I recently spoke at the Digital Summit* in Minneapolis, and I talked about using Google Analytics, Google Tag Manager, and Microsoft Power BI to create a nifty, real-time content performance dashboard. Since I only had 30 minutes, I couldn’t cover every little detail. This three-part blog series serves to fill in those gaps and provide a comprehensive tutorial on how to configure the dashboard. Ready? Strap on your data geek hats! (Because data geek hats would have straps, naturally.)
I’m focusing on a specific metric, the bounce rate, and why it can be misleading. I’m then going to show you what I think is a better measure, and we’ll create that measure in Google Tag Manager and Google Analytics. Finally, we’ll create a dashboard using Microsoft Power BI to visualize this information.
- There’s a lot of foundational measurement stuff I’m not covering in this blog post. Measurement strategy, measurement plans, testing of analytics, and much more.
- If you’re new to Web analytics, or you’re a marketer who’s also been tasked with being the “data guy,” check out Google’s Google Analytics Academy and Avinash Kaushik’s blog as starting points. Also check out the Power BI team’s blog.
- The tools and platforms discussed here – Google Analytics, Google Tag Manager, and Power BI – are personal favorites of mine. You you may have some you like better. Great! I’ve chosen these because of their user-friendliness. Oh, and they’re free. 🙂
- This dashboard is only the tip of the iceberg for content insights. Even if you think my tutorial is lame, explore your data with Power BI or another visualization tool. My goal is to get you immersed in your organization’s data.
- Credit where credit is due: This tutorial utilizes a Google Analytics scroll-tracking script that was created by Rob Flaherty and then made into a Google Tag Manager tutorial by Andy Gibson.
- A Google Analytics account and property (you’ll need a Gmail address / Google account as well)
- A Google Tag Manager account and container (again, you’ll need a Google account for this)
- Microsoft Power BI Desktop downloaded and installed (you don’t need a Microsoft account to use Power BI, but you will need one if you want to publish your data to Power BI online)
- Rob Flaherty’s scroll tracking script downloaded to your computer. We’ll need it momentarily.
The bounce rate, as many of you already know, is traditionally defined as the percentage of single-page visits for a page or set of pages. In other words, a bounce means that user left the site from the same page they arrived at, without viewing any subsequent content. Most people view this as “bad,” and sometimes that’s true. It would be hard to argue, for example, that a high bounce rate on your home page is a good thing.
But sometimes a bad bounce rate can be misleading. A bounce must not always be interpreted as bad, because even within a bounced session, a desired action could have occurred! For example, a bounced user could have downloaded an asset, spent a lot of time on your content, or submitted a form. So I’d like to see a deeper content engagement metric before freaking out about the bounce rate. That’s why we’re going to use Google Tag Manager to configure a scroll tracking script. The scroll tracker data will appear in our Google Analytics account, and it will tell us how deep users are scrolling on our pages, in increments of 25%. So we can know the percentage of people who scroll 100% of the way to the bottom of our site’s pages, for example. Keep in mind there are other ways to tweak the bounce rate, but I like to keep the bounce rate “pure” (as a measure of single-page visits), and then use other metrics to explore content performance. Enter the scroll tracker!
Bounce rate – helpful tip
“All data in aggreagate is useless.” –Avinash Kaushik
Don’t look at your website’s overall bounce rate; what does that tell you? Instead, look at individual pages. In Google Analytics, use the Behavior > Site Content > All Pages report to view detailed metrics for each of your website’s pages, including bounce rate. How’s your home page doing? Your about page? Product detail pages? Viewing bounce rates for specific pages allows you to take specific action.
*Originally presented at the 2016 Digital Summit