By Casey Tuggle CMO, SocialBot and Kelly Auto Group, Inc.

Google Analytics has been lying to you for years.

There is an easy way to see this; open GA, find traffic sources, and pick a vendor (source or source/medium) who provides some kind of direct-to-VDP traffic. It can be retargeting, dynamic social ads, or a display campaign – anything that takes your customer directly to a VDP. Look at time on page and session time for that provider. With a standard implementation of Google Analytics, you will see time on page is longer than session time. How is that even possible, you ask? How can someone spend less time doing a total task than they spent doing a component of it?

The fact is Google Analytics is broken, and by exploring this further, you’ll see how this may have caused you to misunderstand your advertising ROI. In this article, we’ll focus specifically on how bounces are defined, and how you can use Google Tag Manager to correct that important metric.

Google Analytics was designed for traditional e-commerce; you find something on a search results page, add it to your cart, and then check out. It was designed to show how much traffic your site is getting from organic search, paid search, and your other channels. It’s a great foundational tool, but some parts of it do not account for the way the modern internet works. Why hasn’t Google fixed this? They have to an extent, but they did it in such a way that it didn’t disrupt the GA product for core e-commerce clients. They created Google Tag Manager which integrates into Google Analytics. Google Tag Manager gives us new pixel capability and a powerful set of tools to provide additional data on site traffic delivered by modern providers.

GTM allows us to create a pixel which fires when an event is triggered instead of just once per page load. The traditional GA tracking pixel fires only when a page loads, and that is why GA can’t tell how long a user spent on a session if only one page was visited. Without the pixel firing again on a second page, there was no second event to measure and Google knew nothing about what happened on the first page visited. In this math, Google understands 1 minute on page 1, plus no visit to page 2, means an average session time of 0 seconds. The session time for this single-page traffic is unrealistically low and the bounce rate high.

A user who visits just one page is a “bounce.” We have been conditioned to believe bounces are bad. This is not true for the auto industry with standard GA. If bounces were bad, how are retargeting vendors still in business? They put users straight onto the VDP every time, and if we believe the data on VDPs being the most important page for conversions, there is a substantial amount of value from a direct-to-VDP advertiser. A bounce does not take traffic quality into account, and the VDP engagement we are looking to create will instead look like a failure. We need GTM to solve this problem.

How to create and connect a Tag Manager account to Google Analytics: 

Start by going to tagmanager.google.com. Click the create account button, and complete the next couple of simple fields with your web address and click to accept the terms. Be advised, the checkbox does have you confirm that you are GDPR compliant. Next, you will get a couple of snippets of code to install on your site, including one for the header and one for the body. This should be familiar, but know that instead of just adding yet another tracking pixel to your page, GTM can act as a “container” for other pixels and can even help speed up your page load speed if you’re using several currently.

Next, establish variables, triggers, and tags:

Click “Variables” in the left-hand navigation. Google helpfully gives you some built-in variables to edit as a starting point, so just edit the Event variable. Click on Configure, then scroll down the list to Scroll Depth Variable and select it without deselecting the others.

Select “Triggers” in the left-hand navigation, and click New. Select the box to define a Trigger, and select “Scroll Depth.” Select Vertical Scroll Depths and enter a percentage of scroll depth that you would like to record as an engagement event. Realistically, most people aren’t scrolling 90% of the way down a page. However, in order to get to the vehicle features area of your VDP they’ll probably need to scroll at least 25% of the way. That’s at least a positive scrolling action the user had to engage in. Let’s put in 25%, at least for now. Name your Trigger something descriptive.

Configure a Tag. Click on GA as the tag type, and Event as the track type. It will ask for a Tracking ID. To get this go to GA, click Admin, find your page, and copy the number that starts with UA. Next, select Triggering, click on the Scroll Depth Trigger you just created, name the tag, and save! That’s your first GTM tag. There is a “Submit” button on the upper right you can use to deploy the code live to your site, automatically, without having to call any developers again.

Once completed, this will fire a pixel to Google Analytics as an “event’ every time a user scrolls to 25% of a page. A user who does this will no longer be factored as a bounce as an event will be recorded. Tip: You can create and track whatever events you find important for engagement, but keep it realistic. It’s probably a good idea to watch user sessions through a recording software like Hotjar or Lucky Orange and see what metrics are realistic for your dealership.

GTM events will redefine the bounce rate in Google Analytics. GA now has pixel data every time this scroll depth event is triggered, and it can tell engaged traffic from unengaged traffic. If a user clicks on several photos, scrolls through features, and then clicks on the dealership phone number from the VDP, that’s a great session and can be shown as engaged by firing multiple events. If a user lands on the VDP, takes no action and backs out (like a mistaken click), a bounce is still recorded just like before. Tada!

Now you can correct the broken GA bounce rates using Google Tag Manager. The implementation of GTM will enrich your reporting, allow you to really measure ROI for traffic that was previously ambiguous or looked low-quality, and give you the ability to evaluate new vendors in the future. Weighing this data correctly will help you get more leads with less money, and sell more vehicles.