Google Analytics (GA) is a great tool to monitor key statistics about your website’s user behavior. You can track everything from how many users come to your website, to what percentage of those users fill out a specific form. It’s a powerful tool to help marketers make informed decisions about their website and marketing campaigns. However, getting the most out of GA is dependent on one thing: accurate data. And bots are the enemy of accurate data.

Data can quickly become inaccurate when GA captures sessions for autonomous bots. These sessions are not driven by your users, but by computers. If the data’s purpose is to be a tool to monitor and interpret user behavior and interactions with your website, we want to eliminate anything that isn’t a user. Fortunately, GA provides the tools needed to clean up and prevent 99.9% of bot traffic.

Why is bot traffic bad?

When making informed decisions about user behavior and campaign efficacy supported by analytical data, we need the analytical data to be driven by real human users. We want our decisions to be for the users, not for the bots. When a large number of bots are logging session data in GA, it’s going to skew the GA data. Analytical data like bounce rate, time on site, pages per visit, and the number of sessions are some of the statistics which will be negatively impacted. These stats are important for making key user behavior decisions for your website and marketing campaigns.

When the standard bot goes to a website, it typically exits a landing page in less than a second. This can count as a bounce session in GA. If there is a large number of bot sessions, this will inflate the number of sessions in GA, drive up the bounce rate average, reduce the average time per session and per page, and reduce the conversion rate. The good news is that we have the ability to clean that skewed data up and prevent it from being compromised moving forward.

What does bot traffic look like?

Before taking any action, make certain your GA data is in fact infected with spambots. We only want to remove illegitimate data. Make certain to properly identify spambots that need to be filtered out of the GA data. So how can you identify traffic as bot sessions? There are various methods of identifying spambot traffic, but here are two examples that should help show what bots in GA look like:

Example One

Spambots can be identified by a sudden high volume of sessions coming from a single location or network. 99% of the time, this traffic will have a >99% bounce rate and <1-sec average time on time on the site.


Example one

Example Two

Bot traffic can also come from various uniquely named networks.


Example Two

Google Service Provider & Network Domain Update

In February of 2020, Google officially stopped collecting user’s service provider and network domain information. The service provider and network domain were two user values that were the most helpful in identifying bots. At this time, those two values can only be used to weed out bot traffic from before February 2020. This has made it harder to identify and filter out bot traffic, but not impossible. Look for a common identifier that has a high bounce rate and lower pages per visit and use it to filter out the bot traffic from the GA data.

How to block bot traffic in Google Analytics

The first method to blocking bot traffic in Google Analytics is to enable the automated bot filter setting in the GA view settings. This setting is not enabled by default, but it’s something I would recommend enabling if you want to purge 99% of bots from your GA data. This setting can be found under the “View Settings” in your GA Admin settings.


Box Filtering

To block any remaining bot traffic, we’ll need to identify it and manually set up a filter to prevent data with similar properties being logged in GA. For example, I have found a lot of bot traffic in the past to come from the same network domain and the same city. These are types of data that can be used as identifiers in a filter to remove the traffic. You may need to use a regular expression to properly target the identifier. I’d recommend only using regular expressions if you are experienced with them. Filters can be created under Admin > View Settings > Personal Tools & Assets > Filters.


Filter Example

Unfortunately, filters do not clean up existing data. They only prevent specified bots from being logged in GA moving forward. We can set up segments in GA in a similar fashion to the view filter to purge the existing data of the bots. When setting up the segment setup the segment using the same settings as the filter and then apply the segment. Segments can be created under Admin > View > Personal Tools & Assets > Segments.


Segment

To apply a segment after it’s created, select “Add Segment” above the graph when viewing sets of data. Select the segment the bot filter segment and this will filter out all bots from the GA data being viewed.


New Segment

Living your best, bot-free life

At the time of this article being written, these are the best solutions for preventing bots from being logged by Google Analytics and skewing your data. It’s possible that more bots will need to be filtered, in the future, as time goes on. Continue monitoring your GA data for irregularities and investigate what is causing those irregularities. Authentic and accurate GA data will lead to more informed decisions that help you develop successful campaigns and user experiences. If you’d like some professional assistance with resolving issues with Google Analytics, please contact us today!