The Basics of Behavioral Analytics

If it’s worth doing, it’s also worth measuring – while it sounds like a famous quote, it’s really just a basic part about doing business in just about any industry. And when your business involves a digital product, the data you generate will you with everything from evolving new features for your product to fixing bugs as well as improving it along the way.

Most businesses should have some understanding of search engine optimization (SEO) and to some degree, app store optimization (ASO), which are heavily driven by their own flavor of analytics usually called behavioral analytics. While there is some overlap with marketing analytics, this process goes much deeper to track and measure granular details from user activity. Here, we’re going to distinguish this from other, more familiar analytics processes and explain a bit about how they work.

Behavioral analytics: not your typical marketing analytics

Every business with a website should have some understanding of SEO and if you’re serious about building a real web presence, you put effort into creating engaging content and studying your competition. You also take the five or so minutes to set up a Google Analytics property and drop the tracking script onto your site and you’re good to go.

Once you break past the learning curve and all the different numbers, it’s just like any other game where the goal is to raise your score – if we ignore the majority of your marketing or sales funnel, the goal here is mainly to get eyes on your stuff. Web analytics services track where users originate and the places they visit on your site, all the while counting the time they spend on each page. Services like Google Analytics 4 does the same things for apps.

Behavioral analytics looks beneath a user’s overall trajectory that’s (mostly) captured by analytics more geared for marketing purposes to understand exactly what they’re doing. Depending on the device and input methods (e.g., a touchscreen, a keyboard and mouse, etc.), user analytics captures and often interprets user actions taken as they interact with the app.

An overview of how user behavioral analytics systems work

There are a few ways that these systems collect data.

Almost everything will generate an “event” in your system when a user takes an action, for example, interacting with an element they (hopefully) see on their screen. An element could be anything from an:

• Image 
• Button 
• Form 
• Lightbox overlay 
• Embedded podcast
• Video live stream 

Basically, an element could be just about anything, even basic text copied to their clipboard. Events are what happens when a user does something like press a key on their keyboard, where they move their mouse, where they touch on the screen, button presses and analog stick or trigger positions on a gaming controller, and so on, to interact with an element they see. Some examples of an event might be:

• Details of how a user interacts with a scroll such as where touch or click it and where they move it to
• Completing tasks while using an interactive element on a webpage – a Google Doodle like Pac-Man is an over-the-top example but you could track things like user score on game over, how many ghosts are chomped in a playthrough, and much more.
• The viewable area on an app screen when a user backs out, closes the app, or when it crashes.

Depending on the design of your app, you’ll have all kinds of different events that will surface. While some programs will be configured out-of-the-box to look at and measure various specific events, the experience will differ greatly between apps which means you’ll need to work with your team to figure out which events are meaningful and in what context. As your apps grow, analytics will need occasional freshen-ups to ensure that monitored events are still meaningful and to understand new or revised portions of your app.

Each analytics platform is different but the general idea is that these events are referenced somewhere usually a TMS (Tag Management System) so they can be tracked. Configuring events for Google Analytics is done using Google Tag Manager where you can identify your events and setup components like triggers and measure code. Watch the short video in the link for a brief of their system as most solutions are similar in design.

What to track?

The design and purpose of your app will dictate what events to track and how to configure triggers. Keep in mind, that the actual implementation is usually handled by a developer who’s familiar with the site or app to configure and maintain some of the more complex tags and the code.

For example, here’s a detailed guide on how to use Google Tag Manager to track comments on a site or app where the author explains the process of identifying the provider who built the commenting plugin, creating a tag and setting parameters, dropping in an AJAX script for configuring a listener to be used as a trigger, and then some. It’s pretty dense but what you should take away from it is that the process can vary substantially where, in some cases, you or your developers will be able to find tried-and-true snippets of code that play nicely for all the underlying technology on your website but in other cases, you’ll need to write it from scratch.

Your best bet to get the most out of your data is to have someone close to your product and familiar with all the technologies to ensure you’re getting the best fit. As such, it’s usually best to ensure that analytics are built alongside your product as you grow, rather than implement these tools as an afterthought.

We use analytics to keep your product & UX on point

Without analytics, you’re doing your product a disservice as there’s no real way to know how users are interacting with your product. Developing and maintaining tags and their logic for user event tracking can reveal everything from issues with the product to opportunities for improvement for everything you care to monitor in your product. Feel free to get in touch with us at Blue Label Labs to learn more about user analytics or to discuss your idea for a project.