How do industry-leading firms create great products? In a word: data.
To build products that people love and want to use, successful companies collect and analyze data on what their current (and potential) customers do and don’t do. In brief, this is what’s called “behavioral analytics.”
Behavioral analytics enable you to keep your finger on the pulse of what customers are doing and what they want to do. With these kinds of insights, you can create the best possible products to anticipate and meet their needs. Read on to learn more about how you can apply behavioral analytics in your own business or organization.
What is behavioral analytics?
Every time visitors engage with your digital channels and products, they convey important clues about their desires, expectations, requirements, and purchasing capability. All of these clues comprise behavioral data. You can use this kind of data to:
- Plan new features and capabilities that match user demand
- Optimize existing products and flows for real user behavioral patterns
- Find and eliminate bugs and other issues that frustrate and block users
- Build up your understanding of your different client profiles or personas
- Provide proactive and well-informed customer care
- Measure and gain deeper insight into digital performance metrics
Although you can use many different types of data and analytics to accomplish these goals, behavioral data is unique. Unlike most other types of digital analytics, it provides a human angle on the performance of your digital products.
Of course, any large digital experience dataset exists by aggregating human behaviors and choices together into patterns and trends. But in behavioral analytics, you can zoom back in from the macro level and evaluate those data points at the level of an individual human experience, with the context that comes with it. In fact, this is usually where the greatest insight comes from.
For example, an aggregated dataset of exit rates for a webpage only has meaning in the aggregate – if you pick any individual user out of the set, looking at them alone doesn’t offer any further insight. Either they exited, or they didn’t.
With a behavioral analytics approach, on the other hand, individual choices like these are not divorced from the story of the user that makes them. In a behavioral dataset, we can pick out one user, and then examine the behaviors that led up to the moment of exit. We can identify some behaviors that indicate user frustration, impatience, or confusion. Then, we can make an educated guess at the goals and intentions of the user.
Essentially, rather than treating user actions just as numbers, behavioral analytics treats them as stories.
By looking at data in this way, you can learn a lot more about current and future customers. You can predict consumers’ needs in advance; you can determine where different users are in the customer lifecycle, what information or interaction options they’ll need next, and what questions or difficulties they may face next.
What types of data are collected in behavioral analytics?
1. Mouse movements
The way a website visitor moves their mouse around as they use your site can contain a lot of information about their thinking. It can help you determine which content the user is most interested in, where they focus their interest, the pace they are working at, their level of certainty, and more.
For example, consider a user who moves their mouse moves slowly and methodically along a body of text, versus one who moves the mouse straight to the button at the bottom, versus yet another whose mouse tends to linger around the images. Even if all 3 visited the same page and took the same action at the end, their mouse movements make clear the differences between their intentions, interests, and mental state.
Combined with other relevant customer behavioral data, you’ll be able to get even deeper into these questions.
2. Clicks (and taps)
What users click on, when they click, and how they click (as well as whether they click at all) can all tell you more about them.
Users may expect to visit a new page, or experience a new interaction, or execute a specific function when they click on an image, piece of text, or other section of the website. However, if the outcome of their click is different from what they expected, they might end up highly irritated.
(Note that in general, what’s true for clicks on desktop devices applies in a similar manner for taps on mobile devices.)
Tracking the number of clicks on an element you want to encourage engagement with can serve as a valuable UX/product KPI. When it comes to analyzing clicks at an individual level, you can look for different telltale signs:
Rage clicks
A rage click is actually a series of clicks, done in quick succession as a user clicks repeatedly on the same element.
Rage clicks are a clear sign of user frustration, and can be caused by a bug, by site slowness, or by mismatched expectations of how your UI will work. Whatever the cause, they’re a great opportunity to find ways you can fix your user experience.
Dead clicks
A dead click is simply a click that has no outcome at all. While they are not a guaranteed sign that something’s wrong, dead clicks can often have similar causes to rage clicks – such as elements that aren’t working, or appear to be clickable even though they aren’t.
A dead click may indicate someone at a lower level of frustration than a rage clicker; it may also indicate someone who is exploring for information that they’ve been unable to find so far. Context will usually help you understand the specific reason, and how to treat it.
Error clicks
An error click is a click that occurs right before an error. These types of clicks are most helpful for finding bugs and analyzing the events that are causing them.
Read more: Identifying user frustration through click behaviors
3. Page scrolling
Scroll behavior is all about how far down the page your visitors read before clicking a link to a new page, scrolling back up again, or exiting the website or app entirely.
You can use page scrolling data to identify the most crucial content on a page, and ensure that it is clearly displayed and readily available in areas where users will see it.
