Analytics tools

Designing both the structure  and experience for the Analytics platform used by MoEngage's users to understand their customers.  The platform included tools for  analysis of user journeys,  user behaviours, funnels,  user sessions and uninstalls

When MoEngage embarked on the journey to flesh out their analytics suite, they designed an experience that allows users to define the rules and constraints of the final analytical graph. The aim was to create a structure that could be utilized across the various analytical offerings provided by MoEngage, thereby reducing cognitive load and ensuring a coherent experience.

The basic structure we devised consists of four sections

1. The red section highlights the analysis-specific rules. Here, the user would input the steps of a funnel, or the start and end points of the user journey, etc., depending on the type of analysis being conducted.

2. The green section emphasizes the rules that define the segments to be filtered out or compared.

3. The blue section accommodates contextual rules that will assist with the analysis, such as the number of steps in a user journey, or whether the steps in a funnel need to follow a strict sequence, etc.

4. Finally, the grey section is where the resulting graph and table of the analysis will be displayed.

We then benchmarked against our structure against our competitors , mix panel and amplitude

Both amplitude and mix panel, have moved to left panels for rule configurations and right panels for the final analysis. This approach was considered initially since we felt it allowed for more exploration and on the fly cusatomisation (Changes made with rules on the left are effected immediately on the right).

We did not follow this approach for 2 reasons, 

1. Our tech stack did not allow for immediate application and clicking of the apply button was necessary

2. We felt this approach was also not ideal because it felt extremely constrained for more intricate rules, the full width appraoch we used allowed us to reuse the recently designed segmentation engine

In the section used to create the rules of the analysis, we applied a logic similar to the one we used in our segmentation that simplified the rules into plain sentences when they were not being actively edited, ensuring clarity without occupying excessive space.

Additionally, we decided to allow for the re-ordering of steps in certain types of analysis like funnels, etc. Re-ordering as an approach was favored over an approach that inserted steps between existing ones as we felt that approach led to an interface that was both confusing and cumbersome

The structure worked for all analysis types as seen in the section below

The segmentation section having been recently redesigned was reused as is

Finally we have the analysis constraint section ( contextual rules/settings) as seen below

So finally we have the redesigned structure that can be used for all the analytics tools that moengage offers its customers