User Path Analysis: Taking the Guesswork Out of Data Analysis

Reading Time: 5 minutes

Walk a mile in your customers’ shoes. At scale.

The world just went through a turmoil which, according to experts, is far from over.

One thing that’s gained popularity during the crisis, especially with marketers and product managers, is certainty. Staying on top of changing trends in customer behavior, comes a close second.

In a normal world, predictive models would do a reasonable job of the former. In the current context though, the latter becomes a much more pragmatic option.

So how do you keep up with trends that change before you can even test a new tactic and establish a conclusion? You might say, “data”, and you wouldn’t be wrong. But it’s not that simple.

Introducing MoEngage User Paths

When you’re looking at hundreds of millions of data points, where do you even start? How do you avoid getting overwhelmed? And how do you not succumb to tunnel vision and go off on a wild goose chase based on a hunch (or its more popular cousin, “hypothesis”)?

We’re bringing the ability to visualize the paths your users take while navigating your app and website. It consumes the behavioral data and distils it into clean sankey charts, so you can make sense of it.

Take the guesswork out of data analysis

Data analysis is a great way to take the guesswork out of customer engagement. But data analysis itself involves guesswork.

User Paths can give you the clarity you need to make the most of data. It can help you answer open-ended questions such as:

• What are my users doing right after installing? Or right after opening the app?
• What do users do right before making a purchase? Or, more importantly, right before uninstalling the app?
• What paths do users follow between receiving a notification and completing a purchase?

Once you have a high-level understanding of your user behavior, you can develop informed hypotheses. This lets you ask the right questions when you dive into the data and get more out of it. Let’s look at a couple of common examples:

• Before user behavior analysis, you can see if there is a polarizing split in how two personas or user segments use your app – rather than treating it as the average behavior of all your users.
• Before conversion funnel analysis, you can start with User Paths to get the big picture and then narrow your focus with Funnels.

That may sound like common sense once you’ve spent enough time working on your current product. If you’re still trying to get a grasp on your product’s strengths (or weak spots), user personas, or how their behaviour is changing, user path analysis is where I recommend you start. It would also test, and eventually train your intuition for making better decisions.

Fix UX issues and improve your product/marketing strategy

Consider these scenarios:

SCENARIO 1: You’re tasked with fixing the onboarding experience of your app/website

SCENARIO 2: You’re responsible for improving conversion rates of your marketing campaigns

SCENARIO 3: You need to reduce your app uninstalls and improve retention

If you’re a MoEngage customer, these are problems that you’re actually trying to solve. User path analysis not only makes it easier. It drives a BIGGER IMPACT.

SCENARIO 1: Fixing the onboarding experience

Here’s what the analysis would look like:

Let’s boil down SCENARIO 2 to its essence.

SCENARIO 2: Improving conversion rates

In this case, the user path analysis will look somewhat like this:

SCENARIO 3: Reducing app uninstalls

Well, reducing app uninstalls is complicated since it involves a lot of moving parts. You need a good understanding of your product’s “aha” moment and its “uh-oh” moment.

There are tactical things you can do like setting up a deliberate onboarding flow, since the bulk of your users typically leave in the first few days. You can track churned users and retarget them through other channels like email and SMS. But by that time, it’s generally too late and they’ve likely moved to a competitor. Someone who just took that call is extremely hard to win back.

You could use predictive algorithms to see which customers are at risk, but to you, the rationale will always stay fuzzy. It’s like a black box made of tinted glasses.

What if you could see the last five things a user did before uninstalling? Now imagine you can also track system events and campaign events in addition to user action.

So, if they received five notifications over a couple of days and then uninstalled, you don’t need complex data crunching to tell you what went wrong.

The verdict

We’ve been testing out User Paths with a closed group of customers. Here are the results they’re seeing.

For Kutuki the challenge is in building a delicate balance between a fun & engaging Kids App and a comprehensive parenting tool. User Paths has enabled us to validate our strategy for improving the app experience. It has also helped us fine-tune the attribution for purchases by understanding the different paths that lead to a purchase.

Gurupad Badithe
Head of Growth, Kutuki

At EWar games, we help gamers compete at the same level as the big sports. For us, the user experience is everything. We needed to answer questions like what users were doing after certain events, like adding money to their account. User Paths gave us that clarity. An interesting thing we found was that a small subset of users were doing nothing after their first deposit. Once we increased the number of tournaments, we got them to become more active.

Chetan Raithatha
Business Manager, EWar Games

We were waiting for User Path Analysis with high expectations, which have been met.
It’s given us a deeper look into things we couldn’t see before regarding the user journey on the app. We had to see Funnels separately before and we couldn’t get the bigger picture. User Paths combines funnels to give us a complete picture.

Vinicius Bazan
Head of Mobile Applications, Empiricus

If you’re a MoEngage customer, tell us what insights you were able to discover through user path analysis.

If you’re not, and you’ve grappled with questions like ‘What are users doing before they uninstall?’, give User Paths a try.