The manner in which a user scrolls through a page can also indicate something about their experience. For example, a slow scroll can mean they are thoroughly reading your content, or looking carefully for a specific call to action. Rapid, random scrolling on the other hand can be a sign that the user feels lost and is trying hopelessly to find something – or perhaps that they are just bored.
Page scrolling data can be insightful in a number of ways, depending on your intentions for a given webpage. For example, you can use the data to observe whether visitors are finding what they want without having to scroll below the fold; or whether they’re actually fully experiencing a page that was designed to be scrolled all the way through; or whether most of them aren’t even reaching the visual assets your team spent lots of time creating.
In this way you can not only make your site easier for the user, but you can also better use your team’s time and resources by better understanding what users are viewing and experiencing.
4. Event completions
You can, of course, track overall conversions for major goals using traditional analytics. Still, behavioral analytics enable you to look at more precise event completions to track the micro-conversions that advance visitors through a user journey or conversion pathway.
This may take the form of monitoring a customer’s use of a specific feature, interactions with a target element, clicks on a preferred CTA, form starts or finishes, and more.
When you combine event completion data with the other pieces of behavioral data you’re collecting, you can learn a lot more about who is or isn’t completing goals, and why.
Imagine, for example, that you identify the users who clicked a specific piece of promotional content, watched it to the end, and then signed up for a free trial in a single session. Then, you identify and look at the users who didn’t finish watching that video, and investigate where they went instead – and whether they ultimately signed up via some other pathway, or never converted.
Just by analyzing this one fork in a single conversion pathway, you’ll be able to reveal something about who you’re successfully reaching with your marketing, and how you can remedy what’s missing for those who aren’t yet swayed.
5. Page navigations
Besides all the different user behaviors that happen within a single page, navigations between pages are also an important part of understanding users’ experiences on your digital platforms.
The biggest and most obvious pieces of data in this category, of course, are where the user is navigating to and from. Even this one piece of information can fuel a variety of investigative directions:
- When there are many navigation options to choose from, you can learn which ones are the most appealing to users, and evaluate whether your CTAs are driving the kind of choices that you intend to.
- When there are specific pathways you are expecting or hoping for users to take, you can learn whether or not actual navigation behaviors confirm or upset your expectations.
- When there is a series of navigations from A to B and back to A again, this is called “backtracking” or “pogo-sticking.” It may well be an indicator of a navigational dead-end or a false lead as the user searches for specific information.
As you find out what content piques users’ interest and which CTAs or visuals tempt them to “turn the page,” so to speak, you may decide to reorganize your page contents, or even your broader informational hierarchy, to improve the user experience and boost conversion rates.
You can also observe metrics like how many navigations a user makes within their session, to learn more about the purpose and trajectory of their visit.
Lastly, when you aggregate navigation data together, it’s extremely helpful for identifying common user paths through your site or app. Although most of us are familiar with the idea of the “user journey,” it’s quite difficult to actually answer the question, “What is my product’s user journey?” With behavioral analytics, answering this question becomes easier (even if it’s still pretty complex).
6. User feedback
Another essential element of behavioral analytics is in-app surveys and nudges for service feedback. Customer behavior insights can be greatly enriched when you actively solicit suggestions from the people who connect with and use your website or app.
By requesting input, you can reach an even better understanding of your audience by minimizing the amount of guesswork in your analysis.
While direct user feedback can be a valuable tool for analyzing behavioral data, always keep in mind the possible biases that may be present (such as who is or is not inclined to provide their responses, and what viewpoints could be missing from the feedback).
Read more: Avoiding bias in user testing
7. Drop-offs
Knowing when users decide to exit your website or app is just as crucial as knowing what pages and content they interact with. Drop-offs are a strong indicator that a user has lost interest or patience, and no longer intend to engage with your site (at least, in this session).
Drop-offs, like event completions, are another data point that you can gather with a standard web analytics tool like Google Analytics; but by approaching drop-off data from a behavioral analytics perspective, you can learn a lot more from them.
Traditionally, drop-off data is reflected in the number of times a specific webpage has served as the “exit page” for a session – i.e., the last page in a user’s session before they left the website for good. This is valuable data, but like many other standard web analytics datapoints, it lacks the individual, human context that makes it meaningful.
When doing behavioral analytics, you start with the big picture – which pages are my most frequent drop-off points? Then, you look at the behaviors of the users who actually dropped off from that page, to answer – why is that page my most frequent drop-off point? For example, what other pages did they commonly view before this one? What kind of interactions did they perform? Were there a lot of indicators of user frustration on this page?
