bot traffic (I'm looking at your server-side analytics). At HockeyStack, we make this possible through three different methods:

navigator.webdriver

This is not the most robust method, but it is the easiest to explain. Webdrivers control browsers remotely and act like real users to perform a variety of tasks, rangin from simple automations to data crawling. By default, they should enable this variable in the navigator object, so we check it to eliminate possible bot traffic.

IP Address

This is a lot more common. Most crawlers and performance tools have specific IP adresses that they reach their website from. We keep and update a list of those in order to recognize the traffic. Keep in mind that we don't store any tracked IP addresses, and these are just deleted once a response has been sent from the server.

User-Agent

Very much like IP addresses, we store a list of terms that bot traffic insert to their user agents. This is updated the most frequently and is checked first on the browser-side in order to prevent any requests being sent to our servers.

<aside> 💡 If you suspect that HockeyStack is not able to detect a certain bot, we are more than happy to update our lists! Just write us through the live support chat or send an email to hello@hockeystack.com.

</aside>

Last edited: 15 June 2021


If you have any other questions or issues, you can always reach us through the live support or just sending an email to hello@hockeystack.com!