By analyzing behavioral clues, you can restore context to your drop-off data, and identify and fix issues that are causing your site visitors to leave instead of converting.
How to start using behavioral analytics
If you’re thinking about starting to use behavioral analytics in your business, don’t be intimidated! Start the process by examining your current goals, and plan accordingly. Here’s how you can do it.
1. Establish your research questions
Defining what you want to analyze and why is essential for creating a productive behavioral analytics practice, as it is with any analytical project.
Once you’ve created a regular behavioral analytics system, you can always choose new goals and research questions later. To start with, think about known problems with your current digital experience. For example:
- Are visitors not clicking past your homepage?
- Do they have trouble locating and utilizing specific features?
- According to your support staff, are there many difficulties using your software?
These can be the initial research questions you focus on as you begin.
2. Define goals and metrics
Define your behavioral analytics goals and key metrics early, so you have a clear direction once you begin.
This helps keep the whole team on the same page, and also allows you to evaluate whether your behavioral analytics strategies are working or not. If you’re putting in a lot of effort but not seeing any impact on the KPIs you’re targeting, you can conclude that you need to try something different. Make sure to allow enough time for changes to be expressed in the data, though.
You should consider the following when defining your behavioral analytics goals:
- What kinds of outcomes are the best indicators of success
- E.g., Number of sales leads? Traction level of a specific feature?
- Which KPIs are the most useful/accurate way to measure them
- If it’s number of leads, should you measure it by total signups, demos scheduled, trials initiated? If it’s feature traction, should you measure total time spent, total usage instances, or percentage of customers who used it? Pick the KPI that fits best for your case.
- How you will evaluate those KPIs and metrics?
- What does success look like? What number means that your efforts are paying off the way they need to? Is it any net positive change? Or is there a minimum threshold that needs to be met? These answers may need to be derived by working backwards from the overall business goals.
3. Start collecting behavioral data
Assuming you’re not going to build your own solution from scratch, there are a number of platforms on the market today that provide behavioral data and the tools to analyze it.
As a rule, in order to collect this kind of data you’ll need to put a tracking code on your website, regardless of which provider you choose. Make sure you have the approval to do so, and that you involve your engineering resources as needed.
These codes are usually pretty lightweight and have little to no impact on your page load speeds – Trymata’s own Product Analytics suite, for example, uses a tracking code which loads asynchronously to product virtually zero effect on your site’s speed.
Ready to start collecting user behavioral data? Start a free trial now >
4. Analyze your data for insights
Now that you have a tool set up to gather data, you can start analyzing. But where to begin?
One of the primary challenges in the world of data and analytics is the overwhelming amount of information you have the option to dive into. If you’re trying to find just a few crucial moments in a handful of user sessions, with the exact insights that will explain why one of your conversion flows isn’t performing well, and you have thousands of visitors on your site every day – you’re not just looking for a needle in a haystack, you’re looking for a specific piece of hay!
This is why we start with research questions. A good research question limits the scope of data that’s relevant to your analysis, and allows you to filter out a lot of the noise. If you know the question you’re trying to answer, you can use your time much more efficiently and focus more on the data that’s most likely to help explain it.
Based on your research question, decide what slice of the dataset you want to investigate. This may be a certain page or set of pages; a certain user profile type; a certain behavioral pattern; or something else. Then, you can set up segments and filters to view only that data, and ignore everything else.
Once you’ve narrowed your haystack down to just the important stuff (for your research question), there’s no exact formula for how to find your answer. The key things to remember are:
-
- Look for patterns – is there anything that keeps cropping up in different user sessions?
- When something catches your eye, follow that rabbit hole – even if it isn’t what you thought you were looking for!
- Behavioral analysis is all about getting into the weeds to discover the why behind an issue – when you find an example of the issue you’re examining, learn everything you can about it and try to understand the human behind the experience.
- Don’t stop after your first insight or breakthrough – keep looking! There may be more to find.
- Behavioral analytics pairs very well with more direct feedback methods like user testing. Once you have a strong understanding of what people are doing and a sound hypothesis for why, you can try running a targeted user test of that flow to ask people exactly what’s going on in their heads!
Conclusion
Your product decision-making can reach a whole new level with the help of behavioral analytics. By understanding your customers’ behavior and its implications, you can create digital experiences that meet and exceed your site visitors’ needs.
With this information, you can create a strategy and provide a solution to a problem that actually exists and needs solving. Not only are you getting to know your prospects, but you are also connecting with them on a personal level and cultivating an interested and loyal audience.
